Wikipedia talk:Verifiability/Archive 59

Archive 55

←

Archive 57

Archive 58

Archive 59

Archive 60

Archive 61

→

Archive 65

all these wiki guidelines/rules pages need to be merged

http://en.wikipedia.org/wiki/Wikipedia:Identifying_reliable_sources http://en.wikipedia.org/wiki/Wikipedia:Verifiability#What_counts_as_a_reliable_source etc.

there's ridiculous overlap and redundancy and it's very very hard to find anything -- there should just be one clean guide and there isnt -- all these pages are extremely wordy and excessive, being a bottleneck to anything good Waveclaira (talk) —Preceding undated comment added 05:38, 26 September 2012 (UTC)

There are only three actual rule pages that matter at Wikipedia that matter: WP:DBAD governs all behavior, Wikipedia:Core content policies covers all content, and WP:IAR covers everything else. All other pages merely serve to elaborate on those principle, for people too dense to get them in their simple forms. --Jayron32 05:46, 26 September 2012 (UTC)

I basically agree with Jayron32. There was an attempt a few years ago to simplify the rules and people came up with WP:Trifecta, tastefully illustrated here by Wikipetan (I have just added the image).

(Wikipedia:Core content policies are WP:Neutral point of view, WP:Verifiability and WP:No original research. Arguably neutrality comes first and verifiability is a way to ensure neutrality and "no original research" is a consequence of verifiability, hence the appearance of a discrepancy between Jayron32 and Wikipetan.)

Yaris678 (talk) 12:27, 1 October 2012 (UTC)

Self-published or questionable sources as sources on themselves

I've just reverted a perfectly good faith edit by Quondam which replaced "themselves" with "the authors", because I don't think that's quite what this section is saying. Could it be slightly clearer.

Self-published and questionable sources, and pages on social networking sites such as Twitter, Tumblr, and Facebook created by or on behalf of individuals, bands, organizations etc, may be used as a source of information about the subject of the source, so long as:'

1. the material is neither unduly self-serving nor an exceptional claim;
2. it does not involve claims about third parties;
3. it does not involve claims about events not directly related to the source;
4. there is no reasonable doubt as to its authenticity;
5. the article is not based primarily on such sources.

It is not required in such cases that the author be a published expert in the field''.

Elen of the Roads (talk) 16:02, 21 September 2012 (UTC)

This suggestion is definitely clearer, but should make it clear that its use as a source must specifically be about the individuals etc. on whose behalf it is created, rather than for the (apparently arbitrary) subject of the source. — Quondum 16:42, 21 September 2012 (UTC)

A self-published source can be reliable for three things: 1) A statement about the opinion of the author, 2) a statement about the individual/group that causes the source to be published (when this is not be the actual author), and 3) a descriptive statement about the contents of the publication itself. I think your suggested change only focuses on the second of the three. Blueboar (talk) 22:20, 21 September 2012 (UTC)

Worth changing it to reflect all three meanings then I think. Elen of the Roads (talk) 23:01, 21 September 2012 (UTC)

• I think self-published sources are good sources for exceptional claims, where phrased correctly. For the sake of argument, let's imagine person X starts a website, and he writes on it: "I am the world's foremost expert on badgers". Editor Y then decides to start a Wikipedia article about person X.

  Editor Y could not write: "Person X is the world's foremost expert on badgers" based on the self-published source. But he could certainly write "Person X claims to be the world's foremost expert on badgers". Couldn't he? We're looking for policy wording that allows self-published sources for anything provided the claims are phrased as reported speech and not as simple statements of fact.—S Marshall T/C 23:50, 21 September 2012 (UTC)

Yes, it could be "Foo claims to be a world expert on badgers", "The Scrotums have tweeted that they are playing Neasden Lido in March", "according to their website, the Acme Drum corporation was founded by a survivor of the Krakatoa explosion" Elen of the Roads (talk) 00:06, 22 September 2012 (UTC) Adding - where you need to be careful is not saying "Rainbow Drops is a magazine specialising in New Age claptrap" and sourcing it to Rainbow Drops. You need a reliable source for the 'new age claptrap' comment. I've seen it done several times with Stormfront - you can source their execrable opinions to their magazine, but not use it to support commentary on the execrability of their opinions, Elen of the Roads (talk) 00:12, 22 September 2012 (UTC)

I'm still trying to visualize tweeting scrotums. --Tryptofish (talk) 00:27, 22 September 2012 (UTC)

People who read Stormfront are definitely tweeting scrotums.—S Marshall T/C 00:32, 22 September 2012 (UTC)

What, this? Blueboar (talk) 00:52, 22 September 2012 (UTC)

I support clarifying the policy per Blueboar's three criteria. I've recently seen evidence that wikilawyers wouldn't accept SPS commentary about a fictional element because the fictional element itself was not the author of the SPS, contra my contention and Blueboar's point that an SPS can indeed be a source for its author's opinions. Jclemens (talk) 01:00, 22 September 2012 (UTC)

I thought the "wikilawyers" reminded you that the use of SPS is not appropriate if it involves "claims about third parties". Of course, SPS cannot be used to source just any of their author's opinion, which is why we have such provisions to their use, otherwise, SPSs would just be RSs if we could use any random blog review of a work.Folken de Fanel (talk) 15:48, 24 September 2012 (UTC)

Perhaps it might help to understand what the underlying principle is, before specifying the who etc. As I see it, a questionable source may be used as evidence of something, provided the "source" constitutes such evidence in itself. With a few conditions designed to filter out abuse and non-notability, which includes in effect that the "source" might reasonably publish whatever; e.g., by definition, someone cannot libel themselves. My point is that we are defining a distinct category of "source" (and allowing it on WP). These categories I see (forgive inaccuracies; you'll understand what I'm getting at) as:

• direct evidence (a nought-ary source, where the source is the subject of the article or contained statement)
• a primary source (where the source's content is used)
• a secondary source (where primary sources are interpreted and synthesized, and this interpretation is used)
• a tertiary source (where secondary sources are summarized)

Allowing a nought-ary source is in effect not requiring citation for generally self-evident facts, and this policy is saying that WP may make a fact self-evident to the reader by providing a link ("see: site X shows pictures of gummy-bears"), again subject to some criteria. To avoid confusion, we should probably not call this a source at all, but rather something like I've done: "providing direct evidence". — Quondum 05:01, 22 September 2012 (UTC)

I think there's a definite need for an etymology of source preference. Too many people have a binary independent-reliable-non trivial-secondary sources are adequate, all other possible evidences are inadequate. In fact, we really have a hierarchy where we want as many of those characteristics as we can muster. For example, triviality is irrelevant to Verifiability, being only really relevant for Notability: an independent, reliable, secondary source that states a bare fact is almost ideal for verifying that fact. Still, I don't think you can get away from calling non-RS'es anything other than sources without causing more confusion and consternation than necessary. Jclemens (talk) 05:09, 22 September 2012 (UTC)

Whatever we call it ("evidence source"?), I feel that we should indicate in policy that it is not the same thing as a primary source. This should help people think a little more clearly about it. — Quondum 10:19, 22 September 2012 (UTC)

Caution... the idea of "providing direct evidence" can easily get us into problems with WP:No original research. That said... Ultimately, all of our sourcing questions come down to a simple concept... is the source appropriate, given the specific context in which we wish to use it. A self-published source will usually be appropriate for a) a direct quote form that source, and b) an attributed statement as to the opinion of it's author (or those who caused it to be published). Self-published sources are usually not appropriate for statements of unattributed fact. (Note: this assumes that it is appropriate to quote or mention the author's opinion in the first place... It may not be... see: WP:Undue weight).

Sounds good. Let's try to work that in:

Self-published and questionable sources, including pages on social networking sites such as Twitter, Tumblr, and Facebook created by or on behalf of individuals, bands, organizations etc, may be used for a direct quote from that source or an attributed statement of the opinion of its author (or those who caused it to be published), so long as:

1. the material is neither unduly self-serving nor an exceptional claim;
2. it does not involve claims about third parties;
3. it does not involve claims about events not directly related to the source;
4. there is no reasonable doubt as to its authenticity;
5. the article is not based primarily on such sources.

It is not required in such cases that the author be a published expert in the field.

There was an earlier suggestion to include "a descriptive statement about the contents of the publication itself", but see no reason to include this. — Quondum 12:46, 22 September 2012 (UTC)

Why not? It is part of "themselves". Blueboar (talk) 15:26, 24 September 2012 (UTC)

"A descriptive statement..." – If this is referring to a statement in WP, e.g. "That site is trashy", then linking to the site so that someone can check their own opinion against the statement ... not allowable, but you didn't mean that. If it is citing a statement in a source (article, site, book) about the publication, then it falls squarely into the description "a direct quote from that source or an attributed statement of the opinion of its author". QED. — Quondum 18:11, 24 September 2012 (UTC)

Something like "that site is trashy" is an editorial comment, and would not be permitted. What I am talking about is something like: "This self-published book discusses five famous battles and compares the strategies of the generals who fought them.<cite the book>" or "Prof. X has acknowledged this error on his personal website<cite profs personal website>". We are not quoting the self published source... nor are we stating an opinion about the self published source... we are stating a fact about the contents of self published source. Blueboar (talk) 18:43, 24 September 2012 (UTC)

I feel that that should be included in citing a source for "...an attributed statement of the opinion of its author (or those who caused it to be published)", though the wording might need broadening; it should not have to be restricted to an opinion (see suggested strikeout), any statement should qualify as much as an opinion. However the topic such an attributed statement may address should be limited; the following numbered points seek to impose the necessary limitation. — Quondum 01:58, 25 September 2012 (UTC)

It's not actually the opinion of the author--it's his stated opinion. We have no way of knowing what his true opinion may be. And it's not his opinion in an universal sense, it's his opinion at the time (which means not just the chronological time, but with respect to where & how he said it).It's rather common here to cherry-pick a quotation from an author's work and use it as his opinion on the subject, when it may actually be an early opinion, or an offhand opinion of a scholar that is not what he says in his formal work, or a particularly inept expression of it. The current US presidential election has had some clear examples of this in the past month. I don't think we need to write out all the possibilities, but we need to remember that interpreting this rule take human judgment. DGG ( talk ) 23:33, 4 October 2012 (UTC)

Russian language references

I have come across a significant number of articles on Russian subjects, such as Verkhovazhsky District, which have very large chunks of illegible references. I would remark that most of them have publishers' names and also dates in Russian; some, such as {{Ru-census}} have both Russian and English terms for the publishers (viz: 'Федеральная служба государственной статистики (Federal State Statistics Service)') that I believe are unnecessary. With my Russian limited to ' da zvidana, I attempted to clean up the situation at one offending template, but was summarily reverted with "rst text which must be included per WP:NOENG". I would therefore seek clarification as to what NOENG is actually supposed to apply to, and whether this final result is compliant with NOENG. Thanks, -- Ohconfucius ^{ping / poke} 06:02, 27 September 2012 (UTC)

The principal issue I see there is that the translation be authoritative. If the citation is linked to a specific record in a published bibliographic index such as the Library of Congress Catalog or a union catalog such as Worldcat then it greatly limits the possibility of misunderstanding. Failing that, an entry at the Open Library is a useful interim step.LeadSongDog come howl! 13:23, 27 September 2012 (UTC)

Remember that all of those "illegible references" are quite legible to our editors who read Russian. I am assuming that your comment about "unnecessary" information refers to giving the publisher in Russian as well as in English (ie you think the English is enough)? If so, I disagree... Having the citation information in the original language is beneficial to verifying the material. Since the material is easily verifiable by Russian speaking editors (and much harder to verify by non-Russian speaking editors) they are the ones most likely to be called upon to locate the source and find out whether it actually supports what is said in the article. They can do this more easily if they know the title, publisher, etc. in the original language. Repeating it in English is in some ways just a courtesy for the rest of us. Blueboar (talk) 21:16, 27 September 2012 (UTC)

Yeah, I have zero problem with references in Russian. If there is an equivalent English source, or if there is a reliable English translation (NOT A MACHINE TRANSLATION) then we should defer to that, but in many cases there isn't, and that is OK too. The English Language does not have a monopoly on reliable scholarship, and we would be losing a huge chunk of what could be potentially very good articles if we had any prohibition against foreign-language sources. --Jayron32 21:46, 27 September 2012 (UTC)

I don't think Ohcon was complaining about using russian sources... I think his question had to do with formatting the citations. Blueboar (talk) 01:26, 28 September 2012 (UTC)

Let me try and rephrase my concern: This is English Wikipedia, and having something in the English Language isn't exactly a courtesy, but a basic necessity particularly where we have a script that is comprehensible only to give or take 300 million people. For the sake of the specific example above, we have as given 'Федеральная служба государственной статистики', which by any means of translation (including G-transl) means "Federal State Statistics Service" The publisher, as an organ of the state, is definitely easily verifiable, so what purpose does it serve to not simplify this clutter? For Chinese articles, I often deal with Chinese sources. I try having the bare minimum of Chinese text. Wikipedia tolerates "|publisher=Xinhua News Agency" instead on insisting upon "|publisher=新华社" or "|publisher=新华社 (Xinhua News Agency)". So my query is why we should not make life a little simpler for the English reader by reducing the wall of unintelligible script by replacing 'Федеральная служба государственной статистики' with 'Federal State Statistics Service' throughout? Similarly, 'Администрация Верховажского муниципального района' could be replaced with 'Verhovazhskogo District Municipal Administration' with no detriment to verifiability. -- Ohconfucius ^{ping / poke} 07:39, 28 September 2012 (UTC)

In what way does changing scripts make it simpler for an English monoglot? They couldn't check the references before this change, and they can't check them after it.—S Marshall T/C 07:47, 28 September 2012 (UTC)

I don't think it's a very good argument, except for doing away with foreign language references entirely, and it isn't what I'm suggesting. -- Ohconfucius ^{ping / poke} 07:52, 28 September 2012 (UTC)

As I said above, including the title of the source, author, publisher, etc. in its original language helps those who wish to find and check the source. So I would include the Non-English.

However, Ohconfucious does have a minor point... this is the English version of Wikipedia. So, I would format the citation giving English pride of place (ie the source info would be given in English first, and then repeated in parenthesis in the original non-English), Example: "Federal State Statistics Service (Федеральная служба государственной статистики)" or "Xinhua News Agency (新华社)".

Actually, to make it all more readable, I would present the entire English citation, and then (in parenthesis) the entire non-English citation... example: <ref>2009 Census of Foobar Province, Slobovian Census Bureau, website/address.gov, 2010 (¶৳№€₤₭ 2009, ₢ƒ¡¿₤, ℳ¶৳♠.₤₫ƒ, 2010)</ref> ) Blueboar (talk) 12:50, 28 September 2012 (UTC)

normally its the other way round: the usual procedure here and in the rest of the world also is to give the original first, and then the translation--the original is the one that matters, the translation is just a convenience. Pride of place has nothing to do with it. Accuracy does. That its the enWP is shown by our giving the English translation. DGG ( talk ) 23:27, 4 October 2012 (UTC)

Errors in reliable sources

Occassionally facts mentioned in the most reliable sources may be wrong. Is there any policy, as opposed to reliance on the good judgment of editors, that explains how to handle disputes over which source to use when this occurs? Say for example because of a typo, Paul Ryan was referred to as the Republican presidential candidate in a reliable source? TFD (talk) 21:17, 12 October 2012 (UTC)

I think that the changes that were implemented a few months ago reinforce that meeting wp:ver is a requirement for inclusion, not a force for inclusion. I.E. that editors usually are free to leave out material e.g. for being in error. I say "usually" because there are special cases, e.g. situations which invoke wp:npov's balancing/weight provisions. North8000 (talk) 21:47, 13 October 2012 (UTC)

North is correct... You may use editorial judgement - with the caveat that before you simply ignore the typo, you may need to talk to other editors and reach a consensus that the typo really is a typo. (I doubt you would need to do this in your Paul Ryan example... but it could be something you would need to do in a less obvious situation.) Blueboar (talk) 16:13, 14 October 2012 (UTC)

I agree also. So the key references for this type of situation are actually probably WP:NPOV and WP:CONSENSUS.--Andrew Lancaster (talk) 10:10, 15 October 2012 (UTC)

Occasionally? Most reliable sources have errors (or distortions that introduce errors). First rule of thumb is to always have at least two independent sources (not one source repeated twice in different pubs) for a disputed fact. Why this isn't enshrined in our policies or guidelines is bizarre. Viriditas (talk) 11:50, 15 October 2012 (UTC)

WP:NOR hints at it, when it talks about "Source based research". Blueboar (talk) 14:52, 15 October 2012 (UTC)

Aren't formal retractions of errata a necessary characteristic of reliable source publishers? If so, where should wp:RS capture that? Right now, it's buried in the wp:NEWSORG section. LeadSongDog come howl! 15:33, 15 October 2012 (UTC)

It depends on the nature of the source... media publishers usually do issue retractions (which is why it is mentioned at wp:NEWSORG)... book publishers rarely issue formal retraction ... instead, they tend to simply correct the typo in subsequent editions without comment (assuming there are subsequent editions). Blueboar (talk) 16:14, 15 October 2012 (UTC)

I think some of the problems now being pointed at here are not problems with WP itself, but with the way academic publishing sometimes works. Here on WP we have to be a bit practical and flexible while the situation is changing so much. Attempts to frame policy in terms of how real academic publishing "should" supposedly work are doomed to fail, because everyone disagrees and there are many new models out there. That's why reference to very fundamental policies like WP:NPOV are probably the best advice for complex cases.--Andrew Lancaster (talk) 18:34, 15 October 2012 (UTC)

WP:NOR can be used to keep out OR, but what can be used to prevent removal of reliably sourced material that has supposedly been shown to be erroneous through OR? --Bob K31416 (talk) 23:05, 15 October 2012 (UTC)

I think that there are too many different things and premises bundled into your question to give a good answer to it. North8000 (talk) 00:30, 16 October 2012 (UTC)

There is no requirement that content be included in Wikipedia simply because it is technically verifiable. Just as there is no requirement to include an attributed opinion simply because it exists. A consensus can always be sought that the erroneous fact is insignificant or just plain inappropriate. Someguy1221 (talk) 00:53, 16 October 2012 (UTC)

verifying through email

Is there any way to verify these rugby league players of Chinese descent through email? anyway here is what was sent to me from ians@nrl.com.au https://mail-attachment.googleusercontent.com/attachment/?ui=2&ik=6520685ce9&view=att&th=13a61827d7ff7f66&attid=0.1&disp=inline&safe=1&zw&saduie=AG9B_P9pTKT5SKAlcnTiooxv7aTN&sadet=1350350639059&sads=Gd8ca6yqIoAtVY_HjWIx6qwdL2Q&sadssc=1 Here is what the list has given me (if it doesn't show); Player surname - Given Name - 2010/11/12/13 club - Contract type - Seasons covered in contract - Highest rep honors - Country born in - Country of Election - 1st parents heritage - 2nd parents heritage (if different from first) - Either grandparents heritage (if different from parents) - Other grandparents heritage (if different from parents or other grandparents).

• Ah Mau, Izaak/25-11-82/Cowboys/Cowboys/none/none/NRL/2011/NZ Schoolboys/NZ/NZ/Samoa/Samoa/China
• Ah-Cheung, Jim/6-2-/95/none/Eels/Eels/Eels/NRL/2013/Samoa/Samoa/China
• Auva'a Kirisome/23-2-92/Storm/Storm/Storm/none/NRL/2012/Samoa/New Zealand/Samoa/China/Kiribati
• Chan, Alex/22-12-74/none/none/none/NRL/0/NZ/NZ/Maori/China
• Chan Sau, Vincent/19-11-89/none/none/none/No contract/NZ/Samoa/Tuvalu/China
• Chan-Ting, Henry/24-7-91/Warriors/Warriors/Vulcans/NSWRL/2012/NZ/NZ/Samoa/China
• Fuller, Dennis/16-2-89/none/none/none/No contract/Australia/Taiwan/Australia/China/England
• Gordon, Kevin/26-12-89/Titans/Titans/Titans/Titans/NRL/2014/CHS Opens 2007/Australia/Australia/China/Australia/Scotland/Ireland
• Groom, Aaron/23-6-87/none/none/none/SG Ball/Fiji/Fiji/Fiji/Fiji/China
• Hoani, Bronson/11-5-95/Eels/Eels/none/NRL/2011/NZ/Australia/Tonga/Maori/China/Germany
• Inu, Krisnan/17-3-87/Eels/Warriors/Bulldogs/Bulldogs/NRL/2015/NSW 19s/NZ/NZ/Samoa/Samoa/China
• King, Denzel/24-6-94/none/none/Bulldogs/Bulldogs/NRL/2013/schoolboys/Australia/Australia/Torres Strait Islands/China
• Lee, Brendan/2-2-86/none/none/none/Other NSWRL/2005/Australia/Australia/Australia/China/China
• McGrady, Cory/15-3/93/Roosters/Roosters/Roosters/NRL/QLD Schoolboys/Australia/Australia/Indigenous/Australia/China
• Sing, Matt/13-3-75/none/none/none/NRL/2006/Australia/Australia/Australia/China/Indigenous

(will finish later on) my point being can I go and make these edits for Aaron Groom, Krisnan Inu, Matt Sing and Ryan Tongia as they all have wiki pages as they were reverted by User:Gibson Flying V talk 02:07, 16 October 2012 (UTC)

• No, verification cannot be done via private, unpublished correspondence. All verification needs to be done through reliable, published sources, the sort of stuff you could find in a library or online repository of reliable sources (JSTOR, online newspaper archives, etc.) We cannot accept emails as "verification" because it's basically "Trust me I know it's true", or at best one step removed from that. If you are basic a player's ethnic ancestry from an email, there's no way that is reliable enough. You'd need to have some firm confirmation in reliable published accounts, otherwise the information shouldn't be part of the article until it is well referenced. --Jayron32 05:30, 16 October 2012 (UTC)

Good idea to link this to Wikipedia:Identifying reliable sources more clearly?

Maybe it would be a good idea to add a Wikipedia:Identifying reliable sources hatnote link to the top of Wikipedia:Verifiability#Reliable sources (at present the hatnote link is only at the bottom)? LittleBen (talk) 00:39, 19 October 2012 (UTC)

 Done. There's a bit of redundancy in the section, as WP:IRS is mentioned a few times, but I agree that the hatnote is useful there. NTox · talk 23:26, 19 October 2012 (UTC)

WP:BURDEN

I am not a cognoscente of the detailed discussions of this policy, so please forgive me if there have been past discussions. But there really seems to be a significant problem with the second sentence of WP:BURDEN. The sentence is: "You may remove any material lacking an inline citation to a reliable source." Not sure if that's a writing error or if the natural semantics are intentional. Is it not inconsistent with all of the instructions around it? Just above WP:BURDEN, we have: "When a reliable source is required ... Anything challenged or likely to be challenged". Then, immediately below it: "You may remove any material lacking an inline citation to a reliable source." If I can remove anything without a reliable inline citation, then it seems quite clear that everything needs a source, not just challengeable stuff. The intent as far as I can tell is that it's trying to say that "You may remove any material lacking an inline citation to a reliable source eventually", but for the hundreds of people like me who don't actually actually write this policy and only read it, that is not going to be clear by reading "You may remove any material lacking an inline citation to a reliable source." Or, perhaps, eventuality is not the intent at all. Either way, something needs to be clarified. As the policy stands, I seem quite justified in removing every single thing I see without a source by citing WP:BURDEN. NTox · talk 01:03, 26 August 2012 (UTC)

That's a big topic. The intent in this area is that if it sincerely challenged, and there is a debate/dispute, the policy weighs in on the side of removal if it does not get sourced. North8000 (talk) 02:18, 26 August 2012 (UTC)

You seem to be referring to the WP:CHALLENGE clause, so I'm going to make a change along those lines to WP:BURDEN. As this is a central content policy, I fully expect that someone will BRD the change, so to that person: please comment here if there is a problem. I'd be happy to discuss. NTox · talk 04:34, 26 August 2012 (UTC)

Well, unilaterally deleting core wording of of a core policy makes that inevitable and I just reverted. The dilemma is easily remedied, and IMHO you approached it in the wrong direction. If it were procedural to challenge material when tagging / deleting then these elements of the policy would be reconciled. Just raising a concern or question besides the lack of sourcing (NO need to have or win a debate) would be enough to move on the the full weight of wp:burden. Not only would such fix the dilemma/conflict, but it would solve a LOT of other problems. North8000 (talk) 12:09, 26 August 2012 (UTC)

North: If the change was surprising to you, I apologize. Sorry. I really didn't think it was that big of a deal. Please note that this wasn't me looking at something and saying "meh, that's stupid. i'm changing policy to something I like better." As I wrote above, I really don't think whoever wrote the text was trying to say "remove anything unsourced". Therefore, I wasn't even trying to make a content change, but a writing change. Of course I left it to BRD to get the details straight, but I was not just ignoring everybody and adding my own opinions to policy. NTox · talk 16:34, 26 August 2012 (UTC)

First one must understand that removal is a form of challenge. It may not be the best form of challenge in a lot of situations, but it is still a form of challenge. With that in mind WP:BURDEN can be understood in the following chronological order ...

1) "Editor A" wishes to add material to an article: If s/he honestly believes the material is verifiable, but thinks that the material is unlikely to be challenged - s/he has a choice as to whether to include a citation when initially adding the material. There is no initial burden to include a citation if you don't think one is needed. So, let us assume s/he decides to not include a citation.

2) "Editor B" comes along later, and does not believe the material is verifiable: he/she may challenge that material, or may choose not to. Let us assume that B chooses to challenge, and does so in the form of removal. At this point the material has actually been challenged.

3) There is now a burden on "Editor A" (or Editor C, D, or E) to supply a citation if s/he wishes to return the material.

Does this mean that every single unsourced sentence in Wikipedia could be challenged and removed?... potentially yes... but in reality no. There are sentences that are so obviously verifiable that no one will ever challenge them.

Are there assholes who challenge obviously verifiable information, and pointedly insist on citations?... yup. The question is, how do you react to them? You can spend hours and hours debating the need to provide a citation ... or... you can mutter "what an asshole" to yourself, spend two minutes tracking down a source and simply return the information with that citation. The second option is far less aggravating in the long run (And, if you really think that including the citation "harms" the perfection of the article, you can always apply WP:IAR, and quietly take it out again once the asshole has moved on to pester some other article... chances are no one will notice or care). Blueboar (talk) 13:58, 26 August 2012 (UTC)

That deserves to be an essay. Perhaps Alternatives to removing content, with something about removing content, which makes articles non-sensical, misleading, or grossly uninformative is strongly discouraged. Alanscottwalker (talk) 15:03, 26 August 2012 (UTC)

We have a policy on that point. See WP:PRESERVE. WhatamIdoing (talk) 15:20, 28 August 2012 (UTC)

I think the biggest thing about a challenge is that it needs to be a challenge based on "I do not believe that text is accurate" not "that text is not sourced". That is, a "challenge" for the purposes of BURDEN on the basis of lack of sourcing is a circular argument, and renders the "likely to be challenged" clause entirely moot: everything unsourced is therefore challengeable by virtue of being unsourced. Thus, I agree with the addition to the policy that was recently reverted--it doesn't add anything to the overall picture, just clarifies away the wrong understanding of what a challenge must be to trigger BURDEN. Jclemens (talk) 15:37, 26 August 2012 (UTC)

re I think the biggest thing about a challenge is that it needs to be a challenge based on "I do not believe that text is accurate" not "that text is not sourced". - that is your interpretation - i am not sure that it is the consensus interpretation. -- The Red Pen of Doom 16:06, 26 August 2012 (UTC)

"Is accurate" might be too narrow a formulation; lack of a citation to check can also contribute to legitimate "challenges" that info is unclear, misplaced, miss-worded, misunderstood, or undue. I don't think "is (not) accurate" unambiguously captures all that. Alanscottwalker (talk) 16:29, 26 August 2012 (UTC)

It isn't just my interpretation... it follows from BURDEN's position immediately following CHALLENGE. Jclemens (talk) 22:22, 26 August 2012 (UTC)

We can have more discussion on the finer points of the issue if people wish to, but my primary interest here as the person who started all this, is: what is the policy /trying to say/ with the statement, "You may remove any material lacking an inline citation to a reliable source." Does it mean that I can literally click the random article button right now and delete everything that's unsourced? If so, let me know, but if that's the case I must have been living in a Wikipedia cave for the last year if I didn't know that. Chuckle. It must mean something else. The change I made to the policy page was a preliminary effort to actually change it to /what it is trying to mean/. Who can illuminate what that is? NTox · talk 16:40, 26 August 2012 (UTC)

The policy makes this clear: Yes... you can (it's called "stubifying" an article)... in fact, you may - if you think removal is the best way to deal with the problem. However, you are cautioned that doing will royally piss people off, and we strongly encourage you to think twice and consider other options. We allow it because sometimes it is the only good way to deal with a problem... but we definitely do not encourage it.

I may be wrong, but I just don't think that's exactly true. I think it's really close to true, but not exactly true. You're saying that I can remove any unsourced content - that I am justified in doing so by this policy - that there is consensus for my ability to do so - but I should note that it might annoy people. Again, if I hit the random article key and went around doing this everywhere, I would get into some kind of trouble. It wouldn't be: "dang, there's that ntox guy being really annoying", it would be "he is violating policy." Now, I can hear the counter-argument already: that the statement in question needs to be read into context, and I understand that. Can we then write it into context? You know; invisibly, without changing meanings, to clarify this? I just feel uneasy about the current wording, as it stands. I don't think it is clear to the layman. What I would be looking for is just a simple change; perhaps something as easy as "You may remove any material lacking an inline citation to a reliable source, if it is in the best interest of the encyclopedia." I don't know what exactly - that just popped into my head - but it needs some kind of basic qualifier. NTox · talk 17:32, 26 August 2012 (UTC)

Upon re-reading this, I suppose there may also be some confusion about what we mean by "you can". Sometimes in practice what we mean by this is "you have a right to do it always", and other times it means "it is an allowed option if the circumstances warrant it". I think in my comment above I was responding to the former, when now I realize that you seem to be referring to the latter. Still I contend that that very issue could be made more invisibly clear, with some kind of basic qualifier so people like me aren't confused by it. NTox · talk 17:43, 26 August 2012 (UTC)

Responding to Jclemens, I agree with you 100% but think you might have made a typo on your last sentence because you said that you agree with an addition which went against what you just said. Possibly you were viewing my reversion of the deletion to be the addition in which case it is consistent. North8000 (talk) 17:18, 26 August 2012 (UTC)

Regarding knocking out material without even questioning it, one must realize that this is generally not due to lack of guidance. It is due to wikilawyering that gos like: How can I use wp:ver to help me conduct my POV/pissing war and knock out the other guys material? North8000 (talk) 17:22, 26 August 2012 (UTC)

Re: It wouldn't be: "dang, there's that ntox guy being really annoying", it would be "he is violating policy." ... Probably... the question is which policy would you actually be violating? Not WP:V. In all likelihood you would be accused of WP:Disruptive editing. And ultimately that's the distinction. Stubifying one random article might be considered be constructive ... stubifying another might be considered disruptive. Being Wikilawyerish and Pointy is almost guaranteed to be considered disruptive. That is one of the things you should consider before you go off on a random article sweep. If it is clear to others that your goal is to assess articles and stubify those that need it, people will probably construe your actions as being constructive. If your goal is to make some sort of Wikilawyerish point about "WP:V says I can do this", your actions will be construed as disruptive. The policy has to allow people to stubify articles, because sometimes that is a necessary action... but we caution them to think twice before doing so(to think about whether they should stubify the article)... because there may be a better option. Blueboar (talk) 18:11, 26 August 2012 (UTC)

Right. I consent to that reasoning (and I'm not /actually/ planning on going on this kind of article sweep!) Just hypothetical. ;) I still would like to see the wording tweaked, so to your ordinary reader their understanding will be exactly consistent with your explanation. The spirit of course would not change, but you must tell I am concerned with clarity, and if there is some wording that can make us both happy as the consensus process recommends, I'd feel good today. NTox · talk 18:20, 26 August 2012 (UTC)

Indiscriminate removal of uncited material (material that is probably verifiable, encyclopedic, etc., not material that is probably unverifiable, spammy, etc.) is a violation of the WP:PRESERVE section of the WP:Editing policy. We don't usually enforce it explicitly, because so few editors are tempted to trash articles, but it is a violation. WhatamIdoing (talk) 15:27, 28 August 2012 (UTC)

Here is why I don't think we should tweak the wording... this is a content policy, not a behavior policy. We have to say "you may remove" because that is how we deal with inappropriate content. I suppose we could say "but don't be disruptive about it"... the problem is that Wikilawyers will argue for days and days about whether an edit was disruptive or not. So we would then try to spell out what is and is not considered disruptive... the policy would become bloated as we try to prevent loopholes for the wikilawyers to exploit (because we forgot to mention something). I think it better to keep behavioral issues to behavior policies, and content issues to content polices. Disruptive behavior is covered... just not here in this policy. Blueboar (talk) 20:42, 26 August 2012 (UTC)

Two points:

• Most editors draw a 'line in the sand' between old content and new contents. Wikipedia has lots of older content that is unsourced or poorly sourced. Sometimes it gets fixed, but most of the time it doesn't. We don't have deadlines so sources can always be added later. However, if you add something to an article that is unsourced, most editors will expect you (or someone) to find a source.
• See WP:PRESERVE.

A Quest For Knowledge (talk) 20:51, 26 August 2012 (UTC)

The last time we discussed this issue was at Wikipedia talk:Verifiability/Archive 58#A problem with BURDEN. In my opinion, there was considerable sentiment, if not actual consensus, for making some significant revisions to BURDEN. Basically, the mediation about the lead was going on then, and most of us wanted to wait until that sorted out, and the full-protection of the page had been lifted, before making other large changes. But I think we can seriously discuss such changes now.

I support the change proposed by NTox. I think it makes very good sense. JClemens pointed out, correctly I think, one of the ad absurdum consequences of the current wording, but more simply NTox is correct that the lead itself specifies the consensus about "likely to be challenged" etc., and BURDEN, as currently worded, reaches beyond what the lead says and actually contradicts it. Like Alanscottwalker, I'd like to see Blueboar's behavioral advice in an essay. But the wording proposed here is not behavioral at all, just common sense. --Tryptofish (talk) 20:59, 26 August 2012 (UTC)

OK... let's explore it... I believe (correct me if I am wrong) the proposal is to say: "You may remove any material lacking an inline citation to a reliable source eventually". The first question a Wikilawyer wishing to remove is going to ask is... how soon is "eventially"? Tomorrow? Next week? Next month? When can I remove the unsourced text? Blueboar (talk) 21:27, 26 August 2012 (UTC)

No, as I understood it, the proposal was this: [1]. And I wouldn't mind looking back to the discussion archived here while we're at it. --Tryptofish (talk) 21:33, 26 August 2012 (UTC)

• I've often thought that it's a pity WP:Editorial judgment is a redlink. This is yet another place where the problem isn't the policy, so much as editorial judgment about when and how to apply it.—S Marshall T/C 22:10, 26 August 2012 (UTC)

I'm getting confused. If you are talking about adding "challenged or likely to be challenged" in another place I'm for it, if it's about removing it from a place I'm against it. But I think that using my idea instead would provide a more decisive resolution which would have many other benefits. Sincerely, North8000 (talk) 03:18, 27 August 2012 (UTC)

Hold on... cutting the excess verbage, the links language says "you may remove if the unsourced material is challenged... " but since a removal is a form of challenge, that would mean the act of removal gives permission to remove. I removed, thus I have challenged, ergo I may remove. Blueboar (talk) 03:40, 27 August 2012 (UTC)

Yes, that is what reconciles the two. IMHO not in the most ideal way (most would say that "challenge" means more than that), but it reconciles them. North8000 (talk) 03:54, 27 August 2012 (UTC)

Removal for any good-faith reason other than "it's not sourced!" would indeed be a challenge within the meaning of BURDEN. But if anything removed as unsourced just because it was unsourced, every unsourced statement throughout all of Wikipedia could be removed at once without discussion, and not reinserted without a reliable source. That's not at all compatible with the notion that Wikipedia is a work in progress. Jclemens (talk) 04:12, 27 August 2012 (UTC)

Jclemens, I very very very very very strongly agree that what you described is how it should be. And that is the clear spirit and intent of "challenged or likely to be challenged". But that is not how it is. We'd need a few wording tweaks to make it so. North8000 (talk) 11:11, 27 August 2012 (UTC)

Agree, it should be something more than just unsourced -- improvement, tempered by judgment, and subject to consensus -- gets somewhat closer (but I don't know if a wording change will help). Alanscottwalker (talk) 12:35, 27 August 2012 (UTC)

What about:

• "You may remove material lacking an inline citation to a reliable source. However, the fact that you are allowed to do so does not mean you necessarily should do so. Whether and how quickly..." (proposed addition in italics)

This addition would (I think) better link the statement of permission to the cautions and exceptions that follow it. Blueboar (talk) 13:22, 27 August 2012 (UTC)

• Or more tersely, "You may remove material lacking an inline citation to a reliable source. "May" does not mean "should". Whether and how quickly..."—S Marshall T/C 13:29, 27 August 2012 (UTC)

First we should realize that the main issues areas aren't people who are just looking for guidance. They are people who are using wp:ver in wiki-lawyering way to knock out material that they don't even question or doubt. The main two purposes are to wage a POV war and to wage a general pissing war. And usually we aren't talking about unsourced material, they use this on sourced material in synergy with / in a domino effect with requirements about the grade of sourcing. E.G. look for a flaw in the given source fully meting wp:rs and then snowball that into removal of the opponent's material. So, while "guidance" carries some influence in such situations, what is really needed is something to prevent such mis-uses.

My thought would be to require expressing some question about the material (in addition to sourcing) when tagging or deleting. It could be something as simple as "I'm not so sure about this, please provide a source" to fulfill that. It is emphasized that this is just a quick vetting procedural requirement for tagging/deleting; after that wp:burden 100% rules, any discussion on the question raised is completely irrelevant to wp:burden ruling. (So we do not create a monster of debating the question instead of a strong application of wp:burden.) My wording is just describing the underlying structure of a proposed change, it is not suitable wording for one. I have been thinking about this for years wp:strategic issues with core policies and believe that this change, more than any other that I can think of, would provide an immense benefit to Wikipedia, so please don't blow this idea off too quickly. North8000 (talk) 14:18, 27 August 2012 (UTC)

I understand the concern, but I don't think your solution does anything to resolve the problem you are attempting to resolve... requiring a statement of explanation won't stop the Wikilawyering POV crusader from deleting... he will simply learn to tack on an edit summary of "doubted - source needed" every time he wants to delete. Blueboar (talk) 15:44, 27 August 2012 (UTC)

If clarity of language can facilitate dispute resolution, thats good. but from what I have seen described here ( i have not actually looked as the specific incident(s)), it does not seem that a language change here would solve behavioral misapplication (of any wording we might land on) there. changing broadly used policy to attempt to meet specific annecdotal issues is generally bad idea.-- The Red Pen of Doom 15:58, 27 August 2012 (UTC)

There is a raison d'etre for the {{citation needed}} and {{verification needed}} tags and their |date= parameter. Even long-established text can reasonably be challenged in good faith without engaging in or being accused of wp:TE. We should not infer otherwise simply because it is obvious to us: editors have widely divergent backgrounds and this is a very good thing for the encyclopedia. Flagging cases of missing or weak sourcing may sometimes be a bit frustrating to other editors, but it does contribute to a stronger 'pedia. Whilst some wikilawyering POV pushers may abuse such tags, they generally find such actions backfire: the arguments against their pushed viewpoints become better sourced. It's one reason that Wikipedia doesn't completely break down in the face of battleground behaviour. LeadSongDog come howl! 16:40, 27 August 2012 (UTC)

I don't think we are discussing tagging, are we? We are discussing the more "drastic" decision on removal. Alanscottwalker (talk) 16:48, 27 August 2012 (UTC)

Well, systematic preemptory deletions in favour of one POV will rapidly draw attention, even if the edit comments are added. With a few key exceptions, such as for potentially libelous BLPs, readding the deleted statement while also adding the necessary dated tags should be seen as a good-faith attempt to wp:preserve content and eventually correct the weak sourcing. Re-deleting such is definitely going to be taken as escalation unless it has been explained on the article talkpage. LeadSongDog come howl! 17:03, 27 August 2012 (UTC)

Responding to Blueboar. Usually these statements are sky-is-blue types and having to say that they doubt them would be a deterrent for even POV warriors. Incidentally sky-is-blue statements can be harder to source because fewer RS's write restating the obvious. North8000 (talk) 17:30, 27 August 2012 (UTC)

Responding to LeadSongDog, that's not it happens. In even those cases the deleter quotes wp:burden and the material stays out. North8000 (talk) 17:30, 27 August 2012 (UTC)

I have always understood BURDEN as erring on the side of caution... leaning towards not returning the material until a citation can be provided with it. That we should (temporarily) not preserve when a challenge rises to the level removal. I suppose that if someone were to "return with a tag", I would respect it as a request for time to find a source... but I would definitely hold the requesting editor's feet to the fire. I would want a see a source provided fairly quickly, and if one did not appear I would remove again.

@North... I disagree... "sky is blue" statements are usually very easy to source. And if one isn't, it's probably OR. Blueboar (talk) 18:08, 27 August 2012 (UTC)

(added later) OK, can you find me a RS that supports this: "There is no record of any major airliner crashing in New York city on August 27th, 2012"?  :-) But I digressed. North8000 (talk) 19:45, 27 August 2012 (UTC)

Been paying some attention to this discussion, and this is one of the statements I can say I agree with entirely. I find it a bit unusual when someone adds information with a "citation needed" tag...it suggests to me that they're expecting someone else to go to the trouble of sourcing their information, while my view is that in theory the person best qualified to provide a source (if not necessarily the citation) should be the person who adds the information to begin with. I'll also admit to being jaded enough that I suspect some editors will/would do that believing, in some cases correctly, that once the information is part of the article other editors will be less inclined to remove it. In other words, it's easier to get the information into the article and leave it for other editors to clean up one way or another, than to do the work of documentating a source oneself at the time of inclusion. Doniago (talk) 18:35, 27 August 2012 (UTC)

Folks, I think that what happens after someone tags or deletes material while sincerely questioning it is a non-issue. I think we all pretty much agree that wp:ver works well in those cases which is leaning towards deletion or leaving it tagged. The place where the policy somewhat conflicts with itself is where somebody tags or deletes it WITHOUT expressing any questions or reservation about the material. Especially where this represents wiki-lawyering to pursue POV wars or pissing wars. North8000 (talk) 19:53, 27 August 2012 (UTC)

• I suspect that we will most quickly come to a conclusion if we focus on specific changes in the language, as opposed to discussing things conceptually. (Your mileage may differ.)
• North, you suggested your idea as an alternative to what NTox proposed. I've read through this discussion a couple of times, and I still don't know exactly what your idea is! For the benefit of nitwits like me, please state specifically what language changes you are proposing.
• NTox's proposal, which is to add the words "if it is challenged or likely to be challenged" at the end of the sentence "You may remove any material lacking an inline citation to a reliable source" continues to be something that I support (whatever else we might do). It has the effects of (1) making BURDEN consistent with what the lead says, and (2) making the reasons for removing material more specific, not less specific. The current wording implies that you can cheerfully remove everything that does not have an inline cite at the end of the sentence; the new wording would limit that to when you consider it to be challenge-able. Blueboar pointed out the logical issue of "I removed it, so de facto it is challenged." That's true. But it isn't a problem. If someone (other than a complete troll) goes to the effort of removing material, then they have challenged it. Maybe you and I agree with them, or maybe we don't, but that's a challenge, and the validity of the challenge is determined by other policies and guidelines, not by verifiability. And if they are a complete troll, the current wording makes it easier for them, not harder. --Tryptofish (talk) 22:52, 27 August 2012 (UTC)

I don't really understand how adding that wording would substantially change the text other than to make it a bit confusing, given that, as you note, removing text is considered a legitimate form of challenge. To my mind, if we add the additional text, what we'd have would read as, "You may challenge material lacking an inline citation by removing it, if it is challenged." Isn't that a bit redundant? Of course, if we say that an editor shouldn't remove material for lacking a citation unless it is likely to be challenged, I think we're opening the door to an unproductive debate regarding how an editor is supposed to determine whether material is "likely" to be challenged. I guess I'm just not sure what we're trying to fix or how adding that particular text would fix...well, anything. Doniago (talk) 12:23, 28 August 2012 (UTC)

As I already said, I agree with Blueboar and you that there is a logical flaw. Your rewording emphasizes the flaw to the point of ad absurdum, and sure, it sounds ridiculous that way. But "You may remove any material lacking an inline citation to a reliable source if it is challenged or likely to be challenged" doesn't have the same kind of comical tin ear, unless one sits around over-analyzing it. What I'm trying to fix (can't speak for anyone else) is editors using BURDEN as a blunt instrument to remove or prevent content with which they disagree, beyond the actual intent of the section. It's the difference between "I'm removing every sentence that doesn't have an inline cite because I can" and "I'm removing this because it's uncited and because I challenge the claim that reliable sources really exist." The latter sentence conveys what BURDEN is actually about; the former goes way beyond its intent. --Tryptofish (talk) 18:43, 28 August 2012 (UTC)

Agreed. The proposed addition is an attempt to get to the heart of the issue - especially, I would say, for those people who are just scanning this page or are unfamiliar with this policy. NTox · talk 18:53, 28 August 2012 (UTC)

Well, I'm glad we agree it sounds ridiculous the way I phrased it. :p I'm still concerned that this addition would confuse editors rather than clarifying anything though. Anyone who's aware of WP:MINREF, or even just assumes that it's okay to delete unsourced material (which, "best practice" aside, it essentially is, currently), is likely to think, "But...why are they saying I may remove anything likely to be challenged when I'm challenging it by removing it to begin with?" It strikes me as rather paradoxical. Now, if an editor's removing material with an eye on WP:BURDEN without intending it to be construed as a challenge...erm, I'm not sure how that could actually happen.

I'm a bit troubled by your addition of "I challenge the claim that reliable sources really exist". That's not how BURDEN currently reads to me (i.e. I don't interpret it as saying that that's intended to be the grounds for exercising the policy), and I'm not sure it was ever intended to be taken that way. My interpretation has always been that the point is that the material isn't sourced, not whether or not an editor believes material can be sourced. Of course, if an editor believes material can be sourced there's probably an argument that removal isn't the best course of action. But that's a whole 'nother kettle of fish.

I have a feeling I worded this rather badly, but I hope I got my points across in some vaguely articulate manner. Doniago (talk) 19:34, 28 August 2012 (UTC)

I've hesitated a bit to be too involved here, because, well, I am not privy to the fine-tuned meanings that this policy is using (I don't have the historical context to know what they exactly are), but I suppose some of disagreement on this point (at least on my end) is that we are understanding 'likely to be challenged' differently. I suppose it is because I believe it involves evaluating the merit of challenges through the lens of consensus; i.e., there are good challenges and bad challenges. If someone comes in and puts a CN tag after every single sentence in an article, I won't consider many of them legitimate challenges. A legitimate challenge to me is reasoned, and is (theoretically) agreed to be legitimate by the community. Therefore, I would not consider it necessary to provide an inline citation for something even if I know if will be challenged by someone, if I have reason to believe that that challenge would be labeled ridiculous by the community. To put this simply: something to me is 'likely to be challenged' if the community is likely to challenge it, not just one person. But maybe this is not the intention of the section and that is something I am open to. NTox · talk 20:48, 28 August 2012 (UTC)

WP:LIKELY is defined. You don't have to wonder about the secret meaning; just go read the page. Or just read this simple summary: something is "likely to be challenged" if you (using your Best Judgment based on your Editorial Experience) believe that it has a ≥50% chance of getting fact-tagged/disputed on the talk page/otherwise challenged. Feel free to assume that's ≥50% chance during a reasonable time period, like a year, rather than ≥50% chance between now and the heat death of the universe. You may also feel free to ignore the likelihood of random or vandalistic "challenges", like the joker who fact-tagged the statement about how many fingers most humans have on their hands. WhatamIdoing (talk) 05:00, 29 August 2012 (UTC)

You know, I think I actually saw a diff like that once. Humorous. Anyway: yeah, I've read WP:LIKELY before. My point merely is that a text like BURDEN often has more to it than it is capable of expressing. Usually you need some historical context to get the whole picture. e.g., you just brought up the concepts of timelines, and ignoring vandalistic challenges, re LIKELY. That is not expressed there, nor in this policy. It naturally leads a person to wonder what else there is under the surface. But more on topic, I think that all of us understand one thing about BURDEN in particular: don't just mindlessly revert people who don't cite sources, and that's really all I'm interested in crystallizing here. Sure we have PRESERVE and that jazz but still I think BURDEN per se implies to newbies that they are free to revert anything unsourced. They're going to read it out of context, whether purposefully or not. Why not make a tweak to avoid misleading people like this, as long as it maintains the integrity of the policy? I think in theory the '...if it is challenged or likely to be challenged' addition makes this work, but I recognize that that might not be clear in practice. It's just a matter of encouraging people to think reasonably, which I think we are all on the same page about. NTox · talk 06:36, 29 August 2012 (UTC)

Doniago, the whole point (as I see it) of not only BURDEN but all of WP:V is that content must be verifiable. So if someone is challenging something on the basis of BURDEN, it is only valid to do so because they believe it is not verifiable. In other words, if the challenge is saying that it is not verifiable, the challenge is saying that reliable sources really don't exist. If they are removing content simply because there is no inline cite, even though they don't have doubts about verifiability (ie, they don't doubt that sources exist), they should not be citing BURDEN as the reason. And yet I frequently see editors, often ones who are pushing a POV, who remove stuff they don't like using exactly that rationale. Please see also the comment I make to PBS and Bob, below. --Tryptofish (talk) 21:51, 28 August 2012 (UTC)

I think part of the issue is that editors tend to interpret "verifiable" in (at least) two significantly different ways- 1) Someone could do independent research to verify the information (i.e. no reference has been provided, but they do exist), or 2) a citation or other acceptable form of reference is provided so that a reader can verify the information through an explicitly-stated source. I'll admit I've always felt that (2) was the better way to handle things. I don't think WP is nearly as good a tool if all readers are given is "well, you can look this up...somewhere". You seem to be leaning towards (1), and obviously if we differ in our understanding of what "verifiable" means then we're going to end up with differing views of what BURDEN should say or is supposed to mean.

Even if you read WP:V, it's left ambiguous. "In Wikipedia, verifiability means that people reading and editing the encyclopedia can check that information comes from a reliable source." But is this supposed to mean that a reader can check that the information comes from an RS because one has been provided...or merely that one could go to the library and find an RS on their own? Again, my understanding is that "verifiable" information is verifiable because there's a citation or other form of reference directing the reader to an actual source for the information, but clearly other editors interpret it differently. Doniago (talk) 13:47, 29 August 2012 (UTC)

I think you've pinpointed exactly where we are seeing things differently. My reaction to your interpretation of WP:V is something akin to "Yikes!". You are right that I've been thinking more the other way. But if I parse it carefully, it's actually something in between. I don't think that V is satisfied simply by saying something akin to "I don't have any sourcing for this, but I figure someone else could go and find a source", which is what (1) is kind of like. But I don't think it's (2), either. I guess it's either: (a) a reliable source is cited on the page, (b) there is consensus amongst editors that it would be possible to find such a source, or (c) there is consensus amongst editors that the material is so obviously true that it would be silly to cite one. Any one of (a), (b), or (c) satisfies WP:V, speaking practically. But if anyone challenges the consensus for (b) or (c), then WP:V considers the material to be "challenged". My thinking about the wording we have been discussing is that I want anyone making such a challenge to, explicitly, question the assertion that the material could be sourced or doesn't need to be sourced. That's not the same thing as saying that there isn't an inline cite so I'm removing it. --Tryptofish (talk) 22:44, 29 August 2012 (UTC)

I agree with this assessment. The challenger should have done some of the footwork and failed to find verification. The challenger should be prevented from unthinking or trivial deletion. Binksternet (talk) 22:51, 29 August 2012 (UTC)

NTox, I think this is the source of your confusion: "If I can remove anything without a reliable inline citation, then it seems quite clear that everything needs a source, not just challengeable stuff."

What does "just challengeable stuff" mean to you? Everything could be challenged. Everything is challengeable.

The policy requires inline citations if, and only if, the material:

1. is a direct quotation
2. has already been challenged, or
3. is WP:LIKELY to be challenged.

Anything could be challenged, but BURDEN does not apply until it has been challenged. WhatamIdoing (talk) 15:19, 28 August 2012 (UTC)

Right. What I meant by 'challengeable', was: any material that 'has already been challenged' (with removal, cn tag, etc.), and any material that 'is likely to be challenged' (has a great probability of removal, a cn tag, etc.). Forgive my shoddy use of the word. According to my change to the policy that people are discussing, I would think that BURDEN does/would in fact apply before something is challenged, because it involves material that is likely to be challenged, but has not been challenged yet. NTox · talk 16:51, 28 August 2012 (UTC)

Comment - I think the original wording of BURDEN is fine: "You may remove any material lacking an inline citation to a reliable source." It is a brief statement that is consistent with WP policy. Of course, the sentence has to be interpreted in context, and the surrounding paragraph & section provide the key guidance that a challenge - actual or likely - must be involved. The word "may" is key: if it said "should" instead, then I agree it should be changed; but "may" is an accurate representation of the policy. BURDEN, in practice, is only invoked in situations when material is being challenged, true? --Noleander (talk) 16:58, 28 August 2012 (UTC)

A deletion without an edit comment is not clearly intended as a sourcing challenge, it could be an editing error, vandalism, POV rejection, or simply a reasonable editorial consideration that the deleted matter didn't belong in the article. Many editors consistently fail to use edit comments. If an edit comment does make it clear that it was a challenge for inadequate sourcing then clearly BURDEN is and should be invoked. The presumption that deletion is a sourcing challenge has never seemed entirely reasonable to me. That said, I'm inclined the other way: that attribution should be provide even for "the sky is blue", since it often is black (with bright spots) or grey, sometimes orange, red, or even green. There's no real reason editors shouldn't cite sources before inserting content on the strength of those sources. We just haven't chosen to mandate that discipline.

LeadSongDog come howl! 17:25, 28 August 2012 (UTC)

I added "but" to the subject part of Burden.[2] --Bob K31416 (talk) 17:30, 28 August 2012 (UTC)

Responding to Tryptofish, I didn't have good final wording but here's my current best shot: Add the following: "When tagging or removing material for being uncited or insufficiently sourced, please indicate that you have a concern or question aside from citing and sourcing." Sincerely, North8000 (talk) 20:00, 28 August 2012 (UTC)

OK, that seems to me to come down to saying that one should provide a good reason for removing material, something I'm broadly receptive to in principle. But the problem is defining a "good" reason. Your preliminary wording seems to say that if you're removing material for these reasons, you also have to say that you have another reason, which I find confusing. --Tryptofish (talk) 21:51, 28 August 2012 (UTC)

I am reverting the edit "You may remove any material lacking an inline citation to a reliable source." There is no but about it. If you want to change the next sentence to says "Instead of deletion consider ..." as an option then OK, but the first two sentences ought not be compromised so that someone can wikilawyer that the burden and deletion can be worked around with a qualification which placing a "but" at the end introduces. -- PBS (talk) 20:19, 28 August 2012 (UTC)

(Here's PBS's revert.[3])

PBS, I didn't understand your above reason for reverting. Here's the sentence before you reverted the change, i.e. before you deleted "but" and restored it to two sentences.

"You may remove any material lacking an inline citation to a reliable source but whether and how quickly removal should happen depends on the material and the overall state of the article; consider adding a citation needed tag as an interim step."

Could you explain with a hypothetical example your reason for reverting, i.e. "...the first two sentences ought not be compromised so that someone can wikilawyer that the burden and deletion can be worked around with a qualification which placing a "but" at the end introduces." Thanks. --Bob K31416 (talk) 21:15, 28 August 2012 (UTC)

Bob, I'm OK with leaving it reverted until we work out all the other issues being discussed here. For example, your edit would make it logistically difficult to make the edit that NTox and I are suggesting, and I don't think that your edit would really satisfy the concerns that we expressed. On the other hand, I partially disagree with PBS's reasons, because I think that PBS is actually advocating for the kind of misinterpretation of BURDEN that I described to Doniago above. --Tryptofish (talk) 21:51, 28 August 2012 (UTC)

So far, I see it's a simple edit that's being blocked for no good reason.[4] You and PBS can do that; I certainly can't stop you. But it doesn't look good. --Bob K31416 (talk) 22:31, 28 August 2012 (UTC)

I would not have reverted it myself. I just meant that while it's been reverted, we should consider all the alternatives that are being discussed. --Tryptofish (talk) 22:43, 28 August 2012 (UTC)

In that case, why don't you now list all the alternatives that are being discussed, and include the edit I made that was reverted.[5] Thanks. --Bob K31416 (talk) 23:04, 28 August 2012 (UTC)

I readily agree that this discussion section has turned into a wall-o-text, but I don't think that my paygrade includes being class secretary. --Tryptofish (talk) 23:12, 28 August 2012 (UTC)

Responding to Tryptofish, that goes a bit beyond my intent. My intent was just a quick procedural one in order to tag or delete. It doesn't even have to be a "good" reason, just merely a statement like "I question that statement, please provide a source" with no further need to support/debate it would be enough. This may seem like a trivial easily-blown-through requirement, but, long story short, after two years of noodling on it, I think that it it would have a HUGE good effect. I can think of many many situations which even that trivial requirement would have helped. North8000 (talk) 23:23, 28 August 2012 (UTC)

From the debaters here worrying about this issue primarily as the wording encouraging deleting in bad faith they clearly have not come across "editors" like the G.-M. Cupertino who inserts subtle unsourced nonsense which can be hard to spot until one see the pattern. Asking his socks to prove it, is the first step on the road to producing evidence that it is yet another sock. So let us put aside issues of bad faith and frame the debate in terms of good faith.

I said at the time that splitting the original section into two ("Anything challenged or likely to be challenged", and "Burden of evidence") was a mistake and I still think it is. It is much better if the two sections are read together and that splitting them would lead people to want to expand them. I personally am still be in favour of recombining them.

What here is being described as BURDEN has an old redirect called WP:PROVIT the point being that if someone does not believe something to be "obvious" then the person making unsubstantiated claims should either prove it (with a reliable source) or leave it out (to use pun on a London expression[6]). It is better to have nothing than to have incorrect text on a subject -- I agree with footnote 3.

I also do not agree with the point that some editors are making about old text and new text. It is often useful for those articles that were written before the quality/quantity debate of 2006 and have not been substantially updated since which often contain the most blatant examples of text that falls within PROVIT.

Perhaps we can rewrite the second sentence to include the word "reasonable" as that should address the issues of bad faith being raised here. -- PBS (talk) 23:41, 28 August 2012 (UTC)

I'm not following. My idea is that you shouldn't have to have or win a debate to invoke wp:burden tag/take out unsourced material, but should have to say that you have a question/concern about the material itself. And that such would reduce mis-use of wp:burden, but not affect legitimate use of wp:burden. North8000 (talk) 22:06, 29 August 2012 (UTC)

WP:Burden arbitrary break 1

I didn't make this arbitrary break, but I'm glad that North did. At this point, I suspect we're going from "wall-o-text" to "no consensus". Maybe we should start posting a series of drafts for a revised BURDEN section, sort of like the way the recent process made a series of drafts for the lead section. I suspect that, unless we focus on concrete content, we aren't going to be able to agree on a way of writing it – or even understand what each of us is arguing for! --Tryptofish (talk) 22:50, 29 August 2012 (UTC)

The core topic is that "challenged or likely to be challenged" can be (arguably) seen as being in conflict with wp:burden. May I suggest we focus on that? North8000 (talk) 23:59, 29 August 2012 (UTC)

Then how about reintegrating the two sections. The initial separation was made on 22 October 2010 here is how it looked just before it was separated. Reading it in the format that existed before October 2010 shows that the two where not then considered to be in conflict. I suspect it is reading it as two separate sections that may make you think that they are. -- PBS (talk) 12:50, 30 August 2012 (UTC)

Yes, I'm starting to be very much persuaded by that. I like the idea of looking at that pre-October version, and working from there. Something that really stands out to me is that it did not have the current language about "You may remove any material lacking an inline citation to a reliable source", which is a very big part of what I have seen as a source of problems, and yet it conveyed the idea of what "you may remove" much better than the current page does. For me, that right there would be a strong argument for going back to a single section. --Tryptofish (talk) 19:44, 30 August 2012 (UTC)

I think like this idea as well... "Any material lacking a reliable source directly supporting it may be removed" was a much better word choice than "You may removed any material...". While the old language could be construed as granting editors permission to remove problematic material, it mainly served as a warning (telling editors that problematic material might be removed). Blueboar (talk) 21:40, 30 August 2012 (UTC)

I like it too. We should clarify here exactly what the change would be the wait a couple days for other comments, and it it looks positive put it in. North8000 (talk) 00:18, 31 August 2012 (UTC)

Yes, maybe someone could copy the text from that version of the page to here on this talk page. Then, we could discuss any revisions to that that we would like to make, and then we go with it if there is consensus. --Tryptofish (talk) 00:42, 31 August 2012 (UTC)

Proposed change?

Change the following two sections under "When a reliable source is required":

====Anything challenged or likely to be challenged====

All quotations and any material challenged or likely to be challenged must be attributed to a reliable published source using an inline citation. The citation should fully identify the source, and the location within the source (specifying page, section, or such divisions as may be appropriate) where the material is to be found. See Citing sources for details of how to do this.

====Burden of evidence====

The burden of evidence lies with the editor who adds or restores material. You may remove any material lacking an inline citation to a reliable source, but whether and how quickly removal should happen depends on the material and the overall state of the article; consider adding a citation needed tag as an interim step. Editors might object if you remove material without giving them time to provide references. It has always been good practice to try to find and cite supporting sources yourself. Do not leave unsourced or poorly sourced material in an article if it might damage the reputation of living people; you should also be aware of how the BLP policy applies to groups.

To the circa 10:15 October 22 2010 version which is:

For how to write citations, see Wikipedia:Citing sources

The burden of evidence lies with the editor who adds or restores material. All quotations and any material challenged or likely to be challenged must be attributed to a reliable, published source using an inline citation. The source should be cited clearly and precisely, with page numbers where appropriate, and must clearly support the material as presented in the article. Drawing inferences from multiple sources to advance a novel position is prohibited by the no original research policy. If no reliable third-party sources can be found on a topic, Wikipedia should not have an article on it.

Any material lacking a reliable source directly supporting it may be removed. How quickly this should happen depends on the material and the overall state of the article. Editors might object if you remove material without giving them time to provide references. It has always been good practice to make reasonable efforts to find sources yourself that support such material, and cite them. Do not leave unsourced or poorly sourced material in an article if it might damage the reputation of living persons or organizations, and do not move it to the talk page.

Except the "notes" cited in this section (removed from both of the above) would not change.

Is this right? North8000 (talk) 01:29, 31 August 2012 (UTC)

Now, how is this supposed to be an improvement? --Bob K31416 (talk) 05:39, 31 August 2012 (UTC)

It's discussed in the talk sub-section immediately above. --Tryptofish (talk) 18:24, 31 August 2012 (UTC)

Just to clarify, I am just trying to be specific about what someone else proposed. There have been some discussion on this above. But the main goal is to reduce what many would call a conflict between "challenged or likely to be challenged" and wp:burden. North8000 (talk) 11:10, 31 August 2012 (UTC)

The sentence starting "If no reliable third-party" should be removed as it was move over to NOR. Likewise the sentence that starts "Drawing inferences from multiple sources" is a summary of WP:SYN is not needed here. Both sentences bloat this section, are better left to the NOR as including them here can cause confusion for inexperienced editors over which policy to follow for this specific advise. -- PBS (talk) 16:09, 31 August 2012 (UTC)

Those two sentences are not in the present version of WP:V, so why work with an old version that has problems. BTW, have you considered simply removing the subsection section titles from the present version and start working from there to get what you want? In any case, suggest keeping in mind the overriding section title ""When a reliable source is required" so that what goes in the section fits the title. --Bob K31416 (talk) 17:49, 31 August 2012 (UTC)

I took them out.North8000 (talk) 18:06, 31 August 2012 (UTC)

Please restore and use strikeouts instead of removal, per talk page guideline. Thanks. --Bob K31416 (talk) 18:29, 31 August 2012 (UTC)

If it will make you happy.  :-). North8000 (talk) 18:31, 31 August 2012 (UTC)

Suggested revisions

Seeing the old and new versions side-by-side, let me suggest this as the To:

====Burden of evidence====

The burden of evidence lies with the editor who adds or restores material. All quotations and any material challenged or likely to be challenged must be attributed to a reliable, published source using an inline citation. The source should be cited clearly and precisely (specifying page, section, or such divisions as may be appropriate) and must clearly support the material as presented in the article. See Citing sources for details of how to do this.

Any material lacking a reliable source directly supporting it may be removed. Whether and how quickly this should happen depends on the material and the overall state of the article. Editors might object if you remove material without giving them time to provide references; consider adding a citation needed tag as an interim step. It has always been good practice to make reasonable efforts to find sources yourself that support such material, and cite them. Do not leave unsourced or poorly sourced material in an article if it might damage the reputation of living people, and do not move it to the talk page. You should also be aware of how the BLP policy applies to groups.

I've copied into it what I can see as better in the current version than in the old one, including some of the wording, and also a shortcut. I also took into account the comments in this discussion so far, including about the header. As I've said above, for me, the key thing that is a big improvement here is the way that it talks about when material may be removed: it doesn't frame it in terms of merely the absence of an inline citation. --Tryptofish (talk) 18:46, 31 August 2012 (UTC)

I noticed that you didn't use "but" to connect the first two sentences in the second paragraph, as is done for the somewhat similar first two sentences in the present version of WP:V. If you did use "but" it would look like the following:

"Any material lacking a reliable source directly supporting it may be removed, but how quickly this should happen depends on the material and the overall state of the article."

By connecting the two sentences with "but", this reduces the chance that the first sentence will be interpreted alone or quoted out of context. --Bob K31416 (talk) 19:23, 31 August 2012 (UTC)

Yes, that would be fine with me personally. --Tryptofish (talk) 19:27, 31 August 2012 (UTC)

I've just added "Whether" to that sentence, per discussion lower down on this page. I'm fine with merging the two sentences, with "...may be removed, but whether and how quickly...", but I don't see it as something I would fight over. --Tryptofish (talk) 20:07, 1 September 2012 (UTC)

I think it is a good idea to show the edits with [] for new words and strike out for any removals. I favour the changes so far (but I am aganist Bob's proposed change) and suggest that unless there are any objections that we implement the combining of the two sections. -- PBS (talk) 08:55, 2 September 2012 (UTC)

Thanks, PBS. I feel good about this approach as well. But before we implement anything, I'd like to clarify a few more issues:

1. Will we also add the footnote discussed just below?
2. Where do other editors stand on the disagreement between you and Bob about one versus two sentences? (At this point, we don't really have a consensus, so much as a stalemate.)
3. What do we want to do with North's proposal?

--Tryptofish (talk) 21Look:45, 2 September 2012 (UTC)

Looks good except "You should also be aware of how the BLP policy applies to groups" appears to be a specialized tangent that doesn't belong here? North8000 (talk) 21:50, 2 September 2012 (UTC)

The BLP groups wording is what the page says now. I suspect the BLP true believers will want it to stay. --Tryptofish (talk) 21:54, 2 September 2012 (UTC)

PBS's reformatting

====Burden of evidence====

The burden of evidence lies with the editor who adds or restores material. All quotations and any material challenged or likely to be challenged must be attributed to a reliable, published source using an inline citation. The source should be cited clearly and precisely, with page numbers where appropriate, and must clearly support the material as presented in the article.[(specifying page, section, or such divisions as may be appropriate) and must clearly support the material as presented in the article. See Citing sources for details of how to do this.]

Any material lacking a reliable source directly supporting it may be removed. [Whether and] how quickly this should happen depends on the material and the overall state of the article. Editors might object if you remove material without giving them time to provide references[; consider adding a citation needed tag as an interim step]. It has always been good practice to make reasonable efforts to find sources yourself that support such material, and cite them. Do not leave unsourced or poorly sourced material in an article if it might damage the reputation of persons or organizations [ living people ], and do not move it to the talk page.[You should also be aware of how the BLP policy applies to groups.]

I have taken the liberty of striking out the old text that is not wanted and adding in the new text in green, so that it is clear what the changes are. To the three points Tryptofish raises: I suggest that we do not include any of those points in this edit, as those are sill being discussed. This change was to address a specific issue and it does not directly affect the others, as they can be added to either this version or the current version if there is a consensus for any of those changes. -- PBS (talk) 14:26, 4 September 2012 (UTC)

No, I really want those three points to be resolved first. WP:There is no deadline. There's nothing urgent about changing what the page says. I think we have made good progress about agreeing on the basic text, and we just need now for more editors to comment on the three issues. My own response is that we should (1) add the footnote, but maybe revise it a bit first; (2) go with 2 sentences, not one; and (3) include North's language, as revised by me. --Tryptofish (talk) 23:21, 4 September 2012 (UTC)

Suggested footnote

I want to make another suggestion. Because it can really be considered as a separate question, I made a separate discussion sub-header. Please look back at Wikipedia talk:Verifiability/Archive 58#A problem with BURDEN. That was a pretty recent discussion that got sidetracked by the discussion about the lead section, but there were good ideas there, and I don't want to see them get lost. In particular, the sub-section about Footnote had something that I think might perhaps get consensus. I'll update it here, as a proposal to add a new footnote at the end of the first sentence, about "burden of evidence". It would only be a footnote, no change to the main text:

Footnote: The burden of evidence is the initial obligation of an editor to provide reliable sources that directly support the material being added. If a good-faith effort has been made to provide such sources, it is then the obligation of editors who do not wish to include the material to articulate problems that would justify its exclusion from Wikipedia. All editors are then expected to help achieve consensus, so that any potential problems with the text or sourcing are fixed.

The archived discussion was largely about the issue of cases where BURDEN is over-emphasized, as though all of the burden always remains on only one side, and the other side has no obligation to discuss anything. There was sentiment then against actually weakening the requirement, but I think there may be consensus that this footnote would be a helpful clarification. --Tryptofish (talk) 19:16, 31 August 2012 (UTC)

I like the idea behind this, but think the requirement on those who wish to keep challenged material is a bit higher than simply making a good faith effort to find a source... that effort actually has to succeed... a source actually has to be provided. Blueboar (talk) 19:52, 31 August 2012 (UTC)

By "a good-faith effort has been made to provide such sources" I presume you mean a good faith effort to find such sources. If that's the case, the failure to find sources would suggest that the material is less likely to be verifiable and shouldn't be included, unless there is reason to include it using WP:IAR. --Bob K31416 (talk) 20:04, 31 August 2012 (UTC)

I see this as a bit of a complex "walking the edge". On one hand, we have widespread mis-use by people who use it to knowck out material that they don't even question to pursue POV wars and pissing wars. On the other hand, we don't want to fundamentally weaken wp:burden to where removal of sincerely questioned unsourced material requires winning a debate to remove it. I think that the closer integration of "challenged or likely to be challenged" with wp:burden safely helps this a bit. I think that my other proposal would help it more without weakening wp:burden. But structurally I don't understand this new idea. Much material (e.g. not challenged or likely to be challenged) does not need to be sourced. North8000 (talk) 20:54, 31 August 2012 (UTC)

I think Blueboar and Bob are right about that. How about changing "If a good-faith effort has been made to provide such sources" to "If sources that an editor believes in good faith to be sufficient have been provided"? That takes it out of the realm of "effort", I think.

Similarly, I could also see changing the last sentence to "All editors are then expected to help achieve consensus, and any potential problems with the text or sourcing should be fixed before the material is added." That also moves the burden back in the direction of not including material without verifiability.

(North, I think you have to make your other proposal specific.) --Tryptofish (talk) 21:40, 31 August 2012 (UTC)

Trypto's change would resolve my concern. We might also work in something about being reasonable if someone asks for more time to find a source. Blueboar (talk) 02:02, 1 September 2012 (UTC)

Tryptofish, Regarding the situation that you described in your last message, where an editor has provided an inline citation but another editor disputes that the ref supports the material — In this case, it seems that you are suggesting that the material should stay in the article until the disputing editor has gained consensus to remove it. I don't think that is good because an editor could provide in good faith an inline citation, which turns out to be part of a violation of WP:Synth, and the material couldn't be removed without consensus. That seems backwards. --Bob K31416 (talk) 02:26, 1 September 2012 (UTC)

When an editor supplies a source in good faith, they have complied with WP:BURDEN. There may still be problems with the material or with the source, but once a source has been provided WP:BURDEN no longer applies. Indeed, the burden of evidence shifts ... it is now up to the challenger to explain what the problem actually is. Blueboar (talk) 12:33, 1 September 2012 (UTC)

You mentioned that "it is now up to the challenger to explain what the problem actually is." Does this mean that only an explanation is required, or does the challenger have to get consensus before removal?

If consensus is required for removal when it wasn't required for the addition of material with a questionable source, it seems that a greater burden is being placed on the challenger. This doesn't seem right that an editor can add questionable material with a source that doesn't directly support the material, and there has to be consensus before it can be removed. --Bob K31416 (talk) 14:07, 1 September 2012 (UTC)

Comment by North moved by me to a separate section below. --Tryptofish (talk) 19:37, 1 September 2012 (UTC)

If someone needs more time to find a source, there is no reason why the unsourced material needs to stay in article space. "If a good-faith effort has been made to provide such sources" I am having a discussion about this at the moment elsewhere! Suppose that a good faith attempt has been made to provide a source, but the source does not meet reliability standards. The debate may go to WP:RS/N but while it is debated there is no reason that the text has to stay in the article. The point made about this is that well known information does not need sourcing, but if it is well known then it will be relatively easy to find a reliable source it is more likely to be obscure information that is difficult to source and in which case it is not unreasonable to ask for a reliable source to be provided. -- PBS (talk) 14:09, 1 September 2012 (UTC)

That's another way of looking at it which seems consistent with my 14:07, 1 September message above. In your case the source may not be reliable, and in the case I mentioned the source does not directly support the material. (BTW none of this seems very relevant to your edit warring on the project page, where you claim it is.) --Bob K31416 (talk) 15:24, 1 September 2012 (UTC)

Footnote: The burden of evidence is the initial obligation of an editor to provide reliable sources that directly support the material being added. If sources that an editor believes in good faith to be sufficient have been provided, it is then the obligation of editors who do not wish to include the material to articulate problems that would justify its exclusion from Wikipedia. All editors are then expected to help achieve consensus, and any potential problems with the text or sourcing should be fixed before the material is added.

Here's a revised draft, reflecting what I suggested and what Blueboar said would satisfy his concerns. I think that both Bob and PBS are raising valid points, and I think that the change I made to the last sentence should address that, because it says "before the material is added". In other words, an editor adds material, with a source. Another editor removes it, challenging the source. The first editor makes a good-faith argument that the source satisfies WP:V. The second editor now cannot simply stonewall, but is expected to take part in consensus building. Discussion follows, before the material is added back. (Should we italicize "before" in the last sentence?) Does that work? --Tryptofish (talk) 19:53, 1 September 2012 (UTC)

I do quite a bit of work in areas where editors, acting in good faith usually because they have no idea what we mean by reliable sources and assume anything on the web and self-published books can be used, add patently unreliable sources. This seems to suggest that the reliability of the sources doesn't matter so long as we know that the editor has looked for sources. Is that what is meant? Dougweller (talk) 07:33, 2 September 2012 (UTC)

It may be that more guidance is needed, but as is shown by the complications that have been highlighted in the proposed text, I think if more guidance is needed then that guidance should be in the guidelines not in a footnote to policy. -- PBS (talk) 08:50, 2 September 2012 (UTC)

I think that Dougweller has raised a good point, and I want to ask whether the wording needs to be made clearer in this regard. My intention is to convey that it's OK to remove material in such cases, but then there must be discussion if the editor who had added the material objects to the removal. I would expect that the discussion would then explain the sourcing issues to that editor. (In other words, you can remove the material, and "articulate problems" at the time of removal, even in the edit summary.) Is that insufficiently clear? --Tryptofish (talk) 21:52, 2 September 2012 (UTC)

Ping! In my mind (others of course may disagree), this is the main thing that needs more discussion. What do other editors think about the questions that Dougweller raised? Should the footnote be revised accordingly? How? --Tryptofish (talk) 23:26, 4 September 2012 (UTC)

Re the sentence, "If sources that an editor believes in good faith to be sufficient have been provided, it is then the obligation of editors who do not wish to include the material to articulate problems that would justify its exclusion from Wikipedia." — This seems to be saying that the material should stay in the article until editors have justified excluding it.

Then there's the sentence, "All editors are then expected to help achieve consensus, and any potential problems with the text or sourcing should be fixed before the material is added." — This essentially says that the material should be deleted before editors have justified excluding it, and consensus is needed before adding it.

These sentences seem somewhat contradictory about whether or not to delete the material before the deletion is justified. --Bob K31416 (talk) 17:11, 5 September 2012 (UTC)

Yes, I think you are right about that, as being where the problem arises. How about this:

Footnote: The burden of evidence is the initial obligation of an editor to provide reliable sources that directly support the material being added. If sources that an editor believes in good faith to be sufficient have been provided, it is the obligation of editors who then remove the material to articulate problems that would justify its exclusion from Wikipedia. All editors are then expected to help achieve consensus, and any potential problems with the text or sourcing should be fixed before the material is added back.

My intention there, is to change the second sentence, such that editors who object per BURDEN are clearly able to remove the material. I think the problem with previous drafts of the footnote was that it sounded like the second sentence was forbidding the removal of sourced material. The revised wording attempts to say that one can remove material if one thinks the sourcing fails WP:V, but you have to explain why instead of just removing it without being willing to articulate reasons. I think that's more accurate as to what is meant by "burden". (It's also a clearer sequence of WP:BRD.) What do other editors think? --Tryptofish (talk) 18:23, 5 September 2012 (UTC)

It's unclear. For example, the first sentence looks like the complete obligation, rather than the initial obligation. Note that the complete obligation is "to provide reliable sources that directly support the material being added." --Bob K31416 (talk) 20:30, 5 September 2012 (UTC)

I see: it's like conflating two obligations. OK, then:

Footnote: The burden of evidence is the initial obligation of an editor to provide reliable sources that directly support the material being added. Once sources that an editor believes in good faith to be sufficient have been provided, editors who then remove the material need to articulate problems that would justify its exclusion from Wikipedia. All editors are then expected to help achieve consensus, and any potential problems with the text or sourcing should be fixed before the material is added back.

The burden per se is still met by the "initial" obligation, but then there are expectations that we place on the challengers, not just the first editor. --Tryptofish (talk) 21:04, 5 September 2012 (UTC)

As before, a problem is the first sentence, which hasn't changed in your rewrite. The initial obligation described in the first sentence and the complete obligation are the same, i.e. "to provide reliable sources that directly support the material being added." I think you want them to be different. Right? --Bob K31416 (talk) 22:32, 5 September 2012 (UTC)

Yes, I see that now. Hey, I'll keep trying 'til I get it right!

Footnote: The burden of evidence is the obligation of an editor to provide reliable sources initially, that directly support the material being added. Once sources that an editor believes in good faith to be sufficient have been provided, editors who then remove the material need to articulate problems that would justify its exclusion from Wikipedia. All editors are then expected to help achieve consensus, and any potential problems with the text or sourcing should be fixed before the material is added back.

--Tryptofish (talk) 00:14, 6 September 2012 (UTC)

Suggested footnote — break1

I don't think the use of "initially" works. I think it's unclear what it is supposed to mean. Here's another approach for what I think you want to do. In the following, I've included a modified version of the sentence of the main text that precedes the footnote. I've underlined additions and used strikeouts for deletions.

The burden of evidence lies with the editor who adds or restores material, and is satisfied by providing reliable sources that directly support the material.^[1]

1. The burden of evidence is the obligation of an editor to provide reliable sources initially, that directly support the material being added. Once sources that an editor believes in good faith to be sufficient have been provided, editors who then remove the material need to articulate problems that would justify its exclusion from Wikipedia. All editors are then expected to help achieve consensus, and any potential problems with the text or sourcing should be fixed before the material is added back.

The footnote needs further work, but let's first see if these changes are OK. --Bob K31416 (talk) 01:18, 6 September 2012 (UTC)

Yes, I like that much better! If no one else objects, I'd say we should go with that. --Tryptofish (talk) 19:17, 6 September 2012 (UTC)

Fine. Regarding the first sentence of the footnote, suggest clarifying in that sentence that the justification for removal needs to be given on the talk page immediately before the material is removed or sooner, if that's what you meant. --Bob K31416 (talk) 14:12, 7 September 2012 (UTC)

Thanks, that's helpful. Actually, I don't feel strongly about whether it's before or after, or on the talk page or in the edit summary, but I can see a way to tighten up the wording in that regard.

The burden of evidence lies with the editor who adds or restores material, and is satisfied by providing reliable sources that directly support the material.^[1]

1. Once sources that an editor believes in good faith to be sufficient have been provided, editors who then remove the material need have an obligation to articulate specific problems that would justify its exclusion from Wikipedia. All editors are then expected to help achieve consensus, and any potential problems with the text or sourcing should be fixed before the material is added back.

With that, I think we're ready to pull all the ideas we have together in one place. --Tryptofish (talk) 21:39, 7 September 2012 (UTC)

Not quite. Suggested rewording:

The burden of evidence lies with the editor who adds or restores material, and is satisfied by providing reliable sources that directly support the material.^[1]

1. Once an editor provides sources that an editor believes in good faith to be sufficient have been provided, any editors editor who then remove removes the material have has an obligation to articulate specific problems that would justify its exclusion from Wikipedia. All editors are then expected to help achieve consensus, and any Any potential problems with the text or sourcing should be fixed and consensus should be achieved before the material is added back.

Reasons for some of the above rewording.

• Previous wording gave the impression that the material with sources was probably going to be removed, i.e. "editors who then remove the material".
• "All editors" in what sense "all"? Rearranged to avoid "all editors" language.

--Bob K31416 (talk) 23:32, 7 September 2012 (UTC)

Maybe I'm just becoming jaded with all the tinkering, but I have low enthusiasm for most of those changes, because they seem to downgrade the importance of consensus-building, and I'm not really sold that this isn't just endless tinkering with diminishing returns. --Tryptofish (talk) 19:11, 8 September 2012 (UTC)

North's proposal

I moved this comment here, so that it can be discussed more fully, in its own right. --Tryptofish (talk) 19:37, 1 September 2012 (UTC)

Responding to Tryptofish, my proposal is to add the following sentence, (or a better written one to the same effect) "When tagging or removing material for being uncited or insufficiently sourced, please indicate that you have a concern or question aside from citing and sourcing." This would thin out the people mis-using wp:ver to conduct warfare, while leaving wp:burden at full strength. North8000 (talk) 02:40, 1 September 2012 (UTC)

• Where would you add the sentence? I don't know, I'm not really comfortable with this idea. It seems to me that one should have the option of challenging material exactly because of citing and sourcing. --Tryptofish (talk) 19:41, 1 September 2012 (UTC)

  If you know that the material is good, and the only problem is the source, why would you want to challenge/remove the material?
  I understand why you'd be unhappy to have "The brain is part of the central nervous system" followed by a citation to Joe's Fishing Blog, but why would you challenge "The brain is part of the central nervous system" in that instance? WhatamIdoing (talk) 19:58, 1 September 2012 (UTC)

  (edit conflict) Oh! So the issue is of material that is thought to be good, but it's just a matter of there being no source or a wrong source. I didn't understand that before. OK! Then let me suggest this: "When tagging or removing material for being uncited or insufficiently sourced, please indicate that you have a concern that the material itself is unverifiable. If you believe the material to be verifiable, then it is better to provide sourcing yourself, instead of removing the material." --Tryptofish (talk) 20:00, 1 September 2012 (UTC)

• North, I think that this is just a bureaucratic step. We need something that can't be accomplished by setting AWB or Twinkle to automatically add a standard edit summary any time a fact-tag gets added. It would be more pointful to say, "You may not challenge any material for which you personally believe a high-quality reliable source could be supplied by anyone who took the time and effort to search for one." WhatamIdoing (talk) 19:58, 1 September 2012 (UTC)

This is to address a pervasive problem in Wikipedia, and the most common form of Wikilaywering, which is misusing policies to go against the objectives of Wikipedia. This is people who do NOT question the material, and do NOT express any reservations or questions about the material. Instead they use this section to knock out material (to further their objectives in a POV war or pissing war) on procedural grounds. Moreover, they seldom apply this on unsourced material so we should not be thinking about unsourced material when we decide this. They apply this on SOURCED material in tandem/synergy/domino effect with the requirement that the sourcing be a wp:rs and so they find technicality problems with the sourcing, and the use this clause to knock out that material to pursue their POV/pissing war. I have the ability/curse to see the underlying structure of the exchanges that happen on Wikipedia and policy-related aspects, and have been testing my thoughts on this during most of my 23,000 edits of experience, so I beg you to not blow me off too quickly on this. I believe that this little change (or a better-worded version of it than I came up with if someone could do that) would have an immense positive effect on Wikipedia, ranging from reducing battles, starting to fix the contested-topic articles where Wikipedia is a dismal failure, reducing chasing away good editors in frustration, to civility in general. So I ask you to give this a thorough hard look and consideration. North8000 (talk) 20:59, 1 September 2012 (UTC)

I understand that problem you're trying to solve. Requiring an edit summary of "I have a concern or question aside from citing and sourcing" is not going to stop POV pushers from removing good material. After all, they do have "a concern or question aside from citing and sourcing", namely, they're very concerned that some other POV is being presented in the article.

BURDEN is not designed to deal with sourced material. It is designed to deal only with unsourced material. Removals of poorly sourced material happen under the authority of other policies. WhatamIdoing (talk) 23:22, 1 September 2012 (UTC)

Thanks for the thoughts. On your first point, I didn't word it well, I implicitly meant a question about the veracity or accuracy of the material.

On your second point, that may very well be the intent, but in practice it is pervasively coupled with the wp:rs requirements, and then domino-ing finding some "weakness" in the source into deletion of the material. I'm guessing that you've often seen material deleted with a "not a RS" type justification, which is an example of this. Sincerely, North8000 (talk) 23:31, 1 September 2012 (UTC)

• I suggested a clarification above, and I think it got lost in this subsequent talk. To repeat, how about:

"When tagging or removing material for being uncited or insufficiently sourced, please indicate that you have a concern that the material itself is unverifiable. If you believe the material to be verifiable, then it is better to provide sourcing yourself, instead of removing the material."

Now that I understand what North was getting at, I think that this wording expresses it more clearly, and I'm quite receptive to adding it. I realize, per WhatamIdoing, that it's hard to stop determined POV warriors, but I accept that this would be a step in the right direction. --Tryptofish (talk) 21:41, 2 September 2012 (UTC)

• Nobody has said anything about where this proposed language might go, and we really need to pin that down. I'll suggest near the middle of the second paragraph, just before "It has always been good practice...". --Tryptofish (talk) 21:59, 2 September 2012 (UTC)

  Sounds good. North8000 (talk) 22:28, 4 September 2012 (UTC)

Bob K31416 question

From the history of the policy.

• 15:04, 1 September 2012‎ Bob K31416 (+4)‎ . . (Rv PBS. I don't see the connection. Please discuss at Talk. And I'm still waiting for your response to my message 4 days ago at Talk. Your reverting without good reason and refusal to discuss at Talk is edit warring.)

I am reverting to the stable version. It is you who is proposing a change, so to call a revert to a stable version when simultaneously stating there are open questions is a misuse of the term "edit warring". As far as I know I have not refused to discuss anything. So:

• What question has not been answered?
• "And I'm still waiting for your response to my message 4 days ago at Talk" what message? A few diffs would help.

-- PBS (talk) 15:35, 1 September 2012 (UTC)

Re " 'And I'm still waiting for your response to my message 4 days ago at Talk' what message? A few diffs would help." — diff 28 Aug --Bob K31416 (talk) 15:47, 1 September 2012 (UTC)

I though I had answered it with a long reply and more specifically:

What here is being described as BURDEN has an old redirect called WP:PROVIT the point being that if someone does not believe something to be "obvious" then the person making unsubstantiated claims should either prove it (with a reliable source) or leave it out (to use pun on a London expression [7]). It is better to have nothing than to have incorrect text on a subject -- I agree with footnote 3

But to give an example. This edit with the comment "Commissioners)". Let us suppose the revert of that edit had been reverted, and then reverted with a comment of WP:PROVIT. Using your wording "but" wording it could have been reverted yet again with the comment See PROVIT "but ... the overall state of the article" allows me time to find a citation. The debate is then framed around the the state of the article and the time needed to find a citation rather than the validity of the information in the first place. Without the qualification of "but" the policy debate ends with PROVIT. -- PBS (talk) 16:20, 1 September 2012 (UTC)

However, the editor could have reverted it just as well with the comment

See PROVIT “Whether and how quickly removal should happen depends on the material and the overall state of the article;” allows me time to find a citation.

Seems like the issue you are raising is with the above phrase, which remains after your revert,[8] not with the word "but". --Bob K31416 (talk) 17:16, 1 September 2012 (UTC)

P.S. Please note that the first sentence in the version you reverted to,

"You may remove any material lacking an inline citation to a reliable source."

is qualified by the second sentence,

"Whether and how quickly removal should happen depends on the material and the overall state of the article;...".

even without the "but". The purpose of the "but" connecting the first and second sentences is to reduce the chance that the first sentence is interpreted alone or taken out of context. Here is what it looks like with the "but".

"You may remove any material lacking an inline citation to a reliable source, but whether and how quickly removal should happen depends on the material and the overall state of the article;..."

--Bob K31416 (talk) 17:43, 1 September 2012 (UTC)

Your argument is precisely why the but should not be there. The difference is that as two sentences the first gives a clear statement of what PROVIT means. The second sentence is advise on whether it ought to be done. Combining the sentences qualifies the first clause it does not advise. -- PBS (talk) 20:12, 1 September 2012 (UTC)

The second sentence is stronger than just advice since it uses the words "should" and "depends". It's a qualification of the first sentence. --Bob K31416 (talk) 20:32, 1 September 2012 (UTC)

BTW, I got involved in this because I was trying to help out after seeing a long discussion that originated with Ntox's message at the beginning of the main part of the section that this subsection is a part of. Ntox wrote,

"But there really seems to be a significant problem with the second sentence of WP:BURDEN. The sentence is: 'You may remove any material lacking an inline citation to a reliable source.' Not sure if that's a writing error or if the natural semantics are intentional. Is it not inconsistent with all of the instructions around it?"

I tried to help out by making a simple benign edit that I thought would be acceptable to everyone and that connects the sentence Ntox mentioned with the sentence that follows it. Seems like you're the only one opposed to it. Even so, I tried to understand your objection but you cut off communication on the talk page about it. However I am glad that you resumed discussion today and I understand your point better, although I don't agree with it. Much if not all of the conflict could have been avoided if we had this conversation 4 days ago. Anyhow, I'll wait and see how the comments to the proposal come out. Regards, --Bob K31416 (talk) 21:45, 1 September 2012 (UTC)

• As I've said directly below, this issue isn't a big deal to me, but I want to caution both of you that you are both in violation of the spirit, and maybe or maybe not the letter, of WP:3RR. --Tryptofish (talk) 19:33, 1 September 2012 (UTC)

The reverting has already stopped after PBS made the last revert and this Talk section started. --Bob K31416 (talk) 21:06, 1 September 2012 (UTC)

Proposal re connecting two sentences with “but”

Please see the discussion in the above section and comment on whether two sentences in WP:Burden should be connected with “but”. In other words, whether the following change should be made in WP:Burden

from

"You may remove any material lacking an inline citation to a reliable source. Whether and how quickly removal should happen depends on the material and the overall state of the article; ... “

to

"You may remove any material lacking an inline citation to a reliable source, but whether and how quickly removal should happen depends on the material and the overall state of the article; ... “

The purpose of the change is to reduce the chance that the first sentence is interpreted alone or taken out of context. Thanks, --Bob K31416 (talk) 18:21, 1 September 2012 (UTC)

---

Comments:

• I'd lean in favor of connecting with a comma, but I don't feel strongly about it. --Tryptofish (talk) 19:30, 1 September 2012 (UTC)
• I don't much care about whether this is one sentence or two. I very much care that we continue to say "whether and how quickly" instead of just "how quickly". WhatamIdoing (talk) 20:01, 1 September 2012 (UTC)

  I agree about "whether". --Tryptofish (talk) 20:03, 1 September 2012 (UTC)

I do wish that people would stop creating new sections about thinks that are being discussed in other sections!

Bob, if there is no conjunction, then I do not agree that the second sentence has the ability to be interpreted as you interpret it for means of reversal (your edits at 17:16 and 17:43, 1 September 2012). If the two sentence are joined then it can be interpreted as you suggest. If that is not true are you pressing for them to be combined?

I wrote "The difference is that as two sentences the first gives a clear statement of what PROVIT means. The second sentence is advise on whether it ought to be done." and you replied "stronger than just advice since it uses the words 'should'" what do you think is the difference between ought and should -- does 'should' mean 'must'? -- PBS (talk) 08:41, 2 September 2012 (UTC)

• If it will help get to consensus quicker, let me float this idea. We go with PBS's approach of keeping the two sentences separate, but we insert a revised form of North's language as I have suggested above. That way, the first sentence really states the PROVEIT concept strongly, but then we will have a fuller explanation of the reasons to be cautious. --Tryptofish (talk) 22:03, 2 September 2012 (UTC)

Reasonable doubt of verifiability

Please comment on the idea expressed in the following,

"If there is reasonable doubt about the verifiability of unsourced material, then it should be either tagged with {{citation needed}} or removed."

Thanks. --Bob K31416 (talk) 10:44, 5 September 2012 (UTC)

seems just like bait for wikilawyering -you dont have a reasonable doubt - leading to pointless discussions about editors intentions rather than leading the efforts to improve the sourcing of articles. -- The Red Pen of Doom 10:48, 5 September 2012 (UTC)

Could you suggest a rewording? (It seems that you didn't express any opposition to the idea that editors should tag or delete if they have reasonable doubt about verifiability.) --Bob K31416 (talk) 11:01, 5 September 2012 (UTC)

I'd rather speak in terms of good faith instead of reasonable doubt. But it seems to me that North's proposal, in a sub-section above, gets at something pretty similar to this. As I see it, we want to indicate that tagging or removing under BURDEN should be because the material is thought to be unverifiable, not because the material fails policies or guidelines other than WP:V. In other words, BURDEN applies specifically to WP:V, not to WP:everything-else-out-there. Of course, one can and should remove material for all kinds of other reasons unrelated to verifiability, but that isn't what BURDEN is for. That said, I don't think that we would need this in addition to North's proposal. --Tryptofish (talk) 18:31, 5 September 2012 (UTC)

I've been thinking about my proposal for 2 years, including mentally "testing" it on real world situations and believe that it would do a significant or immense amount amount of good. Sincerely, North8000 (talk) 20:02, 5 September 2012 (UTC)

I think the proposed wording should be modified to account for situations where larger amounts of unsourced text need to be dealt with and CN isn't the best tag to add. Editors may not realize that there are other templates that are more appropriate for larger-scale issues. Doniago (talk) 20:20, 5 September 2012 (UTC)

It isn't visible on this talk page, but on the page itself there is already a footnote about alternatives to CN. My understanding is that the two existing footnotes are to be retained in whatever we come up with here. --Tryptofish (talk) 20:48, 5 September 2012 (UTC)

(edit conflict)There's currently a footnote in the Burden section that is used for that.[9] --Bob K31416 (talk) 20:56, 5 September 2012 (UTC)

This is getting pretty complicated

This is getting pretty complicated and hard to follow without a pretty big effort. I think that there are just three people left on the train and I'm not one of them. Once this gets hashed out a bit more we should clarify to get more input. North8000 (talk) 01:09, 6 September 2012 (UTC)

Agreed. How about this: I think the main area where discussion is still in need of a few more eyes would be #Suggested footnote — break1. If one or two more editors can say OK there, then I'll create, just below here, a new sub-section, with everything incorporated into it. Then, we can all look at what we have, in one place, and take it from there. --Tryptofish (talk) 19:21, 6 September 2012 (UTC)

I've had little time on Wiki and haven't absorbed this enough to understand. But whatever your idea is I'm sure its good. North8000 (talk) 09:10, 7 September 2012 (UTC)

Sound of evil laughter! Just kidding, thanks. --Tryptofish (talk) 21:30, 7 September 2012 (UTC)

Now don't be gettin' toooo crazy  :-) North8000 (talk) 21:43, 7 September 2012 (UTC)

Summing up

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

---

The result was to implement the proposal. There is a broad base of support for the proposal among the editors who commented here. At twelve participants, this discussion has been commented on by a lot fewer editors than other major policy discussions (for example, the 100 or so editors that commented on the July 2012 RfC about the lead section). However, I do not see this as an obstacle to implementing the proposal, as it is a clarification of the existing policy, and not a major policy change. I note that support for the proposal is not unanimous, and in particular there is a significant minority who would like to remove the mention of "burden of evidence" from the first sentence and move it to the second sentence, or possibly the second paragraph. However, the support for the proposal as stated is strong enough to overcome these objections, at least until further discussions are had about possible amendments. On that note, I would like to stress that this close is not the final word on this subject; editors are encouraged to open up new discussions about further changes to the proposal, whether those changes have already been considered in this thread or whether they are completely new ideas. If you have any questions about this close, you are welcome to ask them on my talk page. — Mr. Stradivarius ^{(have a chat)} 08:19, 20 October 2012 (UTC)

---

Combining the discussions in each of the sub-threads above, I think (please check) that the following is everything that is being proposed. The proposal is to replace WP:V#When a reliable source is required with a single section, as follows:

==Burden of evidence (1)==

The burden of evidence lies with the editor who adds or restores material, and is satisfied by providing reliable sources that directly support the material. All quotations and any material challenged or likely to be challenged must be attributed to a reliable, published source using an inline citation. The source should be cited clearly and precisely (specifying page, section, or such divisions as may be appropriate) and must clearly support the material as presented in the article. See Citing sources for details of how to do this.

Any material lacking a reliable source directly supporting it may be removed. Whether and how quickly this should happen depends on the material and the overall state of the article. Editors might object if you remove material without giving them time to provide references; consider adding a citation needed tag as an interim step. When tagging or removing material for being uncited or insufficiently sourced, please indicate that you have a concern that the material itself is unverifiable. If you believe the material to be verifiable, then it is better to provide sourcing yourself, instead of removing the material. It has always been good practice to make reasonable efforts to find sources yourself that support such material, and cite them. Do not leave unsourced or poorly sourced material in an article if it might damage the reputation of living people, and do not move it to the talk page. You should also be aware of how the BLP policy applies to groups.

• And I support the proposal. --Tryptofish (talk) 23:04, 7 September 2012 (UTC)

• Oppose — I don't think it works. Burden of evidence is a secondary topic here, not the primary. How to show verifiability should be mentioned first, then discuss where the burden of evidence lies. And original section title is thus more appropriate. Bob K31416 (talk) 04:11, 8 September 2012 (UTC)

Given the surprising quiet that has descended on this discussion, I figure I ought to reply to that. Strictly speaking, this section of the page has never addressed "how to show verifiability". I think that's well-covered over the rest of the page, and readers are introduced to it in the lead. The current main header is "When a reliable source is required". Below it, with no introductory text, are "Anything challenged or likely to be challenged" and "Burden of evidence". The reason I put "Burden of evidence" as the suggested header here is that it was what we had been discussing up to this point, so I wasn't going to change it now. But I'm receptive to revising the header if that's where consensus points. Up to this time, one of the main points of agreement in the discussion was that we would combine the two sections into one. But I would also consider moving the first sentence of the first paragraph, above, to the beginning of the second paragraph. But, that said, I don't really see any problem with doing it as shown above, because I don't think that readers will be confused about when material might be challenged as not showing verifiability. --Tryptofish (talk) 00:49, 13 September 2012 (UTC)

Re your comment, "Strictly speaking, this section of the page has never addressed "how to show verifiability"." — I didn't understand this comment since both the current and proposed versions address it. For example, here is an excerpt from the first paragraph of your version. "...must be attributed to a reliable, published source using an inline citation. The source should be cited clearly and precisely (specifying page, section, or such divisions as may be appropriate) and must clearly support the material as presented in the article. See Citing sources for details of how to do this." --Bob K31416 (talk) 16:26, 19 September 2012 (UTC)

Oh, now I see what you meant, and I understand. I guess, with respect to that wording, I don't see any potential for confusion in having it in the order shown here. --Tryptofish (talk) 20:37, 19 September 2012 (UTC)

• Support Seems to sum up and clarify the intent of current policy and usage in the community. LK (talk) 03:50, 13 September 2012 (UTC)

I do not agree with the new wording. "The burden of evidence lies with the editor who adds or restores material." the addition "and is satisfied by providing reliable sources that directly support the material" is not true in all cases. For example it is widely agreed that statement of common knowledge are satisfied without the need to provide reliable sources. In practice there is another area where it is generally accepted that a source is not needed and that is where a statement contains a Wikipedia link to another article. One example given in the past is a hypothetical sourced list of mountains in Africa where they are listed by height, and another article says mount xyz is the "fourth highest mountain in Africa" (with a link to the article). The question then is does that need to be sourced in the main article or if the information is easy to find in the list does it need to be also sourced the article as well. That is open to debate but what is clear is that the proposed second phrase would be used as a club against such an argument.

I was recently asked why ancestor trees need citations but navigation boxes do not. The reason navigation boxes do not is because they usually contain a link to a list that the navigation box relies on. So if citation is provided then it should be provided in the list article. If it is then usually the need to have a citation on the navigation box disappears as the information is not open to a reasonable challenge.

Another example is summary in the lead about a fact that is cited lower down the article. The wording of the second phrase would false citations into the lead even when the information is a reasonable summary of information in the body of the article, because the proposed qualification does not allow for that interpenetration because it says "and is satisfied by providing reliable sources that directly support the material". If it is not meant to mean "and is only satisfied by ..." then the phrase is superfluous and confusing. -- PBS (talk) 09:38, 14 September 2012 (UTC)

• I'd suggest "and can be satisfied by ..." as a replacement, since that doesn't imply "and can only be satisfied by ..."—S Marshall T/C 11:45, 14 September 2012 (UTC)

But it can also be satisfied in other ways, so why extend the sentence? -- PBS (talk) 17:58, 14 September 2012 (UTC)

• Because this policy needs to be newbie-friendly. WP:OBVIOUS applies to policies as well as articles. I like the flow of ideas here: material can be challenged; anyone can challenge it; if it gets challenged here's how you can respond.—S Marshall T/C 19:30, 14 September 2012 (UTC)

I feel like I'm herding cats here, because the phrase to which PBS objects was added to accommodate Bob K.

1. Personally, I would be fine with taking it out of the lead sentence, and moving it back to the footnote.
2. Another option would be sort of a combination of what S Marshall and Bob K have suggested: move the first sentence to the beginning of the second paragraph, so that we start with material that could be challenged, and move the burden part to the next paragraph.
3. I would also support S Marshall's "can be" language, or other language adjustments, as well as another footnote, listing the cases where inline cites are generally not expected.
4. My first choice, though, would be to agree with S Marshall that we are in the realm of OBVIOUS, and the wording here is fine as it is. It's a tortuous argument to say that a "burden of evidence" exists in cases where no one is requesting evidence.

Another alternative would be to keep the organization of the page as it is now, in two sections instead of one, especially if editors want to either have things exactly their own way or have nothing at all. --Tryptofish (talk) 20:34, 14 September 2012 (UTC)

• That would be unfortunate, though, because this is (give or take a few tweaks) a small but genuine improvement.—S Marshall T/C 20:54, 14 September 2012 (UTC)

Summing up the summing up, so far: LK, S Marshall, and I have expressed favorable views of the proposed change. Bob K and PBS have each expressed objections. These objections go in opposing directions, and there hasn't been any movement to bridge those differences. A number of alternatives and tweaks have been discussed, but none seems to have more support than the proposal above. A lot of editors were very active in the discussions leading up to the proposal, but have not commented recently. Like S Marshall, I feel that it would be unfortunate to let this proposal just fade away, because it is a genuine improvement. Therefore, I think it would be a good idea to get more input. Besides, a potential change to a core policy really ought to be widely discussed. So, I'm opening an RfC. --Tryptofish (talk) 19:21, 16 September 2012 (UTC)

• Support the proposal. Seems like it captures the intent of our policies, and leads to the gradual growth of Wikipedia into becoming a better encyclopedia, while not justifying boorish editor behavior. Cheers. N2e (talk) 18:52, 30 September 2012 (UTC)

RfC

Should WP:CHALLENGE and WP:BURDEN be revised as shown above? --Tryptofish (talk) 19:21, 16 September 2012 (UTC)

Versions side by side (2)

Original Proposed Anything challenged or likely to be challenged All quotations and any material challenged or likely to be challenged must be attributed to a reliable published source using an inline citation. The citation should fully identify the source, and the location within the source (specifying page, section, or such divisions as may be appropriate) where the material is to be found. See Citing sources for details of how to do this. Burden of evidence The burden of evidence lies with the editor who adds or restores material. You may remove any material lacking an inline citation to a reliable source. Whether and how quickly removal should happen depends on the material and the overall state of the article; consider adding a citation needed tag as an interim step. Editors might object if you remove material without giving them time to provide references. It has always been good practice to try to find and cite supporting sources yourself. Do not leave unsourced or poorly sourced material in an article if it might damage the reputation of living people; you should also be aware of how the BLP policy applies to groups. Burden of evidence The burden of evidence lies with the editor who adds or restores material, and is satisfied by providing reliable sources that directly support the material. All quotations and any material challenged or likely to be challenged must be attributed to a reliable, published source using an inline citation. The source should be cited clearly and precisely (specifying page, section, or such divisions as may be appropriate) and must clearly support the material as presented in the article. See Citing sources for details of how to do this. Any material lacking a reliable source directly supporting it may be removed. Whether and how quickly this should happen depends on the material and the overall state of the article. Editors might object if you remove material without giving them time to provide references; consider adding a citation needed tag as an interim step. When tagging or removing material for being uncited or insufficiently sourced, please indicate that you have a concern that the material itself is unverifiable. If you believe the material to be verifiable, then it is better to provide sourcing yourself, instead of removing the material. It has always been good practice to make reasonable efforts to find sources yourself that support such material, and cite them. Do not leave unsourced or poorly sourced material in an article if it might damage the reputation of living people, and do not move it to the talk page. You should also be aware of how the BLP policy applies to groups. The version on left uses footnotes 2 and 3, in that order. The version on the right uses footnotes 1, 2, and 3, in that order.

• You're talking about this proposal, correct? I will be happy to take a look at it shortly. I wonder though (and I hate to generate work for you), if you can summarize all of the main things that are different in it from the current version. NTox · talk 19:50, 16 September 2012 (UTC)
  • No problem. Yes, you are correct: the current version is this, and the RfC is asking opinions about changing it to this. The proposal grows out of the entire long talk thread that you see above, and is intended to clarify things that editors in that discussion have seen as misinterpretations of WP:BURDEN, particularly misinterpretations that are sometimes used by POV pushers to Wikilawyer. The substantive changes are:

1. Combining two sub-sections into one.
2. Adding the first footnote, along with a succinct definition, in the first sentence of the main text, of how BURDEN is satisfied.
3. Some re-wording of what is now the first sentence of the second paragraph.
4. Adding two new sentences to the second paragraph: "When tagging or removing... If you believe...".

--Tryptofish (talk) 19:42, 17 September 2012 (UTC)

• Added later: Please see #Mentioning burden of evidence first, below. --Tryptofish (talk) 14:15, 29 September 2012 (UTC)

• Support (Is this the right place?) North8000 (talk) 20:11, 17 September 2012 (UTC)
• I don't think the proposal changes the policy provision in any way... it simply rephrases what the current language says more concisely... so... I Support Blueboar (talk) 20:38, 17 September 2012 (UTC)
  • See my comments in the previous section. How can additional qualifications to a sentence such as "The burden of evidence lies with the editor who adds or restores material." make it more concise? -- PBS (talk) 11:39, 18 September 2012 (UTC)

Because those qualifications are in the policy already... only spread out over several paragraphs and sections. the proposal puts them all together in one place, using less verbiage. Blueboar (talk) 13:26, 18 September 2012 (UTC)

• EhSupport as minor I don't like any policy that gives people the idea they can remove everything that doesn't have a citation just by saying they believe it's unverifiable, without regard to whether they believe it to be true or not. I know truth is a can of worms and all, but if no one is even questioning the truth of a claim, its verifiability should not be an issue. Ultimately people should not be "challenging" claims they believe to be true, even if "truth" is not the ultimate standard we go by when deciding whether to keep challenged material. Gigs (talk) 19:47, 18 September 2012 (UTC)

I agree with your overall view, but I believe that this proposal represents a tiny step / improvement towards your view. This at least says that the person has to say a bit more than "unsourced", i.e. goes a bit towards implementing the spirit of "challenged or likely to be challenged". Sincerely, North8000 (talk) 21:37, 18 September 2012 (UTC)

I dont see how you can get more to the spirit of material being challenged than someone removing the content. -- The Red Pen of Doom 14:05, 19 September 2012 (UTC)

"Jesus is the Son of God" is a statement I believe to be true, but know is not verifiable. If I saw that statement in an article I would either remove it, or rewrite it (as "Christians believe that Jesus is the Son of God")... even though I believe the original statement to be true. Blueboar (talk) 22:15, 18 September 2012 (UTC)

North8000, I can see what you are saying, and I've proposed something similar in the past "challenges should be based on good faith doubts about a claim". Right now the policy is silent on what should motivate challenges, which is a mixed blessing, I think. To me the proposal would be better if "please indicate that you have a concern", should read "please indicate why you have a concern ". I don't think it's reversing burden to at least ask people think for 5 seconds about why they have a problem with some content, and articulate that.

Blueboar, On the balance, there would be someone along to challenge the factual content of that claim, even if you believed it to be true and left it as it were. I just don't want to see people challenging uncontroversial claims merely because of a lack of citation. Gigs (talk) 13:27, 19 September 2012 (UTC)

And there is the difference between us... I don't worry about people removing uncontroversial claims merely because of a lack of citation. It may be annoying when it happens, but if the information really is uncontroversial, it will usually take all of two minutes to find a citation and return the information. Returning the information with a citation fixes the problem and improves the article. Blueboar (talk) 13:56, 19 September 2012 (UTC)

I'm with you both on this but I'm thinking about pervasive mis-uses. And it goes like this: To use wp:policies conduct a POV war and POV an article my way, I want to knock out material which I KNOW is accurate. The most effective way to do this is to use wp:burden COMBINED with detailed sourcing guidelines, knowing that perfection on the latter is impossible or immensely time-consuming to achieve. I can just knock out material without questioning it based on the imperfection in the sourcing in TANDEM with wp:burden. Asking folks to question the sourcability of material when knocking it would reduce this misuse. North8000 (talk) 14:37, 19 September 2012 (UTC)

Well in any case, I guess I can support this as a small improvement that doesn't really change much. Gigs (talk)

I think that in POV wars it is ESPECIALLY critical to have a source to back all claims.-- The Red Pen of Doom 17:29, 19 September 2012 (UTC)

Agree, at least for "challenged or likely to be challenged"; this proposal implements that. North8000 (talk) 17:33, 19 September 2012 (UTC)

• I like all of changes made here and I think it is an improvement on the current version. The addition to the first sentence ("...and is satisfied by providing reliable sources that directly support the material.") is something I particularly like, because it emphasizes the importance of having sources that actually support stuff. I think there is a temptation with the current version to indulge in a "hey, I added a source after it; there's a ref tag" and be done without enough concern that the source is really supportive. I also like the two added sentences in the second paragraph. I have always found "unsourced" to be a very poor reason for removing content; the better reason is "I doubt this is possible to verify". The proposed version encourages editors to say such a thing instead. The fact that it encourages editors to PRESERVE material they think is verifiable is also welcoming. We tend to forget about the principles of the editing policy. I would say however that there are indeed exceptions here — we do not and should not keep contentious claims about living people around without a good, cited source, even if we think they are verifiable. Further, I'm inclined to say that the combination of these sections into one is good, but I may care less about that than others — the benefit I see is that it emphasizes the burden we have for citing challenged and likely-challenged stuff particularly. And thank you to whoever tweaked the wording of that pesky introductory sentence in the second paragraph (current/bad version: "You may remove any material lacking an inline citation to a reliable source."). The new version, "Any material lacking a reliable source directly supporting it may be removed," is much better. The former implies, as I still believe, that people are justified in unilaterally removing things that don't have inline citations. The latter says (assuming I've understood the intentions), that material that can't be verified is what can be removed. Perhaps it is still not exactly clear enough for my taste because the verifiable/verified/cited issue is fuzzy, but I support what we have at this stage and encourage more tweaks to come. NTox · talk 23:22, 18 September 2012 (UTC)

  For what it's worth, in my opinion you have done an excellent job of explaining why I, too, consider this revision to be a step in the right direction. --Tryptofish (talk) 23:28, 18 September 2012 (UTC)

• Comment We should decide on a process. My thought: Run it for exactly 2 weeks. This page is watched by 1,454 people plus seen by more. That should be the main input. If there is an obvious strong consensus either way, implement it. If not, ask for an admin close. Sincerely, North8000 (talk) 13:24, 19 September 2012 (UTC)

I have no objection to doing it that way. Maybe with the caveat that we can leave it open longer if the discussion looks like it would benefit from doing so (not looking that way to me now). --Tryptofish (talk) 20:37, 19 September 2012 (UTC)

Depending upon how one counts it, we will be coming up on 2 weeks in another day or so (if anything, a bit more than 2 weeks). I suggest that in approximately another day, I'll ask an uninvolved administrator to close the RfC. Are there any objections to doing it that way? --Tryptofish (talk) 14:27, 1 October 2012 (UTC)

[10], [11], [12]. --Tryptofish (talk) 21:56, 3 October 2012 (UTC)

Update: [13], [14]. --Tryptofish (talk) 18:21, 9 October 2012 (UTC)

• Comment - I took the liberty of presenting the two alternative texts, side by side, at the top of this RfC. If I inadvertently captured the wrong text, feel free to correct it. --Noleander (talk) 16:54, 19 September 2012 (UTC)

OK, but users should note that the footnotes for the old version are the second and third footnotes; the footnotes for the proposed version are as shown. --Tryptofish (talk) 20:37, 19 September 2012 (UTC)

• Support - I have some quibbles with the wording of the proposal, but I always thought that Challenge and Burden were two sides to the same coin, so merging them makes sense. The proposed text seems more straightforward, hence more understandable for newbie editors. --Noleander (talk) 16:58, 19 September 2012 (UTC)

Note that merging can be accomplished by simply removing the subsection titles of the current version. --Bob K31416 (talk) 18:33, 19 September 2012 (UTC)

Sure, and that is the major improvement of the proposal: policies should be as simple and streamlined as possible. The wording of the proposal is not ideal, but it is at least as good as the original. In particular, I like the new sentence "If you believe the material to be verifiable, then it is better to provide sourcing yourself, instead of removing the material" ... it is a good intro to the following sentence re PRESERVE. --Noleander (talk) 20:43, 19 September 2012 (UTC)

Re your comment, "policies should be as simple and streamlined as possible" — Note the relative size of the proposed version compared to the present one, keeping in mind the added footnote too. --Bob K31416 (talk) 13:40, 20 September 2012 (UTC)

• Comment — It's kind of awkward for me to give a "support" or "oppose" to this proposal because some parts are good, some aren't, and some need rewrite. I'd be more comfortable if it was broken up into more than one proposal. For example, the first proposal could be removal of the two subsection titles in the present version of the section When a source is required, and changing from two policy shortcuts boxes to the proposed one. --Bob K31416 (talk) 13:23, 20 September 2012 (UTC)

What the heck, this is a relatively small change that doesn't seem to be controversial so I'll make it now and see what happens. --Bob K31416 (talk) 14:00, 20 September 2012 (UTC)

Cool. We could view this as a net step forward and then (if it flies and gets implemented) figure that we'll reopen the discussion for another round of tweaks on it. North8000 (talk) 14:08, 20 September 2012 (UTC)

I agree that it's probably a non-controversial change, but I wouldn't want it to interfere with the views that have already been expressed here. By my count, 10 editors have commented so far. Bob and PBS have objections to the proposal, although I don't think PBS would object to Bob's interim change. However, 8 other editors have expressed support for the proposal as a whole (some advocating further, additional tweaks), including the language changes that are not implemented via Bob's change. I accept that Bob feels awkward about supporting or opposing the proposal in its entirety, and that's fine, but it actually appears that 8 editors are at least reasonably comfortable also supporting the proposal as it was presented. Therefore, I don't want this interim change to be where things stop. --Tryptofish (talk) 21:04, 20 September 2012 (UTC)

• Comment — Here's the side-by-side comparison updated for the current version in WP:V, which no longer has the two subsection titles.

Versions side by side, updated (3)

Current Proposed When a reliable source is required Burden of evidence All quotations and any material challenged or likely to be challenged must be attributed to a reliable published source using an inline citation. The citation should fully identify the source, and the location within the source (specifying page, section, or such divisions as may be appropriate) where the material is to be found. See Citing sources for details of how to do this. The burden of evidence lies with the editor who adds or restores material. You may remove any material lacking an inline citation to a reliable source. Whether and how quickly removal should happen depends on the material and the overall state of the article; consider adding a citation needed tag as an interim step.[2] Editors might object if you remove material without giving them time to provide references. It has always been good practice to try to find and cite supporting sources yourself. Do not leave unsourced or poorly sourced material in an article if it might damage the reputation of living people; you should also be aware of how the BLP policy applies to groups.[3] The burden of evidence lies with the editor who adds or restores material, and is satisfied by providing reliable sources that directly support the material.[1] All quotations and any material challenged or likely to be challenged must be attributed to a reliable, published source using an inline citation. The source should be cited clearly and precisely (specifying page, section, or such divisions as may be appropriate) and must clearly support the material as presented in the article. See Citing sources for details of how to do this. Any material lacking a reliable source directly supporting it may be removed. Whether and how quickly this should happen depends on the material and the overall state of the article. Editors might object if you remove material without giving them time to provide references; consider adding a citation needed tag as an interim step.[2] When tagging or removing material for being uncited or insufficiently sourced, please indicate that you have a concern that the material itself is unverifiable. If you believe the material to be verifiable, then it is better to provide sourcing yourself, instead of removing the material. It has always been good practice to make reasonable efforts to find sources yourself that support such material, and cite them. Do not leave unsourced or poorly sourced material in an article if it might damage the reputation of living people, and do not move it to the talk page. You should also be aware of how the BLP policy applies to groups.[3] The version on left uses footnotes 2 and 3, in that order. The version on the right uses footnotes 1, 2, and 3, in that order. 1. Once sources that an editor believes in good faith to be sufficient have been provided, editors who then remove the material have an obligation to articulate specific problems that would justify its exclusion from Wikipedia. All editors are then expected to help achieve consensus, and any potential problems with the text or sourcing should be fixed before the material is added back. 2. It may be that the article contains so few citations that it is impractical to add specific citation needed tags, in which case consider tagging a section with {{unreferencedsection}}, or the article with {{refimprove}} or {{unreferenced}}. 3. Wales, Jimmy. "Zero information is preferred to misleading or false information", WikiEN-l, May 16, 2006: "I can NOT emphasize this enough. There seems to be a terrible bias among some editors that some sort of random speculative "I heard it somewhere" pseudo information is to be tagged with a "needs a cite" tag. Wrong. It should be removed, aggressively, unless it can be sourced. This is true of all information, but it is particularly true of negative information about living persons."

--Bob K31416 (talk) 11:38, 22 September 2012 (UTC)

Discussion about location of material in a source

• Comment — The proposed version removed significant information that is in the current version regarding location of material in a source.

The proposed version changes the following sentence of the first paragraph of the present version,

"The citation should fully identify the source, and the location within the source (specifying page, section, or such divisions as may be appropriate) where the material is to be found."

to the following sentence of the first paragraph of the proposed version.

"The source should be cited clearly and precisely (specifying page, section, or such divisions as may be appropriate) and must clearly support the material as presented in the article."

The proposed version left out the part about "the location within the source" "where the material is to be found", which is an important point of the guidance. For example, citing only a book can make it very difficult to verify material if the location within the book where the material is to be found is not specified. Furthermore, if the book does not support the material, it would be very difficult to determine that without knowing where in the book the supporting information is supposed to be. Just giving the guidance "as may be appropriate" is vague and insufficient without "the location within the source" "where the material is to be found". --Bob K31416 (talk) 14:45, 28 September 2012 (UTC)

I see what you are referring to, and I suspect that restoring that language would not be controversial (and in fact, I don't recollect any specific decision to take those words out, although my recollection may simply be atrophying by now). However, I'm not really seeing any loss of information, as opposed to providing the same information in a more succinct way. It still says "(specifying page, section, or such divisions as may be appropriate)", which spells out what "the location within the source" is. That still communicates, per your example, that it's often not good enough to simply reference a book in its entirety, without specifying a page, etc. After all, it would be nonsense to specify those things if they referred to another place within the source, unrelated to the relevant information. And it also says "and must clearly support the material", which makes it clear that the citation must be about where the relevant material must be found. Maybe I'm missing something, but it's very difficult for me to envision how someone could either misunderstand or willfully misconstrue the proposed language to mean something other than what the old language also said. --Tryptofish (talk) 15:24, 28 September 2012 (UTC)

Re "It still says '(specifying page, section, or such divisions as may be appropriate)' " — Appropriate for what? In the proposed version it is appropriate for citing the source clearly and precisely, but it doesn't give the purpose for doing this, which makes it more difficult to decide how specific to be. In the present version it suggests that the purpose of being specific is to be able to find the material in the source, so the editor has some guidance about how specific to be. --Bob K31416 (talk) 18:26, 28 September 2012 (UTC)

I don't know, maybe I'm just being dense, and maybe it would really be good for some other editors to take a close look at this, because I really might be missing something. But it seems to me that the reply to your question of "Appropriate for what?" is "What else could it possibly be appropriate for?" To me, it seems self-evident. --Tryptofish (talk) 21:44, 28 September 2012 (UTC)

Maybe it would be clearer for you if there wasn't that question, so I struck it out. Try reading it again. --Bob K31416 (talk) 22:40, 28 September 2012 (UTC)

Thanks. I re-read it, and I'm still not seeing it. I guess we just need to see what more editors think. --Tryptofish (talk) 14:03, 29 September 2012 (UTC)

After giving some thought to this, my opinion is that the second version doesn't destroy anything we are trying to accomplish, but that I do like the original wording of this sentence better. To me, this is a case in which we should state the obvious, even though I agree with Tryptofish generally and would say you'd probably have to be a bit dense not to get it. For verifiability purposes, the source location—article location concept is important (text-source integrity), and any way we can emphasize that is IMO advised. Both say this in their own ways, but I get a better feeling of it from the original. NTox · talk 15:38, 29 September 2012 (UTC)

I guess, then, that I could agree that there would be a little bit of good accomplished by using the existing wording of the sentence instead of the proposed wording, but I also don't see it as a major or urgent issue. Unless the remaining days of the RfC lead to an outpouring of more editors endorsing this point, I think we can reasonably leave it as a fairly non-controversial edit that will probably be made after the RfC is concluded. --Tryptofish (talk) 20:28, 29 September 2012 (UTC)

Possible revision

"When tagging or removing material for being uncited or insufficiently sourced, please indicate that you have a concern that the material itself is unverifiable." This sentence still rubs me the wrong way, but I can support it if nothing better happens. It still sounds to me like it's OK to remove material merely "for being uncited or insufficiently sourced", as long as you state that you think it's unverifiable, whether you actually think that or not. Can we still tweak this wording? Gigs (talk) 17:13, 19 September 2012 (UTC)

That might depend on what changes you would want to make. In my opinion, someone saying that they think something is unverifiable when they actually believe it is verifiable is pretty far out there, and probably difficult to legislate against, no matter how we word it. --Tryptofish (talk) 20:51, 19 September 2012 (UTC)

I don't think you are considering the "real world" (ab)use of these policies, if you find that difficult to imagine. My concern could be addressed by rephrasing the sentence as such: "When challenging material, please indicate that you have a concern that the material itself is unverifiable". I don't think this changes the intended meaning, and it closes the potential wikilawyer-able implication that I think the current proposal has. Gigs (talk) 03:17, 20 September 2012 (UTC)

"real world abuse"?? i am pretty sure that there is more "real world abuse" potential from POV and vandalism pushers inclusion of unsourced false material and wikilawyering over "you dont really think that its not verifiable" than requiring somone to spend two minutes looking up an easily verified "fact" that they want to have included in an article. -- The Red Pen of Doom 14:03, 20 September 2012 (UTC)

You are right, it goes both ways. I absolutely agree we can't put the burden on the remover, that would require them to prove something that's not falsifiable. We just need to be careful with our wording so we don't give POV warriors a new tool for wikilawyering. Gigs (talk) 14:38, 20 September 2012 (UTC)

If I understand correctly, the change would be from "When tagging or removing material for being uncited or insufficiently sourced," to "When challenging material,". For me, that wouldn't do anything about users who claim that something is unverifiable when they actually believe it to be verifiable, but it loses a lot of the specificity of the language that has been proposed, so I guess I would regard it as something that would not be a friendly amendment, but instead as something that could be proposed separately from this proposal. --Tryptofish (talk) 20:54, 20 September 2012 (UTC)

It removes the language that makes it sound like it's OK to remove material merely because it's unsourced, without regard to verifiability. Another possible rewording would be to say "When removing or tagging unverifiable material, please indicate..." Gigs (talk) 22:40, 20 September 2012 (UTC)

Personally, I'm not seeing it that way. I agree with you that we should want to discourage people from removing material just because there isn't an inline cite, but the language proposed, taken as a whole, moves us in just that direction. Taken as a whole, it says that if you remove something for being uncited etc, you need to actually present a reason that it isn't verifiable, not just that there wasn't an inline cite. It seems to me that the changes you suggest actually would be a step back from that. But if other editors want to support those changes, I'm prepared to listen. --Tryptofish (talk) 23:13, 20 September 2012 (UTC)

Agree with Tryptofish's last response. Gigs, I also agree with your overall view. But IMHO I'm not sure that you realize that the proposed change is a small improvement in the direction that you prefer. North8000 (talk) 00:56, 21 September 2012 (UTC)

I can see that argument, but I'm just afraid that someone will read it the wrong way, the way it's written now. Gigs (talk) 13:25, 21 September 2012 (UTC)

Mentioning burden of evidence first

I don't get why you'd want to mention burden of evidence first. To me "All quotations and any material challenged or likely to be challenged must be attributed to a reliable published source using an inline citation." is the definitive statement, the principle we are trying to follow.

"The burden of evidence lies with the editor who adds or restores material." explains whose responsibility it is to put that principle into practice. I can see why someone might think that might be necessary with a contentious issue like this, but I'm pretty sure it would help to state the principle first.

Yaris678 (talk) 16:19, 28 September 2012 (UTC)

Your comment made me realize that some things got messed up when other editors created the side-by-side comparisons of the drafts. I've tried to correct them, but I'd appreciate it if someone else would check my corrections for me. That said, I'm not claiming that the corrections I've made would really be responsive to the issue you raised here, although you might perhaps want to take a second look in the context of the correct section headers. Assuming that this does not satisfy your concern, please note that in the discussion above, I discussed this same issue with Bob K, and noted that a very reasonable option would be to move the sentence about the burden down. --Tryptofish (talk) 21:35, 28 September 2012 (UTC)

I think Yaris has a good point. The section would probably read better if we moved that sentence down. addendum: Perhaps so it read "All quotations and any material challenged or likely to be challenged must be attributed to a reliable, published source using an inline citation. The burden of evidence lies with the editor who adds or restores material, and is satisfied by providing reliable sources that directly support the material.^[3]" Maybe with a "the" before the bolded "material" in the last sentence. NTox · talk 22:36, 28 September 2012 (UTC)

Yes, I would tend to agree (although it's not something that I feel strongly about either way). I think that NTox's wording, including the "the", is probably the best way to go about it. Please check that the flow into the third sentence still works. (I think it's still OK, but I'd like to be sure.) Given preference for this revision of the proposal by BobK, Yaris, and NTox, and certainly no objection from me, I'd say that there is some consensus emerging for making the change. At the same time, several editors supported the existing proposal before the revision was suggested, so I'd like to ask everyone still following this discussion to look closely at this new idea. If need be, we can always leave the RfC open a little longer than we originally planned. --Tryptofish (talk) 14:12, 29 September 2012 (UTC)

Tryptofish's remark regarding my position needs clarification. I agree that the burden of evidence sentence shouldn't be at the beginning of the first paragraph, but I don't think it should be in the first paragraph at all. Just moving it to the second sentence of the first paragraph interrupts the flow of the paragraph.

I think the Burden sentence should be at the beginning of the second paragraph, as in the current version of WP:V. If we look at the current and proposed versions of the section, for example in the updated side by side comparison above, we can see Yaris's original point that the order in the proposed version is not appropriate. Compare that to the order of information in the current version in the updated side by side comparison above. --Bob K31416 (talk) 16:17, 29 September 2012 (UTC)

I'm not so sure. I mean, I wouldn't be vehemently opposed to what you suggest, but I like how BURDEN and CHALLENGE are so close to each other in the proposal, because they are, essentially two sides of the same coin. I don't think it really interrupts the flow all that much - we might have to change the word "source" to "sources" in the following sentence (as in "The source should be cited clearly..."), but that's about all IMO. Another thing to consider is that adding it to the second paragraph would make that paragraph really long, and the first one really short. NTox · talk 16:59, 29 September 2012 (UTC)

Your point about paragraph size brings up another issue. The second paragraph of the proposed version without the Burden sentence would still be the biggest paragraph in the policy, so it should probably be split up anyhow. Note that the second paragraph of the current version has a better size, even with its Burden sentence included. See above.

On a related issue, I think the approach of this proposal with multiple changes is unwieldy and I would prefer that we first change the Burden sentence in the current version to the proposed version without moving it. I think that would bring more editors on board. --Bob K31416 (talk) 17:50, 29 September 2012 (UTC)

Actually, it isn't like there are a lot of editors not on board. It's more a question of whether Bob K is on board. I'm wary about what had been a simple revision, switching the order of two sentences and adding the word "the", metasticizing into a complete rewrite. I do not support anything more than what I said earlier in this thread, at least not at this time. Instead of ballooning the RfC into every conceivable permutation of the wording, after a lot of people have already commented, mostly positively, I think we should pretty much work with what we have, and not make drastic changes to the proposal. Instead, let's keep in mind that the page won't get locked in stone the minute the RfC is over. Further changes can still be proposed and made after the RfC. --Tryptofish (talk) 20:23, 29 September 2012 (UTC)

I think your comment is full of errors and misunderstanding, so in response I'll just paraphrase one part of it, "I think we should pretty much work with what we have, and not make drastic changes to the proposal section." At least make the changes in steps, and not try to do it all at once with the questionable reassurance that anything anyone doesn't like can be changed after the proposal is implemented. --Bob K31416 (talk) 20:54, 29 September 2012 (UTC)

Then at least let's make sure that I understand one thing correctly. Unless we stop considering this proposal and instead conduct a series of smaller changes one-by-one, you are opposed to the change in this proposal, proposed by Yaris and NTox, correct? --Tryptofish (talk) 21:26, 29 September 2012 (UTC)

While I may have inspired that proposal, the proposal was by NTox. I think I am happy with the section as it stands after Bob's change. Maybe we should have another side-by-side of "The Bob version" and "The NTox version". Yaris678 (talk) 12:09, 1 October 2012 (UTC)

I would support doing that after the current RfC. A bunch of editors have already supported the proposal in this RfC, and it becomes impractical to keep changing the premise of the RfC while it is in progress. Again, the result of the present RfC is not going to be etched in stone. --Tryptofish (talk) 13:47, 1 October 2012 (UTC)

Yaris678, Thanks but actually I just implemented a part of Tryptofish's proposal that I agreed with and everyone so far has accepted. I think the fundamental problem with the way Tryptofish has proposed multiple changes is that they all don't have to be proposed together. Note that if editors see something they like, they may support the proposal without carefully considering whether the rest of the proposal is an improvement.

I agree with Tryptofish about the impracticality of continually revising the proposal but WP:V should not be used as a sandbox where you implement a version on the policy page and then work on it there. These problems can be solved and each of the changes can be given due consideration, by working on each of the multiple changes one at a time on this Talk page, and then implementing each one, revised as necessary. --Bob K31416 (talk) 15:56, 1 October 2012 (UTC)

Thanks for saying it that way, Bob. For what it's worth, let me say that I do not consider it to be "my" proposal, but more like my attempt to help by putting together the output of the many, many, many talk sections starting under the "WP:BURDEN" section header at the top of this talk page. Unless anyone objects, I expect that an uninvolved administrator will close the RfC in a few days, and we'll work with what we have at that point in time. Personally, I think that it will be just fine to consider, one-by-one, the ideas in this sub-thread, the ideas in the sub-thread about where in the source, and any other ideas about this section. My hope is that, following the very long process that led to the revisions of the lead section, this RfC will lead to revisions of BURDEN that will be accepted as having community consensus, in a more substantial way than BRD editing would have yielded, but not that it will result in anything that is etched in stone and cannot be improved further. --Tryptofish (talk) 20:00, 1 October 2012 (UTC)

I feel compelled to give a "thank you" to Tryptofish; I have found your dedication to this (very long) talk thread to be admirable, and that is authentic. A thank you also to Bob, for his continued involvement. I look forward to the implementation of the RfC results, and to the conversations after that. NTox · talk 02:02, 2 October 2012 (UTC)

It is not at all clear to me from the above discussion which version is being proposed as there are at least two new versions and the version currently on the policy page is not the version that was in place at the top of this RfC. -- PBS (talk) 18:34, 4 October 2012 (UTC)

The one that you numbered (1) remains the same throughout; it is what is being proposed. What you numbered (2) is a comparison of the way the page was at the start of the RfC with what is being proposed. Partway through the RfC, Bob K changed the page (ask him, not me), and provided what you numbered as (3) as an updated comparison. --Tryptofish (talk) 20:58, 4 October 2012 (UTC)

Addition to burden of evidence sentence

I do not see that my worries expressed at 09:38 on 14 September 2012 have been addressed, it is possible for the burden to be satisfied in way other than with "providing reliable sources that directly support the material" (detail is the previous posting) -- PBS (talk) 20:03, 6 October 2012 (UTC)

Hi, PBS. Speaking for myself, I am sorry I missed your comments in that edit. I'll offer my opinion on the issue: while I do think you have a respectable point, that a burden is not always created when material is challenged or likely to be challenged, at the moment I believe that addressing that point in this policy is not something that is strictly necessary. Perhaps I am showing my bias (I tend to fall in the WP:NOTBLUE camp, save the most obvious cases like "Paris is the capital of France"), but I believe that the percentage of statements that cannot in fact involve a burden are low enough so that we would not need to elongate the section further with exceptions and qualifications to accommodate them. To me, they are small enough issues to fall under the umbrella of WP:COMMON and WP:IAR. Worst case: addressing these exceptions in a policy as significant as this one may be a step in the general direction of encouraging people not to cite, when I think we should mostly be moving in the opposite direction. Moreover, there are MOS guidelines that inform already that citations are usually not needed in particular places (e.g., leads), so there is a great defense there if someone tries to add a bunch in places they don't need to be. Finally, taking the proposal to its logical extreme, I would say that statement itself is sound ("The burden of evidence lies with the editor who adds or restores material, and is satisfied by providing reliable sources that directly support the material"), because a burden (or a 'responsibility to cite') is not in fact created in the first place for those situations you describe, because there is existing consensus that citations are not necessary there. NTox · talk 23:06, 6 October 2012 (UTC)

Let me add that if anyone can propose a change that would simultaneously satisfy both PBS and Bob K, as opposed to satisfying one of them while making it worse for the other, I'd very likely be happy to support it. --Tryptofish (talk) 18:38, 7 October 2012 (UTC)

I think that the idea of the addition originally came about because Tryptofish ( and others?) was concerned that the burden of evidence has been interpreted to mean burden of evidence in general, not just for verifiabiity, and I agreed with Typtofish on that. Perhaps just adding "for verifiability" without the contested phrase would satisfy everyone? In other words, "The burden of evidence for verifiability...", without the contested phrase. Seems like adding just "for verifiability" and not the contested phrase, would concisely serve the desired purpose of not addressing burden of evidence in general and addressing only the burden of evidence for verifiability. It is possible that PBS wants the sentence to be for burden of evidence in general, not just verifiability, so if that's the case PBS may want to comment. --Bob K31416 (talk) 20:53, 7 October 2012 (UTC)

In a policy called Verifiability, I don't understand what advantage explicitly adding "for verifiability" would do. I think it would just make the sentence more verbose. For the same reason "challenged" does not need to become "subject to a reasonable challenge" as "subject to any challenge" would not be a reasonable inference to draw from the sentence. NTox you write "Moreover, there are MOS guidelines ..." precisely! This is one of three content policies and the edition of an explicit phrase "and is satisfied by providing reliable sources that directly support the material" undermines that MOS guidance and other exceptions I mentioned previously (see above PBS-09:38-14-September). -- PBS (talk) 17:09, 8 October 2012 (UTC)

Otherwise it may look like the burden of evidence on the adder is a general Wikipedia principle that is being applied here to the special case of verifiability. --Bob K31416 (talk) 17:29, 9 October 2012 (UTC)

I suppose we may disagree here, because as I wrote above, I don't think there really even is a burden to cite in a given place if there is a MOS guideline that says you should not cite there. Even if we did say that a burden is created to cite things in leads, navboxes, etc., when it is disallowed by the MOS, one could argue that it is indeed 'satisfied by providing reliable sources that directly support the material', if you cite the material where it's written in the body (for leads), and in the other article (for navboxes), etc. What would remain with the exceptions you discuss are the statements of common knowledge, which I still maintain have no burden to be cited at all (as per common consensus), but even if we said there was one, these statements IMO are so few as not to require accommodation in policy of this logical exception. NTox · talk 19:23, 8 October 2012 (UTC)

I understand you argument, but it contradicts those who are arguing that the addition is a clarification for new users (because your arguments are negating the meaning of "inline citation" defined in CITE in the section Types of citation "means any citation added close to the material it supports..."). If guidelines contradict policy then they are usually discounted (WP:POLICY), so far from clarifying the situation this addition to the sentence will cause confusion particularly for new editors for whom this phrase is being added. In the long term it will lead to arguments on article talk pages and on the pages of guidelines as editors start to insist that policy is followed while others try to explain that the wording in the guidelines should take presidents because the exceptions in the guidelines pre-date the change to policy, and anyway policy does not mean what it says ... . -- PBS (talk) 08:30, 9 October 2012 (UTC)

Hi, PBS. Sorry for the delay. Went on wikibreak for a bit after your comment. I get the logic on your end and would certainly be pleased to work out a solution that could satisfy all of us. At this point I think the best thing to do would probably be to look at some concrete proposals that would alleviate the problem you discuss (but perhaps after the RfC). I see Bob is working on this kind of thing below, although I do not recall if one of the three statements he has listed in the table originates from you. In all, I see the point that there is a technical contradiction between what I have said, and some of the things implied in BURDEN and WP:CITE. The theme here though is that while I recognize this technical issue, I have deliberately factored in a dose of IAR into the argument, because I don't see the technicality creating a huge number of problems. To be frank, I am just assuming that most people would have enough common sense to know that if the MOS specifically says not to put citations in leads, that they would understand that the very general advice in BURDEN would not have to apply. Perhaps, however, I am wrong in this assessment, and that arguments about such things on talk pages would ensue. I guess I can imagine it happening. Thus, if there is some way we can tighten the logic to pre-empt these scenarios while still keeping a readable policy, I'd be happy to help do that. NTox · talk 23:54, 19 October 2012 (UTC)

The three alternatives being considered for the Burden of evidence sentence are,

1. "The burden of evidence lies with the editor who adds or restores material. (current version in policy)

2. "The burden of evidence lies with the editor who adds or restores material, and is satisfied by providing reliable sources that directly support the material." (version in RfC proposal)

3. "The burden of evidence for verifiability lies with the editor who adds or restores material." (version with "for verifiability")

--Bob K31416 (talk) 10:02, 9 October 2012 (UTC)

Inline citations

• As a minor point "inline" should be replaced by "exact" or "specific". I do not know how it crept into the old version, but its out of line with general WP WP:REF policy-- Inline is used here to mean the <ref> </ref> style of referencing, which is not the only possible or acceptable style. The Harvard or Name-date style or the old MLAs style or anything else is just as good, as long as the references clearly indicate what specific fact is being referenced. DGG ( talk ) 23:42, 4 October 2012 (UTC)

I'm not so sure; if I have understood it correctly, WP:REF defines 'inline' pretty broadly ("any citation added close to the material it supports") and the particular section, WP:INCITE, goes on to describe multiple ways of adding inline citations: "Inline citations are most commonly added using either footnotes (long or short) or parenthetical references." I do however agree that in practice editors usually do understand 'inline' as 'superscript & footnote', but with the material in WP:REF I think we are okay using 'inline'. NTox · talk 23:57, 4 October 2012 (UTC)

I went and re-read WP:REF, and I think I agree with what NTox said. It seems to me that, for the purpose of WP:V, what we want is referencing that comes in the sentence where the material that it supports is found. (In other words, we want the sourcing to be linked to from the actual position of the material in the main text, rather than just having a list of sources at the bottom of the page.) My understanding of Harvard, MLA, etc. is that they are variations of the format of the footnote to which the link goes, and WP:INCITE also explicitly includes parenthetical referencing (in the text, instead of a footnote). So, unless I'm missing something, "inline" covers what is intended. In contrast, I think that "exact" or "specific" correctly describe what we want the sourcing to be like, but they fail to convey where we want to find the sourcing: close to the sentence(s) that are being sourced. --Tryptofish (talk) 00:36, 5 October 2012 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Proposal concerning grammar

I have fished the following proposal out of Archive 58:

I propose to rephrase this policy from the passive infinitive to the simple imperative or declarative.	It is proposed that this policy is rephrased from the passive infinitive to the simple imperative or declarative.
The simple imperative is clearer than the passive infinitive.	It is suggested that the wording might be clarified if it were changed from the passive infinitive to the simple imperative.
Examples	Examples
When reliable sources disagree, their conflict should be presented from a neutral point of view, giving each side its due weight.	When reliable sources disagree, present the conflict from a neutral point of view, giving each side its due weight.
Any material that requires a source but does not have one may be removed, and unsourced contentious material about living people must be removed immediately.	You may remove any material that needs but lacks a source. Please remove unsourced, contentious material about living people on sight.
All quotations and any material challenged or likely to be challenged must be attributed to a reliable published source using an inline citation.	Please attribute all quotations, and any material challenged or likely to be challenged, to a reliable published source. Please use an inline citation to do this.

In some cases I would like to simplify the sentence structure, so that each sentence is shorter and there are fewer ideas per sentence, but more actual sentences in the policy. In others I would like to use shorter words with more impact. The table above contains examples of both.

This has been raised before and the consensus (in archive 58) was cautiously in favour, but it was felt that I should await the RFC result so as not to complicate things, and this is what I have done. If anyone objects please do say so...—S Marshall T/C 20:50, 5 November 2012 (UTC)

• Sounds good to me. No objection from me. --Tryptofish (talk) 21:12, 5 November 2012 (UTC)
• Suggest that we go slowly... amend one sentence, wait a day or two (to see if anyone raises an objection or concern), then move on to the next. If we try to amend lots of sentences all at the same time, it is more likely that someone with an objection to one amended sentence will revert the whole thing (tossing out potentially acceptable edits with the one they object to). Blueboar (talk) 21:54, 5 November 2012 (UTC)
• OK (although Blueboar maybe right). Also for the first example shouldn't that be changed to "disagree . . . disagreement" or "conflict . . . conflict", if you don't use "their," just to make the reference clearer. Alanscottwalker (talk) 22:14, 5 November 2012 (UTC)
  • I endorse Blueboar's idea of doing it one-at-a-time. --Tryptofish (talk) 23:09, 5 November 2012 (UTC)
  • I also endorse taking it one at a time, waiting a bit after each, especially the first ones. But this also means its a small cautious decision and you should start today or soon. North8000 (talk) 00:22, 6 November 2012 (UTC)

Exceptional claims

I'd like a bit of advice about "Exceptional claims". I assume it means that extra care must be taken when an editor wants to introduce an "exceptional", "unusual", "out of the ordinary" claim into an article. So that if an editor wants to add that "Mr A owned a dog" into an article, it is not an "exceptional" claim and one decent source for that info would suffice. But if an editor wants to add that "Mr A had sex with his dog" that would be an "exceptional claim" and it would require "exceptional sources" or "multiple high-quality sources" before an editor could put it into an article. Further, I believe that even quoting someone making an "exceptional claim" requires "exceptional sources". So that if a newspaper reports that "Mr X said that Mr A had sex with dogs" we would want more than just Mr X's opinion before adding to an article "Mr X said that Mr A had sex with dogs". The reason I have formed this view is that although the vast majority of quotes contained in Wiki articles are benign, ordinary and undisputed what policy stops an editor from adding unsubstantiated and/or extreme quotes to an article simply by saying "Mrs X says Mr X is a pedophile". Shouldn't we ask for "multiple high-quality sources" other than Mrs X before linking Mr X with pedophilia for the whole world to Google?Momento (talk) 07:14, 8 November 2012 (UTC)

Yes, people have tried that dodge, and yes, consensus is that such quotes generally violate WP:BLP. If that claim by Mrs. X became the topic of a major controversy, though, it would still be reasonable to mention, though: think of Anita Hill re Clarence Thomas.—Kww(talk) 07:22, 8 November 2012 (UTC)

I think I had added multiple to Exceptional purely because of this reasoning. If my memory serves me right, I had also added challenged claims that are supported purely by primary or self-published sources or those with an apparent conflict of interest to Exceptional, again for this reason. A quote by someone that Mr. A had sex with dogs would be a primary quote. Exceptional sources by definition would mean those that adhere to the highest standards of independence in sources. That is one reason I had added the term multiple sources to even WELLKNOWN. Wifione ^Message 07:29, 8 November 2012 (UTC)

Thank you both.Momento (talk) 09:28, 8 November 2012 (UTC)

Sourcing, Tagging and Removing

One of the edits has gotten out of control and I suspect is going to be a stumbling block... so I think it needs further discussion:

• Original text: If you believe the material to be verifiable, then it is better to provide sourcing yourself, instead of removing the material. It has always been good practice to make reasonable efforts to find sources yourself that support such material, and cite them.

S Marshall started us off with a small edit (in italics):

• S Marshall's edit: If you think the material is verifiable, then it is better to provide sourcing yourself than to remove the material. It has always been good practice to make reasonable efforts to find sources yourself that support such material, and cite them.

A minor change. Well within his goal to just tweak for grammar etc. and not change the meaning of the policy. However, I think it does not accurately lay out the chain of what one should do... I tried a complete re-write

• Blueboar's edit: If you believe the material to be verifiable, try to provide sourcing yourself. If unsuccessful, it is best practice to leave the material with a tag (which alerts other editors to the fact that the material needs a citation) rather than to remove it.

Which I fully admit is a more major change... however, I think it is accurate and lays out the steps (first decide if you think the material is likely to be verifiable... if YES, at least do a quick google search to see if you can provide a citation... if unsuccessful tag so others can be alerted and have a go at verifying... and if that does not work, re-evaluate your assumption (the material may not actually be verifiable after all) and remove.

• Kww's edit: If you believe the material to be verifiable, try to provide sourcing yourself. If unsuccessful, it may be either tagged or removed.

This removes the link to WP:PRESERVE (or WP:FIXTHEPROBLEM if you prefer). We seem to have a difference of opinion on what to do when you think the material is verifiable, but you personally can not find sources to support it. Please discuss. Blueboar (talk) 23:51, 7 November 2012 (UTC)

I think that we need to write a solid informational page or guideline on this issue. We shouldn't put too much detail here, but people ask about this all the time, and the answer is complicated. What to do and when to do depend on (at least) the specific content (BLP or not?), the article's traffic, the article's development, and your personal assessment of the likelihood that the material is accurate and relevant. I don't think that any three sentences will be able to provide the level of detail that less-experienced people need. WhatamIdoing (talk) 01:01, 8 November 2012 (UTC)

WP:PRESERVE was best practice when Wikipedia was still small and growing and had not yet encountered any major BLP problems. As of 2012 and going into 2013, any emphasis on upholding PRESERVE basically ignores the last decade of changes. PRESERVE is no longer best practice as we are more concerned with the impact of leaving unsourced material in an article, so it is astonishing to see editors trying to add it back in as if we still treated it as if was of primary importance. I think I can say with some confidence that we no longer prioritize preserving unsourced content over removing it. Viriditas (talk) 01:20, 8 November 2012 (UTC)

In policy that may be true... but I have to disagree when it comes to actual practice. In practice, when the typical Wikipedian comes across an unsourced statement that they think is verifiable, they don't even realize there is a problem... they will not even think to question it (much less tag or remove it). They will simply leave it in the article (unsourced). It is the rare editor who actually goes out and looks for unsourced (but verifiable) information.

Let's call a spade a spade... the issue here isn't answering the question "what should you do if you come across unsourced information that you think is verifiable" (the answer to that is easy: If you notice it, fix the problem)... the real issue here is answering the question: "what must you do if you come across unsourced information that I think is verifiable". Blueboar (talk) 01:52, 8 November 2012 (UTC)

The change is far superior to the previous version citing PRESERVE and reflects current best practice. "If you believe the material to be verifiable, try to provide sourcing yourself. If unsuccessful, it may be either tagged or removed." Contrast this statement of fact with "If unsuccessful, it is best practice to leave the material with a tag (which alerts other editors to the fact that the material needs a citation) rather than to remove it." That is not true. Best practice is to remove unverified content, not preserve it. It is no longer 2004. Viriditas (talk) 02:03, 8 November 2012 (UTC)

• I think the issue here is that different un- or under-sourced material needs to be treated differently. Aspect of a fictional element? Leave it and tag it. BLP material? Nuke it from orbit, and revision delete as appropriate. I'd endorse a rewrite that distinguishes between innocuous material where tagging and retaining is appropriate and potentially harmful material that should not be retained until and unless sourced. Jclemens (talk) 01:55, 8 November 2012 (UTC)

Here is why firm and fast rules with no exceptions never work... Let us suppose someone adds the fact that Obama won the 2012 presidential election to his bio article... without supplying a source. The information is easily verifiable. Yet this is a BLP... Would you really Nuke this from orbit? Would you spend a few minutes to slap in a source yourself? Or would you simply ignore the problem and leave the sentence unsourced? Blueboar (talk) 03:07, 8 November 2012 (UTC)

That, my good sir, is a textbook example of when WP:IAR clearly applies. :-) Jclemens (talk) 03:21, 8 November 2012 (UTC)

edit conflict, but it reinforces what Blueboar just said. My strongest feeling is just from a process standpoint. Nobody should be able to impose their preference by simply being the most aggressive editor. So, from a process standpoint, we should go to the version before the changes while we discuss it. Lets keep this civilized. I plan to revert it to that again solely for that reason.

Unfortunately, misuse of policies is a far more prevalent problem than violation of policies. In this area, here are the two common misuses (which go in opposite direction) :

1. Use it it IN COMBINATION WITH hairsplitting on source quality as a magic bullet to know out material to pursue a POV or pissing war.
2. Use and "preserve" or "tag first" wording to try to keep crap in. If the remover doesn't have a clear-cut policy to back up their removal of total crap, they will have a hard time doing so.

I think that there are a lot more severe hard-to-resolve abuses under #1 than #2 so my comment would be for keeping the "better to tag first" type wording. Sincerely, North8000 (talk) 03:12, 8 November 2012 (UTC)

What is "best practice is opinion" and editor based rather than community based. In the multiple possible variations an editor encounters, actions that best benefit that article and so the encyclopedia can and do vary, so one size cannot fit all. If this policy is going to prescribe best practice it must be varied enough to cover multiple contingencies and even then there must be room for editors to make individual judgments based on the situations they encounter. Probably only BLP should be restrictive because of the damage we can do to real people with unsourced content. We're bleeding into dictating behaviour here and that's tricky seems to me.(olive (talk) 03:22, 8 November 2012 (UTC))

We've had a fair amount of edit warring on this article in the past. I wonder if KWW would consider dealing with his proposed changes on a talk page before adding them.(olive (talk) 03:24, 8 November 2012 (UTC))

You only reverted my change, and didn't go back to the normal version. I have reverted to Oct. 19, before the latest flurry of edits. WP:BURDEN has never recommended tagging. It shouldn't. Tagging is a generally bad practice: the only time unsourced material should be retained is when it is extremely difficult to provide intelligible coverage of the topic without it. WP:PRESERVE should not be used as a pretext to keep unverifiable material in articles.—Kww(talk) 03:25, 8 November 2012 (UTC)

Kww... the sentences we have been editing does not address unverifiable material ... it is talking about verifiable material that happens to not be sourced. There is a difference. Blueboar (talk) 03:31, 8 November 2012 (UTC)

Edit conflict, but again it reinforces what Blueboar just said. On this first part of your post, I'm confused but think that you may be right......put it at the October 20th version while we discuss. On the second part of your post, you conflated 2 very different situations. Material that is, as of first notice, unverifiED at the moment, and material that is unverfiable. North8000 (talk) 03:36, 8 November 2012 (UTC)

A distinction without a substantial difference. If someone notices uncited material, he should look for source. If he cannot find a source, the correct action is generally to remove it. The situations where it should be retained with a tag are extremely rare, and tagging is a bad practice, not a good one. If, after removal, someone can find a source for it, he should put it back with an inline citation, as this policy has always required. On a related topic, restoring with a tag is beyond bad practice: it's always been forbidden by this policy, which requires an inline citation for challenged material.—Kww(talk) 03:49, 8 November 2012 (UTC)

@Kww. I didn't revert you. Did I misunderstand your comment above.(olive (talk) 03:57, 8 November 2012 (UTC))

Indentation is your friend, Olive: I was responding to North8000.—Kww(talk) 04:07, 8 November 2012 (UTC)

Your indentation comment doesn't make sense to me but thanks for responding. I wanted to make sure I was understanding what you were saying if it was directed at me. I'm responding here and indenting deliberately :) so as not to throw this response into the discussion.(olive (talk) 21:20, 8 November 2012 (UTC))

Kww, here is the common situation that I'm talking about. Person A is in a POV or pissing war with person B. Person B puts something in which IS sourced, and which Person A knows is accurate, and which they do not even question the accuracy of. But they find some wiki-imperfection in the sourcing and use that (in conjunction with wp:ver) to simply delete the material. North8000 (talk) 04:01, 8 November 2012 (UTC)

Kww, I didn't revert you because I think that you did what I was attempting to do. Put it at the pre-change version while we discuss. North8000 (talk) 04:03, 8 November 2012 (UTC)

Then fix the imperfection or get consensus that the source does indeed support the material. There's no rush to keep material included, and when there's a question on the quality of the source, we should err on the side of caution and remove the disputed material.—Kww(talk) 04:07, 8 November 2012 (UTC)

I heard of an example of this recently: Editor A added sourced, fully verifiable information, and editor B removed it (solely) because the citation wasn't perfectly formatted. Do you believe that Editor B was helping us write an encyclopedia by removing material due to an "imperfection"? WhatamIdoing (talk) 02:42, 10 November 2012 (UTC)

• The principal purpose of WP:PRESERVE has to do with editor retention, and it is of higher importance today than ever before because for the first time in Wikipedia's history, the number of active editors is declining.—S Marshall T/C 07:54, 8 November 2012 (UTC)
  • I have never heard that before. As far as I can tell, PRESERVE has nothing whatsoever to do with editor retention, a problem that has, IMO, nothing to do with PRESERVE or any other thing related to it. Viriditas (talk) 08:16, 8 November 2012 (UTC)
    • Oh, that's interesting. What do you think the purpose of PRESERVE might be?—S Marshall T/C 08:19, 8 November 2012 (UTC)
      • Older versions from 2003 lend some support to your claim, with statements like: "With large deletions or replacements, it might be better to suggest changes in a discussion, lest the original author get discouraged and quit posting. One person's improvement is another's desecration, and nobody likes to see their work just flushed without warning." However, this is explained in terms of how to edit a wiki and what kind of editing style might come in handy. Editor retention rates have been linked to many things, including attempts at new article creation, the learning curve, the narrow demographic, the training of admins instead of editors, the lack of civility, etc. But it isn't 2003, and we are more concerned with improving good content rather than preserving bad content. Therefore, I suggest that our priorities have changed. In terms of verification, content should be preserved if an attempt at verification hasn't been made, but I think the older wording implying it is best practice to tag after failing verification is wrong. Viriditas (talk) 09:36, 8 November 2012 (UTC)
        • As I said, editor retention is the reason for it. Contrary to what you say, WP:PRESERVE has never been about preserving "bad" content. It has always been about preserving appropriate content. It's trying to ensure that good edits should stick. Any revision of WP:PRESERVE should retain that element.—S Marshall T/C 10:37, 8 November 2012 (UTC)
          • Understood, but from where I sit, it doesn't reflect current best practice but rather a peculiar software engineering approach to "prose as code" that doesn't actually demonstrate editor retention in practice, only in theory. I never said WP:PRESERVE was about preserving bad content, but I maintain that adding tags instead of removing bad content results in a poor outcome. We need to remove poor content, not tag it. PRESERVE sounds great when you've got 1000 articles and 10 editors, but it doesn't scale. Viriditas (talk) 10:59, 8 November 2012 (UTC)
            • Where content is unfixable, I certainly agree that it should be removed rather than tagged. I think there's a vast grey area between the kind of unfixable content that should be removed on sight and the kind of appropriate content that should be retained unedited. WP:PRESERVE ought to be what governs this grey area. It's saying that where there's a choice between improving prose by adding citations and qualifiers, or by wholescale rewriting, it's the lesser change that should be preferred. It needs a big fat within reason, and I'll once again bemoan the fact that WP:Editorial judgment is still a redlink.

              I'm concerned to tone down wholesale revisions to WP:PRESERVE because PRESERVE serves to restrain people with WP:OWN issues and those who love the sight of their own prose, while encouraging the more hesitant to participate in the knowledge that their contributions won't be excised or overwritten without good reason.—S Marshall T/C 12:03, 8 November 2012 (UTC)

              • Those are interesting observations from a theoretical perspective, but I don't see any practical benefit. People with OWN issues aren't restrained by PRESERVE, they are restrained by talk page RFC's and ANI reports. Timid editors are encouraged by helpful editors who are willing to guide them through the process. As far as I can tell, PRESERVE is a relic of an earlier time, and I'm tempted to write Wikipedia:Don't preserve content based on my experience sifting through hundreds and thousands of articles with maintenance tags littering the prose because nobody has the time to fix it. I'm afraid that based on my practical experience, I'm heavily biased against PRESERVE. Viriditas (talk) 03:30, 9 November 2012 (UTC)

Different situations

Just to be clear... there are 3 different situations we are discussing:

1. Unsourced Material you believe is unverifiable
2. Unsourced Material you believe is verifiable
3. Unsourced Material you are unsure of (ie you have no idea whether it is verifiable or not)

Here is how I think most people deal with each of the three:

1. Depending on the specifics, they either tag or remove... tending towards removal.
2. People usually just ignore the issue entirely and leave the material unsoursed, but as a second choice they tag it so others will find a soruce for it... few people remove (except potentially negative info in BLPs.).
3. This is where it gets really tricky, and what I think this debate is really about. Some people think we should err on the side of preservation (taking the attitude that the material probably is verifiable)... Others think we should err on the side or removal (taking a more pessimistic attitude, that it probably isn't verifiable)... this is where PRESERVE/FIXTHEPROBLEM comes into play for me... I tend to want to find out whether it is verifiable or not... and so will at least do a quick google search. I think that is best practice, but I do understand that others take different attitudes. Blueboar (talk) 04:12, 8 November 2012 (UTC)

And we aren't arguing that people should do a search before acting: we agree on that. WP:PRESERVE does, indeed, encourage us to check for sources, and blindly removing material without a check for reliable sources shouldn't be encouraged. I'm simply saying that when that search fails, this policy shouldn't state that tagging is best practice: sometimes material should be removed, sometimes it should be tagged. Both choices have adherents, and this policy shouldn't say that one choice is best.—Kww(talk) 04:21, 8 November 2012 (UTC)

I an in favour of demoting WP:Editing policy to a guideline. This problem of preserve has been used to undermine WP:V and WP:OR for years, see my last comments under Wikipedia talk:Editing policy/Archive 3#Scope. The current wording of WP:PRESERVE was an attempt to reduce its impact and bring in more into line with the three main content policies (and BLP). If WP:PRESERVE was in a guideline then its arguments could be used as a supplement to the content polices and could not be used to undermine them. -- PBS (talk) 18:55, 8 November 2012 (UTC)

I don't think there is a need to demote WP:EP. But if it can be used to "undermine" this policy then there is a problem that needs to be fixed. We need to hold a centralized discussion so the two Policies get back into sync. Blueboar (talk) 20:09, 8 November 2012 (UTC)

I agree with Kww above. Assigning "best" to an action is for the most part opinion based, and is dictating behaviour for situations that may not be comparable. Better to lay out the possible steps, (given an at least preliminary search has been carried out), editors can take, describing possible scenarios. These would be guides rather than dictating action. (olive (talk) 22:01, 8 November 2012 (UTC))

Blueboar let me give you a practical example of how having PRESERVE as a policy undermines/contradicts WP:BURDEN. There was a paragraph in the article "Battle of Berlin". An edit was made 17:55, 17 November 2004 which included the sentence "That the battle ended after a week of heavy fighting was mainly because the German supply dumps were located outside the Inner Ring and were captured quite early in the battle by the Soviets, ..." by an IP address which stopped editing long ago, and long before general references inline citations were the norm in such articles. There were two problems with that sentence it expresses an expert opinion (OR) and it contains a fact "the German supply dumps were located outside the Inner Ring". I added a "citation needed" to the paragraph on 16 July 2007. The sentence was cut and pasted into the more specific article "Battle in Berlin" and then removed by me on 10 January 2009. The dumps sentence had been "preserved" for four years of which it was tagged "citation needed" for a one and a half years. I removed it because it was not sourced (PROVEIT) in an article in which all the rest of the text was sourced. It was not that I did not think it possible for it to be true, but it could also have been false and to quote the Jimbo footnote in BURDEN "There seems to be a terrible bias among some editors that some sort of random speculative 'I heard it somewhere' pseudo information is to be tagged with a 'needs a cite' tag. Wrong. It should be removed, aggressively, unless it can be sourced."

It was not that the facts in the sentence were unverifiable, (In fact I recently came across a source that verifies it (Tony Le Tissier Race for the Reichstage (2010) [1999] p. 74), but leaving unverified text in an article that turns out to be unverifiable is worse than removing unverified text that turns later is found to be verifiable (In World War II the BBC was trusted because it told the truth (even if not all the truth), German wireless was not trusted because it was know to lie). Part of the process of building a useful encyclopaedia is to have articles that the public expect to be factually accurate. Without that expectation Wikipedia will be seen a second rate source.

The problem with PRESERVE is that in this case it could have been used to justify retention "As long as any of the facts or ideas added to the article would belong in a "finished" article, they should be retained" because the position of the arms dumps is of relevance in a "finished" article and just because the facts had not been verified did not mean that they could not be. The trouble is without the complete set of reliable sources (which even the best university libraries will not have) proving that it is unverifiable is impossible (it is much easier to prove verifiability than to disprove it). So the core of meaning of PRESERVE contradicts WP:BURDEN (which puts the emphasis the other way around).

This contradiction is emphasised by the emphasis placed on "migh" in another sentence in the next section of the editing policy (linked with CANTFIX): "Several of our core policies discuss situations when it might be more appropriate to remove information ..." There is no "might" about it!

An easy fix for this contradiction would be to make the editing policy an editing guideline. The next best alternative would be to qualify the sentence: "As long as any of the facts or ideas added to the article would belong in a "finished" article, they should be retained" with "As long as any of the facts or ideas added to the article would belong in a "finished" article, they should be retained if they meet the requirements of the three content policies and WP:BLP" or a similar a qualification. Ie if the text does not meet NOR, V and NPOV then it need not be retained. This would stop the possibility of PRESERVE being used to contradict the intention behind PROVEIT. --PBS (talk) 17:23, 9 November 2012 (UTC)

You're not describing a problem with WP:PRESERVE. You're describing a problem with editors. WP:PRESERVE says "Preserve appropriate content". If the content doesn't meet NOR, NPOV or V, then it's not appropriate.—S Marshall T/C 17:32, 9 November 2012 (UTC)

"Preserve appropriate content." is not defined in PRESERVE, but is is immediately followed by the sentence I have quoted about "'finished' article" which seems to me to define in Editing policy what is appropriate content in a non finished article. How do you extrapolate that the text currently in an article will not meet NOR, NPOV or V at some future (but undetermined date) when the article is finished, even if it does not meet those policies as it is now. Why should it it not be preserved until such time as it does? The example I have given above, -- where I have removed text because no one provided a source (and so failed PROVEIT) -- still met the PRESERVE requirement of the article as it was then, because that arms dump fact is in fact verifiable that should probably be included in the "finished" article. Your point would carry more weight if the Editing policy was a guideline. As it is a policy who is to say that three content policies carry more weight in an unfinished article (and aren't they all) than the Editing policy? -- PBS (talk) 18:56, 9 November 2012 (UTC)

PBS: Omitting facts from articles distorts and misinforms, as well. Alanscottwalker (talk) 18:25, 9 November 2012 (UTC)

A stub that meets the 3 content polices does not distort or misinform even though it is not a complete article and lacks information. PRESERVE does nothing to prevent the omitting of facts, but its current wording can be used to negate the three content polices. -- PBS (talk) 18:56, 9 November 2012 (UTC)

A stub article is suppose to carry a warning that it is a stub. Alanscottwalker (talk) 19:29, 9 November 2012 (UTC)

I think that the discussions about EP need to happen over at WT:EP. WhatamIdoing (talk) 02:49, 10 November 2012 (UTC)

I like Blueboar's outline of the different situations, and I think we need a page that discusses them in depth. Anybody have an idea for a title? Maybe WP:Dealing with unsourced material? WhatamIdoing (talk) 02:49, 10 November 2012 (UTC)

WhatamIdoing this is not a discussion about Editing policy, the point I am making above with an example is about Blueboar's "3 Unsourced Material you are unsure of (ie you have no idea whether it is verifiable or not)" and these two versions are substantially different on that issue:

• S Marshall's edit: If you think the material is verifiable, then it is better to provide sourcing yourself than to remove the material. It has always been good practice to make reasonable efforts to find sources yourself that support such material, and cite them.

A minor change. Well within his goal to just tweak for grammar etc. and not change the meaning of the policy. However, I think it does not accurately lay out the chain of what one should do... I tried a complete re-write

• Blueboar's edit: If you believe the material to be verifiable, try to provide sourcing yourself. If unsuccessful, it is best practice to leave the material with a tag (which alerts other editors to the fact that the material needs a citation) rather than to remove it.

The use of "If unsuccessful, it is best practice to leave the material with a tag" undermines PROVIT, and it leans the wrong way. I think that the former wording is better and should not be replaced with Blueboar's. -- PBS (talk) 16:11, 12 November 2012 (UTC)

How does it undermine PROVIT? When you tag something that you think IS verifiable, you are not challenging the verifiability of the statement, you are just leaving a note for other editors saying that you think the statement would benefit from a citation. Let's take an example... suppose I am surfing Wikipedia and come across an article on a historical battle, that says "The battle took place the same day that the King was crowned"... It is unsourced, but I remember my college history professor stressing this fact. I don't want to remove the statement. However, I am a busy fellow and don't want to spend a lot of time researching a topic I am only marginally interested in. I can conduct a cursory google search in the hopes that I will find a source ... but suppose I don't find anything I think is reliable. I have spent as much time as I want to spend trying to find a citation. However, I hope that someone else will be able do so. So why notleave a "citation needed" tag, to alert other editors to the fact that the statement could benefit from a citation? Blueboar (talk) 18:09, 12 November 2012 (UTC)

Re "When you tag something that you think IS verifiable, you are not challenging the verifiability of the statement, you are just leaving a note for other editors saying that you think the statement would benefit from a citation." — That may not be the motivation for tagging in most cases since the use of {{Citation needed}} is explained at the top of that template page by the following excerpt.

"{{Citation needed}} (also known by the redirects {{Cn}} and {{Fact}}) is a template used to identify questionable claims in articles that lack a citation to a reliable source."

I tend to follow this suggested use of the tag and not leave a tag if I think the material is verifiable. Seems like there is enough unsourced material in Wikipedia that I am unsure about to take up the time I care to spend leaving tags. --Bob K31416 (talk) 19:37, 12 November 2012 (UTC)

Blueboar the point is that WP:BURDEN unequivocally places the emphasis on the person who wants to restore deleted material to provide an source that supports the material. Your proposed wording undermines that. See my example above (where I deleted something after it had remained unsourced for years and then tagged for a year and a half). If someone after my deletion had restored the sentence, then BURDEN states that they must supply an inline citation.

Your wording leaves that open to debate. It undermines that important part of the verification policy which has been in the policy since August 2005. I think that in the long term the project is better off with removal of unsourced material than with leaving unsourced material lying around for years. Which is why I think that the editing policy should be demoted to a guideline, and if that is not done then PRESERVE should be qualified with "[material should be retained] if [it] meet[s] the requirements of the three content policies and WP:BLP" because BURDEN has gone a long way to improving 1,000 of articles through creative destruction and reconstruction cycle, and PRESERVE can be used to undermine BURDEN. -- PBS (talk) 21:31, 13 November 2012 (UTC)

Which version while we talk?

Kww, you have described 2 different date versions which you reverted to. You said the October 20th version in the edit summary and now you said October 19th. There is a very big difference because on October 20th the result of the RFC and a lot of work and discussion was implemented. Unilaterally undoing that would not be right. IMO we should go to the Oct 20th version while we discuss. North8000 (talk) 10:43, 8 November 2012 (UTC)

On Nov 1 I made an edit that has not been disputed and is not related to the present disputes.[15] Please be careful of making destructive blanket reverts. Thanks. --Bob K31416 (talk) 14:28, 8 November 2012 (UTC)

This is exactly why I wanted to go slowly and step by step. May I suggest that the appropriate revert would be to this edit... which would keep the edits that have nothing to do with the sentence under discussion (edits that I think have achieved consensus). Yes, I realize that reverting to this dif would return language of the sentence under discussion to a version that is objected to... but that is a temporary situation, and it will be resolved to a consensus version as soon as we finish discussing and actually reach a consensus. As long as we continue discussing, there is no serious harm in leaving the page at the "WP:wrong version". Blueboar (talk) 16:27, 8 November 2012 (UTC)

Blueboar, this will take more time than I have to figure out, but I know that you take a careful near-middlle-of-the road approach to these things and I support whatever you propose for the "during discussion" version. North8000 (talk) 16:37, 8 November 2012 (UTC)

Thanks for the kind words... I just don't think we should scrap acceptable changes to other sentences (changes that have achieved consensus) while we debate and reach a consensus on the language of this sentence. Blueboar (talk) 16:49, 8 November 2012 (UTC)

I haven't been around for awhile, but will jump in again, and agree that the consensus gained in the lead took a lot of effort and time so while discussion is ongoing on one section perhaps we can leave the rest in place. and I also trust Blueboar to make a sensible decision on this. I did make a edit based on simple logic but no worries at all reverting it.(olive (talk) 21:26, 8 November 2012 (UTC))

S Marshal has more or less set us where we should be... I suggest we leave it at that while we talk. Blueboar (talk) 21:49, 8 November 2012 (UTC)

I smiled at the edit summary here: [16]. Actually, 2017 may end up being overly optimistic. Please let me suggest, entirely in a friendly spirit, not to attempt any substantive changes while these incremental changes are being made. Just simplify wording, without changing any meanings whatsoever. (And simpler wording isn't an end in itself, so getting the meaning right is more important than simplification.) --Tryptofish (talk) 21:19, 12 November 2012 (UTC)

• Actually, I do see simplifying the wording as a valuable goal in its own right. But yes, I'm also trying to get the meaning right. Once I've finished a preliminary pass and made the changes nobody's going to revert, I may return to some of the changes that did get reverted for further discussion, on the grounds that ambiguities aren't a good thing—we should be aiming for one simple sentence per meaning, and one simple meaning per sentence. (I know this isn't achievable in English, but the attempt is worthwhile).—S Marshall T/C 02:31, 13 November 2012 (UTC)

The goal is indeed worthwhile... and you are doing a good job. I think the few reverts are cases where people felt that in the process of making a "simple sentence" you lost the intent of the original, less simple, wording. revisiting is worth doing, because there may be some other simple sentence that would keep the intent. (although some concepts can not be explained clearly in one "simple sentence".) Blueboar (talk) 14:13, 13 November 2012 (UTC)

VNT... parenthesis or footnote

OK... given a recent series of edits, it is time to address an outstanding issue concerning the Verifiability, not Truth statement.

Back in July we had an RFC (see: Wikipedia:Verifiability/2012 RfC), which was closed as follows:

• After examination of the discussion and close consideration of the arguments put forward, closing administrators Coren, jc37 and Sandstein agree that this request for comments concludes with a result of "There is consensus to implement the proposed wording of option D as the lead section of Wikipedia:Verifiability". The detailed closing arguments and rationales of each closer can be read here.

Option D is the current wording of the lead.

However... the closers added an addendum:

• We note that there remain open questions about the exact status and place of the phrase "Verifiability, not truth", and recommend that the community continue discussion on these points.

It is time to settle these open questions. In the version that was chosen, VNT is mentioned in the policy... but it is relegated to a footnote (which sort of "hides it away where it will not be noticed". I think VNT is an important enough concept that we should make it a bit more visible... which can be done by simply turning the footnote into a parenthetical. The language remains the same... it is simply a matter of "status and placement". thoughts and comments please. Blueboar (talk) 17:04, 13 November 2012 (UTC)

• Support Blueboar's change of a footnote to a parenthetical. Wikipedians have used the phrase for many years, and there was no support for completely removing the phrase from the policy, only from the lead. KillerChihuahua^?!? 17:30, 13 November 2012 (UTC)

Well... lets be honest... there was some support for completely removing it... just as there was some support for retaining it prominently. The problem was that neither view gained enough support to overcome the support of the other. We literally came to no consensus on the issue... so it was left as an "open question". Blueboar (talk) 17:59, 13 November 2012 (UTC)

• Totally out of the question. It took nearly two years and a ludicrous amount of work to rid the policy of that toxic phrase and to reinstate it in any way at all is a horrible mistake of the first order.—S Marshall T/C 17:41, 13 November 2012 (UTC)

S Marshal... we did not "rid the policy of that toxic phrase" (it remained in the policy... placed in a footnote) nor does my proposal "reinstate" it (you can not "reinstate" something if it was never removed) ... Please note that my proposal does not change a single word of the policy. All it changes is placement... which is exactly what the closers of the RFC asked us to work on. Blueboar (talk) 17:59, 13 November 2012 (UTC)

• It's not getting out of the footnote and back into the text without a supermajority at a sitewide RFC.—S Marshall T/C 18:16, 13 November 2012 (UTC)

Your opposition is noted. Question... could you explain why you are so adamant about this not being a parenthetical? What do you see as being the difference between a footnote and a parenthetical comment? Blueboar (talk) 18:19, 13 November 2012 (UTC)

• I've always regretted not being allowed to excise that utterly brainless slogan entirely, but at least being buried down there in the footnote reduces the harm that VNT can do by (a) reducing the number of people who actually read it and (b) reducing its prominence and therefore its impact on the people who do read it.—S Marshall T/C 19:01, 13 November 2012 (UTC)

OK... next question... What do you see as being the harm of telling editors that: This principle has been historically and notably expressed on this policy page as "the threshold for inclusion is verifiability, not truth."? Blueboar (talk) 19:31, 13 November 2012 (UTC)

• What could possibly be the benefit of putting this historical, now-superseded text back into the body of the text? We've replaced it with a nuanced discussion that has all the benefits of VNT without the drawbacks.—S Marshall T/C 19:48, 13 November 2012 (UTC)

Hold on... I am NOT suggesting that we put the "historical, now-superseded text back into the body of the text". I am suggesting that we make the current notation that explicitly says the VNT phrasing is historical more prominent... by shifting it from a footnote that no one will see to a main text parenthetical that they will. I am surprised that you don't find doing this beneficial. It helps emphasize that that people should stop using the phrase.

Now, could you answer the question I asked... what harm do you see arising from moving the current sentence from a footnote to a parenthetical. Blueboar (talk) 15:44, 14 November 2012 (UTC)

Blueboar, If you and S Marshall can pardon my intervention in your discussion, I think at least one reason for not moving it from the footnote is that it would be unnecessary clutter as a parenthetical statement in the main text of the lead. --Bob K31416 (talk) 17:00, 14 November 2012 (UTC)

• Indeed. The thrust of everything I've been trying to do since the RFC has been to simplify and de-clutter the policy text, to boil it down to its essentials so that a new person can read the policy and grasp it in one session. VnT is and has always been the biggest single obstacle to that because of cognitive dissonance.—S Marshall T/C 18:45, 14 November 2012 (UTC)

I might agree if the proposal was to restore VNT to the lead... but that is not what is being proposed. What is being proposed is to make a statement saying that VNT is historical more prominent. I would think that, as someone who disliked VNT, you would approve of that. Blueboar (talk) 15:24, 15 November 2012 (UTC)

• The result of the large RFC put the VnT phrase in a footnote. Considering the controversial nature of the phrase, and the extensive effort that went into the previous large RFC, another large RFC may be needed if anyone wants to change the positioning of the phrase, which could include the option of completely removing the phrase from the policy. --Bob K31416 (talk) 20:06, 13 November 2012 (UTC)

• No rush - let it sit for a year or so and see if any cases crop up to show we really need to have it intext. LeadSongDog come howl! 20:53, 13 November 2012 (UTC)

No rush on my part... I raised it simply because it became the subject of a brief flurry of editing/reverting... the closers at the RFC specifically noted that this was an issue we needed to discuss further, so I figured this was a good excuse to start that recommended discussion. Blueboar (talk) 15:20, 14 November 2012 (UTC)

• Short version: I agree with LeadSongDog that there's no rush.
• Long version: I was one of the more conspicuous defenders of VnT in the RfC (in fact, I think I proposed the parenthetical version during the RfC!), so I thank Blueboar for raising the question. But I have to say that I was powerfully persuaded by the consensus of the community, and I really have not been at all unhappy with the outcome. I'm now pretty close to S Marshall and Bob K on the issue. I do not feel that there is much need to move it out of the footnote, just so long as we don't delete the footnote as well. There are other, milder, approaches that are worth taking a look at, though. One issue that came up during the RfC, and I think was well within the closers' statement, is the way that the footnote is worded. Some of us felt at the time that it came off as kind of patronizing. How about just shortening the footnote, by deleting the words "historically and notably"?
  • Thus: "This principle has been historically and notably expressed on this policy page as "the threshold for inclusion is verifiability, not truth." See the essay, WP:Verifiability, not truth."
• --Tryptofish (talk) 21:34, 13 November 2012 (UTC)
• It's absolutely vital to keep the word "historically", which tells the reader that VNT has been replaced by the current version and that we no longer say "verifiability, not truth". I'm quite amenable to removing "and notably".—S Marshall T/C 07:58, 14 November 2012 (UTC)

  I still say it. KillerChihuahua^?!? 15:30, 14 November 2012 (UTC)

• Which is unfortunate, but at least there isn't a policy that backs you up. At the RFC Wikipedians agreed that there are quite a few "thresholds for inclusion", and that the truth matters.—S Marshall T/C 18:47, 14 November 2012 (UTC)

On the other hand, there is a very nice essay page that can be pointed to when someone wants to say it. Blueboar (talk) 19:34, 14 November 2012 (UTC)

S Marshall, would your concern about "historically" be addressed by the fact that it would still say "has been", putting it in the past tense? It seems to me that it's self-evident that that the present version, whatever the present version may be, has replaced all previous versions. Other than the footnote, WP:V does no long say VnT, and any reader of the policy page can plainly see that. (By the way, this is a good reason why leaving it as a footnote has some advantages over making it a parenthetical phrase.) The next sentence of the footnote tell the reader who wants to learn more that they can read the VnT essay, and the essay makes it very clear that this is a past usage. Another thought: delete the words that I suggested deleting, but also change "has been expressed" to "used to be expressed", which might make the past tense more obvious. That would be fine with me. --Tryptofish (talk) 01:20, 15 November 2012 (UTC)

• Oppose The version chosen by the huge RFC included the current handling of VNT. I'm not saying that it precludes other handling (which could include something like your proposal, but could also include complete deletion of any reference to VNT, including deletion from the footnotes) but it certainly puts a lot of weight towards handling as it currently is. North8000 (talk) 18:24, 15 November 2012 (UTC))
• Oppose because VNT is garbage and the closers comments are less important than the view of the surpermajority in two massive RfCs. But, more pragmatically, agree with the view that we should be discussing this when something happens that suggest we need to change things. Formerip (talk) 01:34, 19 November 2012 (UTC)
• Oppose - For a couple of reasons: (1) the VNT can be rather confusing to new editors; and putting it up top (even if it is within parenthesis) could cause problems; (2) The new version of WP:V, as LeadSongDog says above, should sit for about a year, and then we can come back to it. More time needs to elapse. --Noleander (talk) 22:38, 19 November 2012 (UTC)

or (slightly) better footnote

I'd like to propose a much less ambitious alternative proposal, one that really doesn't make any substantive changes, or even any changes in how the footnote mention appears. I suggest revising the footnote...

From:

"This principle has been historically and notably expressed on this policy page as "the threshold for inclusion is verifiability, not truth." See the essay, WP:Verifiability, not truth."

To:

"This principle used to be expressed on this policy page as "the threshold for inclusion is verifiability, not truth." See the essay, WP:Verifiability, not truth."

I even think it's simpler and more direct wording, as well as less implicitly patronizing. --Tryptofish (talk) 00:01, 19 November 2012 (UTC)

• Support... I agree that it says the same thing as the current language, only more concisely and with a less patronizing tone. (I would still prefer it as a parenthetical, but I could certainly live with it as a footnote). Blueboar (talk) 19:54, 19 November 2012 (UTC)
• Support - Seems like a sensible, incremental improvement. The word "notably" is a bit provocative. --Noleander (talk) 22:38, 19 November 2012 (UTC)
• Support - simpler, more direct, and doesn't look like the sentence is trying to justify its own existence, like the current one does. Diego (talk) 07:07, 20 November 2012 (UTC)

  It sounds like the reaction to this idea has been favorable so far. I'm going to wait another day or two, in case anyone has any concerns, and if not I'll go ahead and make the change. --Tryptofish (talk) 15:36, 26 November 2012 (UTC)

• Support - Good, simplifying edit, removes words that people have found jarring. Kalidasa 777 (talk) 11:20, 27 November 2012 (UTC)

Done I think it has consensus, so I have taken the liberty of making the change... revert if necessary. Blueboar (talk) 15:32, 27 November 2012 (UTC)

Thanks, Blueboar! --Tryptofish (talk) 15:34, 27 November 2012 (UTC)

I'm also fine with the subsequent change that KC made. --Tryptofish (talk) 00:15, 29 November 2012 (UTC)

Removal of updates

I've removed these updates [17] as they appear to be reacting to a current discussion on ANI [18] ; while eventually the community may support these changes they should only be made after discussion here. NE Ent 10:12, 21 November 2012 (UTC)

Reading that discussion I agree with Beyond My Ken that WP:BURDEN should not be interpreted in a strict sense that "once an unsourced sentence has been removed from a page, restoring it requires an inline citation or else it's violating WP:V". That would open the way to WikiTrolls removing large blocks of pages at random, and thus forbidding all the unsourced bits in the removed content from being restored.

But I don't think to restrict the wording to "editors in good standing" is the right way to solve the problem. A better solution is this: when there is no consensus to remove or keep the content, require the challenger to identify which sentences or blocks would require inline citations, one by one.

This way the burden policy can be respected, but only for content that has been properly challenged by a user in good faith. This provides symmetry between the processeses of removing content and adding required inline quotes for challenged content; and thus avoids the above asymmetrical scenario by which well-developed articles, with large chunks of content sourced but not with inline references, could be trimmed (requiring a lot of effort to get back with inline refs). Diego (talk) 14:59, 21 November 2012 (UTC)

The problem is that, sometimes, what requires an inline citation is an entire block of text. We have to allow editors to challenge entire paragraphs (or even entire sections) when such a challenge is called for.

As for: "...and thus forbidding all the unsourced bits in the removed content from being restored."... not at all. Nothing in WP:BURDEN forbids restoring all the "unsourced bits"... we just need to provide sources when we do so. If we can not provide sources, then the material should not have been added in the first place.

I have never understood all the angst about WikiTrolls removing blocks of text... While they are annoying pains in the ass, they (perhaps unintentionally) actually force the rest of us to improve Wikipedia... they force us to stop being lazy. They force us to go out and actually do the research we should have done long before, and to provide sources that support what we say in our articles. The end result is: better articles. Blueboar (talk) 15:54, 21 November 2012 (UTC)

What a mess! I've read the now-closed ANI thread, and any semblance of intelligent analysis of BURDEN got lost in the rest of the noise. As for "Once unsourced material is challenged, it cannot be restored without an inline citation by anyone", that's simply not what BURDEN says. That's utter nonsense. It can be removed, even, as Blueboar correctly explains, by trolls. But BURDEN explains what happens next, and I trust editors acting in good faith can read it. I agree with Diego, and with NE Ent, that the suggested changes, especially the part about editors in good standing, do not help. After all, the editor who was blocked self-described as being in good standing, so the new language obviously wouldn't have addressed the root problem. --Tryptofish (talk) 16:21, 21 November 2012 (UTC)

Having read the Wikipedia:Administrators' noticeboard/Incidents#Beyond My Ken and Shaz0t it seems to me that the consensus about text removal was against Beyond My Ken (BMK). I think BMK knows that and hence BMK's recent edits to this policy. However BURDEN was recently the subject of a very long detailed debate before the current compromise wording was agreed upon (see Wikipedia talk:Verifiability/Archive 59#WP:BURDEN), and I do not think that the changes introduced by BMK are an improvement on that wording -- PBS (talk) 01:19, 22 November 2012 (UTC)

As for the argument "That would open the way to WikiTrolls removing large blocks of pages at random, and thus forbidding all the unsourced bits in the removed content from being restored." That cuts both ways without WP:BURDEN someone wanting to push a plausible fact into an article (as was done here) could insist on it staying even if there is no source known to the page's editors to back up the statement. Rather as with the burden of evidence in a trial (Blackstone's famous formulation) it is better that some unsourced correct material is removed until a source is provided than Wikipedia contains lots of plausible but incorrect material.[19] The advantages of WP:BURDEN are: it is a clear simple rule that acts as a fail-safe and simultaneously simplifies edit wars. As TRPoD wrote in the ANI "And if it is indeed trivial to source it, then the established editors in good standing should have simply taken that trivial action themselves and we would not be here". -- PBS (talk) 01:19, 22 November 2012 (UTC)

@Blueboar: If a whole paragraph is challenged as a block, then BURDEN should allow restoring the block with a single inline reference. The point to understand is that is that the deleted material may be already sourced with the article-level references, but not just inline. In that case, removing the block shouldn't require repeating the same reference inline for each restored sentence. Diego (talk) 07:03, 22 November 2012 (UTC)

A single inline reference at the end of a paragraph may suffice if the "reliable source ... directly supports the material" in the paragraph. But in many cases that is not true, as text ought to be a synthesis of several texts. When I am retrofitting inline citations for text attributed to EB1911--in articles that are often more than five years old[20]--I do find additional material that is not covered in the original EB1911 article. When I do, if they do not carry citations, I usually mark those new sentences with {{citation needed}} and place the inline citation to the EB1911 article on both sides of that sentence--one immediately before, and the other at the end of the paragraph.[21] -- PBS (talk) 15:03, 22 November 2012 (UTC)

Couple of thoughts:

• Regardless of what problem this is or isn't, adding the "in good standing" is not a solution. Most problematic editors are "in good standing".
• In our discussions here we seem to repeat the fallacy that most of the contentions cases of application WP:Burden involve UNSOURCED material and CHALLENGED material. This is not accurate. The most common contentious cases cases are Wikilawyering POV and pissing warriors who:

1. Use it on material that they do not question, they just want to knock it out to pursue a pissing war or a POV quest.
2. Use wp:burden in synergy with the wp:rs fine print on SOURCED material. So, they can use some imperfection in the sourcing to domino through wp:burden to knock out sourced material.

• We have inched towards the good answer to #1, perhaps it's time to go another inch. That would be to say that the deleter should state some other concern about the material (e.g. verifiABILITY, veracity) when knocking it out. Just a process note, not something that they have to defend.
• The good answer to #2 would be to say that the more questioned and questionable the material is, the stronger the sourcing must be, and vica-versa.

Sincerely, North8000 (talk) 11:23, 22 November 2012 (UTC)

Amen, bro. (edit conflict) Diego (talk) 13:45, 22 November 2012 (UTC)

North, I don't see the point of a "process note". It won't do a thing to stop the POV warrior (they will simply lie in the edit summary, and accuse you of bad faith if you question it). And requiring a "process note" just makes life harder for other editors who are honestly trying to improve an article by removing unsourced material.

No, by far the easiest (and least stressful) way to deal with someone who has removed unsourced material per WP:BURDEN (no matter who does it, or what you suspect might be the actual reason behind the removal) is to simply return the material with a citation to reliable source. Blueboar (talk) 13:42, 22 November 2012 (UTC)

I agree. -- PBS (talk) 15:03, 22 November 2012 (UTC)

I believe that my idea would satisfy both goals. It's really no big deal. I do it all of the time when tagging/deleting, saying something like "Unsourced, and, as a sidebar, looks implausible". This makes it clear that the issue that needs addressing is the lack o sourcing, not a debate on the latter item. North8000 (talk) 15:32, 22 November 2012 (UTC)

Actually, BURDEN already says: "When tagging or removing material on these grounds, make it clear that you have a concern that the material itself is unverifiable." That sentence actually comes from an idea of North's, that I further revised, and I recently made sure that it didn't get diluted during the "incremental" edit process. Taken with the sentences just after it, it makes clear that BURDEN-based removals should be on the basis of non-verifiability, not on the basis of don't-like-it. I realize that it doesn't go quite as far as asking for a process note, but I think it's something that can be quoted back to the removing editor if someone thinks the removal was ill-advised. --Tryptofish (talk) 16:21, 22 November 2012 (UTC)

And I am not completely happy that we added that sentence. It simply gives wikilawyers and excuse to wikilawyer. Yes, leaving explanatory edit summaries can be helpful, but requiring them would cause more problems than it solves. (and I think most editors would ignore such a requirement).

Suppose a POV warrior wishes to remove unsouced material for POV reasons... he will do so, no matter how many hoops we require him to jump through. If we adopt NOrth's idea of requiring a "process note", the POV warrior will simply lie and leave a fake "process note": "Unsourced, and, as a sidebar, looks implausible". Nothing has changed... The POV warrior can argue that he has complied with policy, and the burden is still with those who wish to retain the material to "prove it" by citation.

On the other hand, if a non-POV editor omits (or simply forgets) to leave a "process note"... North's suggested requirement gives POV warriors an excuse to continue their POV warring and wikilawyering. They can revert a perfectly legitimate removal of unsourced material on the petty procedural grounds that "no process note was given". Blueboar (talk) 16:54, 22 November 2012 (UTC)

Also good points. Maybe what Tryptofish put in is a good middle of the road route on this. But, Blueboar, I submit that the folks just simply trying to keep unsourced material in already quickly lose, whereas the clever wililawyering warriors are are the ones who are harder to rein in and who are the persistent problems. North8000 (talk) 17:02, 22 November 2012 (UTC)

No question that wikilawyers are difficult to reign in... my point is that the recent additions and changes to BURDEN do more to encourage wikilawyering than to reign it in. Remember, there can be wikilawyering on both sides of an unsourced information dispute. The current language actually encourages those on the retention side of an unsourced information dispute to wikilawyer. It focuses them on procedure instead of substance... and to whine that a removing editor "didn't make it clear (or clear enough to satisfy me) that you have a concern that the material itself is unverifiable".

The older versions of WP:BURDEN were simple and clear cut... easy for everyone to understand. It could be summarized in three sentences:

1. You are allowed to add unsourced information to Wikipedia... but WARNING: Someone else is allowed to remove it (for any reason).
2. If you come across unsourced information that someone else wrote, you may remove it if you wish to do so (for any reason).
3. If someone removes unsourced information that you think belongs in the article, you may return it... However, you must provide a citation to a reliable source when you do so.

There was no ground for a wikilawyer (on either side) to stand on with that language. No "procedural quibbles" could be made. The only questions that could arise (and be argued about) were a) whether the information actually had been sourced prior to removal (say later in a paragraph)... or b) whether the provided source was or was not reliable.

Yes, WP:BURDEN was harsh on those who added unsourced information... but that was intentional. We intentionally favored removal, to encourage editors to do sourced based research. To write articles based on "what sources say" rather than on "what you (an editor) know". Blueboar (talk) 18:04, 22 November 2012 (UTC)

This discussion thread was originally about the "good standing" edits, and I think we've established that they were not really an improvement. We can probably go around in circles about what best thwarts determined wikilawyers (hey, how about deleting all policy – then they'll have nothing to point to!), but we can all agree that someone determined enough can willfully misconstrue anything. I don't know if it's unanimous that the new BURDEN is better than the old, but I think it's consensus. Time to move on. --Tryptofish (talk) 20:43, 22 November 2012 (UTC)

Not sure that the new language actually is consensus... (it certainly does not reflect the actual editing practice of most editors)... but I am not up to a big debate about it right now... so I will let it lie. Blueboar (talk) 21:13, 22 November 2012 (UTC)

Maybe I said that a little too strongly, sorry. But anyway, we did have a thorough RfC on it, and Mr. Strad closed it. --Tryptofish (talk) 21:19, 22 November 2012 (UTC)

Yeah, we did... and I actually ended up supporting the change. I now regret that decision. Blueboar (talk) 21:25, 22 November 2012 (UTC)

Oh, well, please don't feel badly about it. --Tryptofish (talk) 21:37, 22 November 2012 (UTC)

I took part in the discussion and I did not support the addition of the footnote for the reasons Blueboar has explained more clearly here than I did in the RFC. I was not longer willing to argue the point further at the RFC simply because I had stated my POV and at least some of those points had been accepted. -- PBS (talk) 16:43, 27 November 2012 (UTC)

All of the good points raised show that almost anything done would be a two-edged sword. Sounds too big and complicated to tackle now. North8000 (talk) 00:46, 23 November 2012 (UTC)

But they will need to be tackled eventually. Blueboar (talk) 15:39, 26 November 2012 (UTC)

I have another idea to solve it. A statement to the effect that "material that is challenged for reasons in addition to lack of sourcing requires stronger sourcing." What that sort of attempts to establish is to leave it easy to knock out material that is unsourced, but if the material IS sourced, but you've going to knock it out anyway because of imperfection in the sourcing, to do that you must challenge the material. Well, that's my first try at it anyway. North8000 (talk) 00:49, 29 November 2012 (UTC)

That confuses the issue, I think... there are lots of reasons why someone might challenge material text/information/sources... not all of them relate to BURDEN. BURDEN only relates to challenges of a specific type ... it only deals with a challenge that information is unsourced (and therefor potentially unverifiable). in other words... a challenge to sourced information is not a WP:BURDEN issue. Blueboar (talk) 14:28, 29 November 2012 (UTC)

Actually they are linked in multiple ways. One is structural/logical, wp:burden places these conditions on the material remaining (emphasis mine)

1. The citation must clearly support the material as presented in the article
2. Any material lacking a reliable source directly supporting it may be removed. (and "relaible source" is taken to mean meeting the rules of wp:rs.)

A majority of cites/sourcings have some imperfection with respect to one or the other under an aggressive literal application of these standards.

A common way to POV an article (and the most common mis-use of wp:ver/WP:burden) is to knock out sourced, unchallenged material based on an aggressive literal interpretation of those two requirements. So my idea raises the bar a tiny bit (only) when knocking out sourced material. Sincerely, North8000 (talk) 14:52, 29 November 2012 (UTC)

OK... I think you are talking about a two step process... step 1) someone challenges a source because it is unreliable (and removes the source on this grounds)... step 2) since the material that was supported by that unreliable source is no longer sourced... the person challenges (and removes) the material per WP:BURDEN. Is this what you are talking about?

If so... I still don't see where the problem is... those who wish to return the removed material can do so... they simply need to provide a reliable source when they do so. While one could always debate whether the original source was (in fact) reliable, it is usually far quicker, easier and less stressful to not argue the point, and simply find another source... one that is reliable (and if you can not do that... it is highly likely that the information shouldn't have been in the article to begin with). Blueboar (talk) 15:24, 29 November 2012 (UTC)

Useful exercise?

New angle, but from previous thread) Blueboar, you and I have had a gentle long running discussion more-or-less on this. I believe that the underlying difference is which angle we come at this from. You come at policy writing from the angle that it should be 95% based on "how is this policy intended to be used?" and the intended situations for that. I come at it from going maybe 60% by "how is this intended to be used?" and 40% from the "how is this being mis-used? and "let's try to tweak it reduce that mis-use". I think I understand "how is this intended to be used?" side but I don't think that you fully understand the "how is this being mis-used?" side. I'd like to propose an experiment. Let's copy Freemasonary (A subject that I know you are active in) into a work space. You play the side of an editor who wants build/keep it as a quality informative article, who is willing to spend some time on it, but not make a life out of it. I'll play the dark side, someone who is mis-using policies to conduct a pov or pissing war, but also isn't willing to make a life out of it. So neither side is allowed to spend a large amount of time on this. I think that it would an exercise that would be very informative to all to see how core policies and guidelines interact with this process. Sincerely, North8000 (talk) 12:17, 30 November 2012 (UTC)

Couldn't you simply give a mock example of misuse of the policy? KillerChihuahua^?!? 13:36, 30 November 2012 (UTC)

Rather than play roll playing games ... I think part of the difference in our views stems from a more fundamental disagreement... I think we actually disagree on what constitutes "mis-use". It isn't that I don't understand the "how is this being mis-used" side... it's that I don't think some of the things you are concerned about are mis-use. You view certain editorial behaviors as being mis-use (and so would like to reduce them). I, on the other hand think these behaviors are perfectly acceptable (and so see no reason to reduce them). Blueboar (talk) 13:43, 30 November 2012 (UTC)

Answering both, in the theater of our minds, I'm in a POV/pissing war, and just deleted the entire first sentence of the lead of Freemasonary, and my edit summary said "not wp:rs, primary source, does not support the material in the sentence". (and note that took me about 50 seconds to do, including glancing at the source) What is your next step / response? And as a sidebar, is what I "did" use or mis-use? North8000 (talk) 14:29, 30 November 2012 (UTC)

Well, my first response would be to ask you to clarify your edit summary... Since there was more than one source, supporting different statements in the sentence, I would have to ask: Which source is not rs? which source is primary? and which source does not support the material in the sentence?" (If you replied I would then engage in discussion and try to explain why the sources are acceptable as used) As for whether what you did was a mis-use... no, I would not call your removal a "misuse" of the policy... I would assume you honestly did not understand the policy, and would attempt to explain it to you. Blueboar (talk) 00:16, 1 December 2012 (UTC)

Oops, I meant to say I just deleted the first 80% of that sentence, up to the first footnote....so the first question of which source/which sentence is now moot.

(in character) (I know the policy, I'm an expert wikilawyer, that's why I know how to use the policy it to conduct the pov/pissing war. I also know that the material is accurate, but I want to knock it out per my pov/pissing war) It doesn't matter. It was a primary source. If you want to restore the material, per wp:burden you must provide a reliable secondary source which directly establishes the numerous assertions made in the deleted phrase and show how it does so. WikilawyerWarrior ( North8000 (talk) 00:34, 1 December 2012 (UTC) )

(I've been following these discussions.) First, let me just make sure that, North, you are actually talking about the second sentence of the lead, right? So your character is challenging the large part of that long sentence that leads up to inline cite number 1. My reaction is that your character is being annoying, but not disruptive. I think that there's a reasonable argument that the clauses your character contests ought to be cited specifically. There's also a reasonable argument that WP:LEAD doesn't require all that, so long as the same information is sourced in the main text. On the other hand, I think that your character is on shaky ground for referring to BURDEN (a part of WP:V, of course) for an issue (primary versus secondary sources) that is, instead, part of WP:RS (and maybe a little of WP:NOR). The wording that we recently added to BURDEN (at your suggestion) gives editors a basis to point out to your character that BURDEN only requires that one can verify that the cited primary source says what the disputed sentence says, and BURDEN does not speak to whether the primary source is a reliable one. (OMG! we're looking at a textbook example of VnT!) I think those are things that can be discussed on the article talk page, and your character can be taken to DR if they engage in WP:IDIDNTHEARTHAT. But it's not unreasonable to ask whether everything in that sentence is sourced by inline cite number 1. --Tryptofish (talk) 00:57, 1 December 2012 (UTC)

Regarding which sentence, you are right. I got that wrong twice; I guess I'm going to have to go back to bad guy school and relearn the basics like looking for punctuation.  :-)  :-). More to come North8000 (talk) 01:56, 1 December 2012 (UTC)

FYI... If you are going to play the part of an evil wikilawyer, it helps to actually quote the policy/guideline correctly when you cite it... for example, in the above, your character complains that WP:BURDEN requires reliable secondary sources... but WP:BURDEN does not actually require that... it does not mention the issue of primary/secondary sources at all (nor does WP:V in general... except for a short cautionary note about original research). Blueboar (talk) 02:12, 1 December 2012 (UTC)

(edit conflict, responding only to Tryptofish) (In character) Regarding being the lead, per policy, there is no exemption for the lead. But, setting that aside for the moment, regarding being on shaky ground in citing wp:burden, I believe not. My character would say "it needs to be sourced by a reliable source per the Wikipedia meaning of "reliable source" This means secondary source unless you can show where this meets the vary narrow and tight restrictions on the use of primary source. WikilawyerWarrior. North8000 (talk) 02:18, 1 December 2012 (UTC)

Reinforcing response to Blueboar. Wp:burden says "reliable source", and then wp:ver has a "reliable sources" section. The first line says to see wp:rs for info, which roughly says that the source should be secondary, plus meet all of the other wp:rs requirements.North8000 (talk) 02:26, 1 December 2012 (UTC)

Man, I'm feeling dirty / creeped out playing this role. But I think that it is useful. North8000 (talk) 02:26, 1 December 2012 (UTC)

Maybe worth it. Blueboar made a nice tweak at wp:rs. Blueboar, this reinforces your spot on the the top of my "people who I most highly admire that I usually disagree with" list. North8000 (talk) 21:50, 1 December 2012 (UTC)

I can see how a difficult editor could look here, and wikilawyer about BURDEN requiring a reliable source, and then deliberately misconstrue the definition of "reliable source". But I'm not convinced that we can solve that here. I think it's a matter of how WP:RS is written (and I'm not adding that to my watchlist :-) ). RS shouldn't say that only secondary sources can ever be reliable. In terms of what WP:V can say, I suppose we could make clear that BURDEN requires only a reliable source, not the best source that could possibly exist, but it strikes me as a bit WP:CREEP to go that far. --Tryptofish (talk) 23:05, 1 December 2012 (UTC)

NPOV & verifiability

I'd like to call attention to this little section of the article that deals with the relationship between verifiability and NPOV:

Sources themselves do not need to maintain a neutral point of view. Indeed, many reliable sources are not neutral. Our job as editors is simply to summarize what the reliable sources say.

If a reliable source is indeed not neutral, does that mean that we should present it in a non-neutral POV? Is "our job as editors" to still report it with a biased POV? Or, if this is the case, should we report it with the POV, and then be sure to use quotations or the name of the author? Charles35 (talk) 16:36, 3 December 2012 (UTC)

( The above excerpt is from Wikipedia:V#Neutrality. --Bob K31416 (talk) 22:33, 3 December 2012 (UTC) )

Balance requires reporting the different substantial points of view. Honesty requires letting the reader know that we've done that. Neither requires the wp:OR step of characterizing them as biased. We need only make it clear that X says Sx and Y says Sy. We don't compare their points of view, the intelligent reader can do that without our help. If neutral Z has published comments on the differences (Sx vs. Sy), we can further report Sz. LeadSongDog come howl! 16:57, 3 December 2012 (UTC)

I don't think I fully understand. If it is clearly a biased view, you should present it as if it weren't? You shouldn't quote it or attribute it? And you said the intelligent reader - what about the fact that some, if not most, readers are not all that intelligent. Shouldn't we keep in mind wikipedia's audience and write for them? Lastly, I'm really interested to hear what you mean with "honesty". Would you be able to elaborate on that? Thanks Charles35 (talk) 20:02, 3 December 2012 (UTC)

No, if it is clearly biased, first we ask if it is worth using at all. If despite being biased, it represents a widely held viewpoint that needs to be represented in order for the articles coverage to be balanced then we make it clear whose viewpoint it is by in-text attribution, but we do not ghettoize the authors by describing them as "advocates", "opponents", etc. We provide balance by similarly covering the opposing viewpoints. However, if an objective secondary source discusses both of the biased viewpoints, then we are far better to look to that objective secondary source than to draw our own conclusions on which aspects of the slanted coverage deserve inclusion. See wp:BALANCE. It is not our job to form opinions for our readers. We simply report (in an encyclopedically structured fashion) the best of what others have already published on a topic. LeadSongDog come howl! 21:49, 3 December 2012 (UTC)

The simple principle is that the Wikipedia article should be neutral, regardless of its sources. --Bob K31416 (talk) 23:21, 3 December 2012 (UTC)

• Here's an example for the OP. Donald Trump has repeatedly stated that he believes that Barack Obama's birth location is in doubt (a stance popularly known as the birthers). Now, Trump has written about this. His own writing is a reliable source for what he has to say: We can trust him to state his own beliefs, so his own writing on the matter is reliable for us to use in the Donald Trump article to note his position. However, is own writings on Barack Obama's birth status are NOT themselves neutral. That doesn't mean we can use them nowhere on Wikipedia. The fact that Trump is a "birther" is reliably sourced to his own writings on the fact, and that's OK to cite is writings for that purpose. Does that make sense? --Jayron32 23:31, 3 December 2012 (UTC)

Umm, sort of. I don't think it's that I don't get your explanation. I think the rules aren't as complete as we like to think and that there is a lot that is ambiguous. Or maybe it's just impossible to make a perfect set of rules for general cases. This is why I made this section. I thought that maybe it might be a good idea to add something between Indeed, many reliable sources are not neutral and our job as editors is simply to summarize what the reliable sources say. I thought it might be helpful to say how we should summarize sources that aren't neutral. ie should we attribute? How should we attribute? In-line citation or saying the actual person's name in the text? Should we quote? Most importantly: in the case where we are not in-material attributing (we can be parenthetically citing, however) or quoting, and are presenting the data as fact, or the opinion as a factual one (ie not saying anything like "it has been said" or "many believe") should we modify the material to make it more reasonable or less biased without removing all bias in an effort to make it more appropriate for wikipedia? For example:

Ensuring of the proper feeling rules of the breast cancer culture is encouraged, including remaining optimistic of a full cure, rationalizing the selfishness of treatment as a temporary measure and feeling guilty that it forces her to put her needs momentarily above the needs of others or due to her perceived inadequacy in caring her family or other women with cancer (Sulik 2010, pages 225–272, 277).

This material is presented as fact. It is cited but not attributed in the material (ie it doesn't say "Sulik says that..."). It is reflective of the source, but in my opinion, it is way to extreme. Is it really worth it to include all of this bias? Or should this be summarized into a lesser biased form that effectively conveys the idea in a less harsh and inappropriate manner? Charles35 (talk) 05:05, 4 December 2012 (UTC)

The decision of whether to quote in full or to summarize is really a matter of WP:Undue weight... not one of WP:Verifiability. Blueboar (talk) 13:34, 4 December 2012 (UTC)

In a case like this it's also a matter of WP:V, that the references must justify the conclusions. It is improper use of a source to use it as a general statement unless it's a recognized tertiary authority. Sulik's book is intended to argue a point,not present a neutral summary. DGG ( talk ) 00:36, 13 December 2012 (UTC)

Sulik's viewpoint seems to be the dominant one among published reliable sources. It just doesn't say what the general public (e.g., Charles) thinks it should say.

Or, to put it another way, if you want to say that a given source is non-neutral, you have to demonstrate that some other published, reliable source holds a different opinion, not merely that you personally disagree with the source (e.g., because you believe that society shouldn't make women with cancer feel guilty, even though the fact that they actually do appears in multiple sources from websites like cancer.gov through popular books to peer-reviewed journal articles). WhatamIdoing (talk) 01:39, 19 December 2012 (UTC)

Whatamidoing, please do not watch my recent contributions. Thank you. And as usual you know nothing about my personal opinion and I would prefer if you kept your assumptions to yourself. My personal opinion is not congruent with the opinion of the general public (whatever that means). It is also irrelevant here. We might be talking about our "personal opinions" on wikipedia policies and the correct way to construct an encyclopedia, but please do not divert the conversation to a false assumption about my personal opinion on the subject matter. It would be much obliged.

Also, I'm not sure if you knew, but This page is for discussions about the Verifiability policy. For questions about the reliability of specific sources, see Wikipedia:Reliable sources/Noticeboard. We aren't having a discussion about breast cancer awareness. I am trying to make the rule as explicit as I can, and I used an example to do so. I have no interest in bringing other things like cancer.gov into this. Charles35 (talk) 21:13, 20 December 2012 (UTC)

My comment is not directed at you. I'm not watching your contributions. If you will click here, you will discover that I am the sixth most active contributor to this talk page in the entire history of the English Wikipedia. It is hardly surprising that I have noticed your question on a page that I've been actively participating in for more than four years now. WhatamIdoing (talk) 07:31, 23 December 2012 (UTC)

Discussion Continued

I wanted to ask one more specific questions and see if it might spark discussion for adding something to the text to make it more precise. Here are two excerpts from the above discussion that I'd like to talk about:

The simple principle is that the Wikipedia article should be neutral, regardless of its sources.

This brought out my question:

[are there ever any cases in which it is appropriate to] modify the material/ideas (ie lessen it) to make them more reasonable or less biased without removing all bias in an effort to make it more appropriate for wikipedia? Charles35 (talk) 16:50, 27 December 2012 (UTC)

I haven't read the previous discussion, but maybe the answer to your present question is "yes". The following hypothetical example doesn't exactly address your question since it removes all apparent bias, but maybe it is appropriate for your question anyhow. Suppose a reliable source says, "The sleazy group X has its headquarters in Y." Seems like it would be acceptable to add to a Wikipedia article the info, "The headquarters of group X is located in Y." --Bob K31416 (talk) 17:34, 27 December 2012 (UTC)

Okay, thank you for your reply, that was very informative. What about this? If the source said "The sleazy, horrible, awful, disgraceful, atrocious, etc group X has its headquarters in Y." It describes group X as, hypothetically, an overall extremely negative thing. But it is a reliable source. Can we translate this to "The headquarters of the questionable group X is located in Y"? Or, the "mediocre" or "poor" or some other lesser adjective in an attempt to soften it to make it more appropriate for wikipedia?

Basically, in these examples, there are two things going on - (1) the fact that Hx is in Y, and (2) the fact that the author considers X sleazy. In your proposal, you removed (2) the authors bias altogether and only utilized the (1) the factual aspect. What I am asking is whether or not we can keep the author's bias but make it less "extreme".

I hope that makes sense. If it doesn't, I can clarify. Charles35 (talk) 17:58, 27 December 2012 (UTC)

Or, more simply, if it were to just say "Group X is sleazy, horrible, awful, disgraceful, and atrocious", can you soften that to "poor" or something like that, effectively retaining the idea but removing some of the harsh bias? Or can you not use something like this at all? Charles35 (talk) 18:30, 27 December 2012 (UTC)

In a section about Y, you could use the fact that X is Y without qualifications (provided the source is truly reliable). In a section about "reception" or "criticism" of X, you should both include that the source thinks X is bad, balanced with diffent views by other groups (including X themselves). Softening the words used to criticise X is an editorial decision - you could soften the expression if including the whole description is against WP:DUE WEIGHT, but you should include them if the exact terms are relevant, because Wikipedia is not censored nor biased. Diego (talk) 18:39, 27 December 2012 (UTC)

It would depend on the context and the rest of the article. There may not be a general formula to use for determining what to do and it may come down to a matter of judgement whether the Wikipedia article appears to be taking sides if a particular edit were made. Ultimately, the consensus of the judgements of various editors might be needed to determine whether it would violate WP:NPOV. If you want to put in the value judgement of the source, it might be helpful to preface the info with, "According to ...". Also, a consensus might be needed to determine whether the source's opinion is appropriately characterized. --Bob K31416 (talk) 18:44, 27 December 2012 (UTC)

Burden of proof

One occasionaly comes across editors who want to delete all unreferenced material regardless of whether they believe it to be unverifiable or not. We already have in policy the statement "When tagging or removing material on these grounds, make it clear that you have a concern that the material itself is unverifiable. If instead you think the material is verifiable, it is better to try to provide an inline citation yourself before considering whether to remove or tag it." Going round tagging and deleting in a mechanical bureaucratic fashion is merely makework for editors, or else results in loss of valuable material when no one bothers to respond. I would like to add something stronger to the policy such as,

Removing material solely for lacking citations where you have no real concern that it is unverifiable can be considered disruptive

to make it clear that tagging and deleting should only be happening when there is a genuine belief that the material might be problematic. There does not need to be a time limit to provide sources for material that is entirely uncontroversial and factual. SpinningSpark 22:02, 17 December 2012 (UTC)

I would support doing that. (By the way, there was just an edit to that section that was reverted, and I agree with the revert.) --Tryptofish (talk) 22:19, 17 December 2012 (UTC)

I support. BTW, I did the revert. I agreed with the spirit of what they are trying to do, but as worded it could be utilized as creating an exception to wp:ver for new articles, IMO that is going too far. Sincerely, North8000 (talk) 22:23, 17 December 2012 (UTC)

• I also support an addition along these lines. NTox · talk 01:36, 18 December 2012 (UTC)
• Oppose The line is horrible and nothing but instruction creep. "No real concern" can never be proven and would creat FAR MORE drama than what it is proposed to avoid. There have been some recent situations were everyone was already too afraid to touch a BLP that was nothing but unsourced claims. So horrid was this article that Jimbo had to step in and reduce it to a stub. This is not the route to go. It creates massive amounts of issues when entire articles are missing reliable sources. Right now people still believe that a primary source is sufficient to add claims they synthesize from reading between the lines or assuming information and then become even more disruptive the those attempting to remove it. No, we do not need to add this line , or any other to protect unsourced claims on Wikipedia. Source it or lose it.--Amadscientist (talk) 01:45, 18 December 2012 (UTC)
• Support the intent, but that wording needs to be fixed; too easy to be gamed by the wikilaywers. --MASEM (t) 01:56, 18 December 2012 (UTC)
• Oppose Changing policy to require us to guess what is in the mind of the remover is not going to end well. The current practice also serves neatly to end arguments over whether a claim is sourceable or not, if its not sourced, it can be removed, and the person arguing that it is sourceable can then go source it and restore it. Monty₈₄₅ 02:05, 18 December 2012 (UTC)

We're not requesting to guess what is in mind of the remover. We're requesting that the remover follows the current guidelines and states it explicitly. And "if its not sourced, it can be removed" is not the current policy; there are important nuances that deleters too often ignore but that are important. Diego (talk) 08:05, 18 December 2012 (UTC)

• Oppose - In most cases, I actually find the complaints about the removal of unsourced information to be more disruptive than the removal itself. By far, the easiest and least disruptive solution to editors who remove unsourced text is to quietly return the material with a citation. Nine times out of ten, doing so will take no more than five minutes of searching the internet (far less time and energy than it takes to complain about the removal in the first place). Blueboar (talk) 02:28, 18 December 2012 (UTC)

In reply to some of the above: I don't think the intent of the line is to defend unsourced information. Rather, it is to remind us that a "no source?→delete" style of editing is not appropriate. There are a lot of people who do this kind of automaton whitewash; others I think even like to "stick it" to people by removing their content for not being sourced, even though it was verifiable the whole time. The end goal is always verifiability and it's not sources. Sources merely prove verifiability, but again what we care about is verifiability, proof or not. If someone has no doubt that some piece of content is verifiable they should not be removing it simply because it has no source. And for any other person who actually does doubt the content's verifiability, well then it's business as usual and the person can remove it for not having the source. The proposed line would not make removing content any harder. It's only there to remind us why we're removing stuff in the first place. NTox · talk 02:34, 18 December 2012 (UTC)

In reply to those that don't think this is about the removal of unsourced content....then we are fine without the change. Not broken...don't try to fix it.--Amadscientist (talk) 02:38, 18 December 2012 (UTC)

The guideline is fine, the problem is that people with your position are not following it as it's written. Diego (talk) 10:10, 18 December 2012 (UTC)

Whack! You've been whacked with a wet trout. Don't take this too seriously. Someone just wants to let you know that you did something silly.

You have been trouted for: Thats a huge and innacurate assumption.--Amadscientist (talk) 11:37, 18 December 2012 (UTC)

Trout assimilated :-P I've striked the wording that I thought assertive but sounds offensive. I do have a concern that you seem to miss an important part of WP:BURDEN and WP:V, the bit about due process and using the power of Wiki. I don't have time now to describe its implications, I'll try to elaborate on it tonight. Diego (talk) 11:48, 18 December 2012 (UTC)

I am reading everything you write so I am sure I will see it when you post, however I should at least mention that I actually understand your concern, but feel you are placing a requirement where none exist. That is my actual concern with the replies you are making about "due process". There is a process and I feel we should let that play out with all situations regarding the removal of content. V does allow (by policy) the removal of unsourced content. Does that mean it should be? Maybe not in some cases, but the majority of content removal is being done in good faith. When it is obvious that it is not, is when we need the process even more. Part of all of this is that editing is part of consensus forming. Placing too many limits or to many suggestions of what is "proper" (but still not required) only confuses editors and makes working on articles more complicated.--Amadscientist (talk) 12:11, 18 December 2012 (UTC)

I don't want to place too many limits, I want the policies and guidelines to explain why placing tags and moving content to talk page is preferred over deleting it and having it lost in the history page. The main concern about what I want to elaborate is that deleted content is "out of sight, out of mind" and this harms the article building process, because editors coming later (maybe years) will not be able to find it and source it. Acting upon this preference to preserve and help is already part of the rules, not an additional requirement. Diego (talk) 12:19, 18 December 2012 (UTC)

• Oppose. As Amadscientist wrote above, the bottom line is "source it or lose it", and it should remain that way. The proposed change would effectively invert the burden, by placing the onus on the remover to demonstrate that the material was unverifiable. Proving a negative is very difficult, so this would generate a lot of drama. --BrownHairedGirl (talk) • (contribs) 07:48, 18 December 2012 (UTC)
• Support although Masem has a good point about wording. It is not "source it or lose it" but "if it is contested, you need to source it or lose it", and thus, the lack of sourcing can never be an essential part of whether a statement is contested or not: BURDEN has always required a good-faith belief that something is wrong before removal. Jclemens (talk) 07:52, 18 December 2012 (UTC)

No, not really. Burden is clear: "The burden of evidence lies with the editor who adds or restores material, and is satisfied by providing a reliable source that directly supports the material." It then goes on to say: "Attribute all quotations and any material challenged or likely to be challenged to a reliable, published source using an inline citation." These two statements do not mean the same thing. The second line is refering to HOW to meet burden. Only material that is likely to be challenged should be cited. What that means is: unquestionable facts do not need a reference. The very first line makes it clear that returning unsourced material that has been removed, must meet burden as well. So if it is removed and you put it back....you MUST provide a reference. Your claim that: "the lack of sourcing can never be an essential part of whether a statement is contested or not" is beside the point. Removing content is not contesting it. It is just removing it as unsourced and "questionable". It is not required (nor should it be) that an editor formally "contest" the information. That is wikilawyering. Our current policy is to remove uncourced information.......not create a discussion to contest the validity of said information. Would that be nice in a perfect world? Perhaps, but it also limits the removal of unsourced content, which is exactly what this proposal is about. Oh....and Masem didn't say "Source it or lose it". I did.--Amadscientist (talk) 08:36, 18 December 2012 (UTC)

(inserting a reply) - Re: "Only material that is likely to be challenged should be cited." No... WP:BURDEN applies to more than material that is likely to be challenged; it also applies to material that has been challenged. Removal for lack of citation is a form of challenge. Blueboar (talk) 13:19, 18 December 2012 (UTC)

The very first line makes it clear that returning unsourced material that has been removed, must meet burden as well. I don't think anyone is contesting that part; I certainly don't. The problem is with how material is being removed, not returned.

Our current policy is to remove unsourced information See, that unqualified assertion is false. That some people holds that misunderstanding is the reason why we're trying to clarify for you what V really means and how it must be used. There are important procedural reasons why the intermediate steps (trying to source disputed content yourself, tagging it if you can't, WP:PRESERVEing at talk pages) must be followed, even if the final result (removing unverifiable information) is the same. Those steps help us to build an encyclopedia step by step through the Wiki process, and short-cutting them is harmful. I don't want to stimulate the battleground mentality ; that's why I oppose the original wording that tried to tag you as being disruptive. Though I think it's important that those opposing this change recognize the need to protect the process by which unverified information is assessed as either to be removed or to be sourced; that process is an important part of the current WP:V, WP:BURDEN and WP:PRESERVE policies. I'll try to expand on these thoughts latter to explain you the full benefits of the process. Diego (talk) 10:10, 18 December 2012 (UTC)

"Any material lacking a reliable source directly supporting it may be removed". That is "qualified". It is what the policy states.--Amadscientist (talk) 11:33, 18 December 2012 (UTC)

Yes, and right after that it describes the proper process to remove it. It doesn't include "instantly, without reason". The only unverified material that can be instantly removed without a reasonable effort to assess its verifiability is BLP; the rest of policies do have requirements to make an effort not to simply lose it; so "source it or lose it" is actually against policy for content removed without care. Diego (talk) 12:14, 18 December 2012 (UTC)

Some comments to the opposers:

• This proposal does not affect the policy regarding BLPs; that remains separate. The following paragraph, opening with "However, do not leave unsourced or poorly sourced material in an article if it might damage the reputation of living people" makes that quite clear. If anything, my proposal is emphasising that by making the difference with BLPs seem even more dramatic.
• It is not the intention here to give an opening to wikilawyering or require us to "guess what is in the mind" of editors. Rather, it is aimed at good faith editors who openly declare this is what they are doing. Suggestions for better wording would be welcome.

SpinningSpark 07:54, 18 December 2012 (UTC)

While it is nice that you explained this here it makes no difference what your intent is....the proposed line DOES actually say "..where you have no real concern". That is indeed an assumption one cannot make of another editor. It assumes bad faith.--Amadscientist (talk) 08:41, 18 December 2012 (UTC)

We can make that assumption when the editor declares that's what they are doing. POVers and trolls will be disruptive no matter what, but there is a class of misguided editors who believe they should delete everything no matter what. That is the group I am trying to address with this. SpinningSpark 19:28, 18 December 2012 (UTC)

• Support BURDEN applies to material challenged or likely challenged, not to all material; removers of content need to make clear that they're removing it on grounds of verifiability; and that they've reasonably certain that sources can't be found - otherwise the correct way according to policy is to tag it as needing citation, not to delete it. It's only and additional clarification of the actual existing requirement to identify the reason why the content is being challenged. Requesting more communication can only be a good thing for consensus-building.

I would include in the text the reason stated by SpinningSpark, that challenged but verifiable content is in risk of being lost, and not the explicit claim that removing content is disruptive if that you don't really think it is unverifiable. Diego (talk) 08:05, 18 December 2012 (UTC)

This misses the point of challenged or likely challenged as simply being unquestionable fact. "The sun rises in the east and sets in the west". You said: "removers of content need to make clear that they're removing it on grounds of verifiability; and that they've reasonably certain that sources can't be found". No, they don't. All editor really NEED to do is not remove unquestionable facts that any reasonable person understands as a fact and make sure they have added an edit summary (we cannot force what that edit summary says, but should be a reasonable explanation). What you are suggesting is that policy requires all removal of content to be thoughly researched. It simply does not say that. That is the requirement of those that add information.

Jimbo Wales states: I really want to encourage a much stronger culture which says: it is better to have no information, than to have information like this, with no sources. Any editor who removes such things, and refuses to allow it back without an actual and appropriate source, should be the recipient of a barnstar. – Jimbo Wales, 19 July 2006

--Amadscientist (talk) 08:58, 18 December 2012 (UTC)

What you are suggesting is that policy requires all removal of content to be thoughly researched. Not really; if you don't want to research the fact, you're expected to tag it as needing verification better than removing it. Only if you have made a reasonable attempt to assess its verifiability should you remove it; I think the very minimum to comply with the it is better to try to provide an inline citation yourself before considering whether to remove or tag it part of BURDEN would be a Google search. Diego (talk) 10:16, 18 December 2012 (UTC)

Uhm...no. Some editor expectations are simply inncorrect. There simply is no requirement to do that. The policy is: "Editors might object if you remove material without giving them time to provide references; consider adding a citation needed tag as an interim step" This is simply not an expectation per policy, but per certain editors. No offense to those that expect this...but if you claim one MUST tag before removal as the method of allowing time for the source to be added.......that is not accurate. Also, there is no "comply" with it is better to.--Amadscientist (talk) 10:31, 18 December 2012 (UTC)

Well, there is a requirement to preserve verifiable content even if unsourced, and to help build WP:CONSENSUS with your fellow editors; and deleting unsourced content without proper consideration goes against both. I'm not requiring to tag all content before removing it; only the one that you're not willing to research yourself. "Do not remove good information solely because it is poorly presented" is a mandate per WP:PRESERVE; if the fact would belong in a "finished" article, you're breaching policy by simply removing it. Diego (talk) 10:37, 18 December 2012 (UTC)

Is that a requirement? First of all the section is titled "Try to fix problems, which is a suggestion and does not assume that every editor is capable of doing so (you must rememeber that policy is made to accommodate all levels of experiance and knowledge, not just experts) Verifiability "means that people reading and editing the encyclopedia can check that information comes from a reliable source". If the content cannot be varified with a reliable source the information need not be preserved. WP:PRESERVE is absolutely not a requirement not to remove unsourced content...at all. "[T]hey should (not "must") be retained if they meet the requirements of the three core content policies (Neutral point of view, Verifiability and No original research)". What many here are trying to say is that "verifiability" is not an actual RS added as an inline citation to content, but only that there be a reliable source out there somewhere. This is true (even Jimbo has weighed in on this as accurate). HOWEVER the problem is...nowhere is there an actual requirement to keep unsourced information just because there is a RS out there somewhere. This is simply a fight over who has the most responsibility to add that RS. WP:PRESERVE "Problems that may justify removal" state clearly that "WP:Verifiability discusses handling unsourced and contentious material." Yes it does, it says "Any material lacking a reliable source directly supporting it may be removed. Whether and how quickly this should happen depends on the material and the overall state of the article." This current proposal CREATES a problem by telling editors that removing unsourced information could be disruptive. Sure...but we do not need a policy to state that. All kinds of things cause disruption. Just editing a page can cause disruption...but that doesn't mean it was the edit that caused it.--Amadscientist (talk) 11:27, 18 December 2012 (UTC)

• Oppose This will lead to far more drama and editwars than the present pratice. The Banner _talk 13:01, 18 December 2012 (UTC)
• oppose - if someone is going around creating disruption by removing everything that is unsourced there is already WP:POINT to deal with that problem. Anything where there would be requirements that someone prove their intent is unworkable. "I can restore this without a source because I think you removed it even though you dont think that there are no sources to support it." Well that is just buckets of assumptions against policy.-- TRPoD aka The Red Pen of Doom 19:46, 18 December 2012 (UTC)
• Support the general idea. There might be a better way of phrasing it. Yes, it's POINTy if you do a lot of it, but it's disruptive even if you only do a little bit of it. If you personally believe that the material is accurate, verifiable, neutral, and encyclopedic, then you personally have no business removing it. WhatamIdoing (talk) 01:45, 19 December 2012 (UTC)

if the removal of unsourced content is disruptive its already covered by our policies against disruptive editing. all this wording would do is ensure that there would be disputes about something that CANNOT BE proved - an editor's belief.How would I possibly ever prove that you were removing content that you suspected might be verifiable? How would you prove that you didnt really think there was a reliable source that supported the claim? Thats just enshrining teh nonproductive dramahz with no value to the encyclopedia on any front. -- TRPoD aka The Red Pen of Doom 02:52, 19 December 2012 (UTC)

I've seen people actually say that they believe the material is correct but they are removing it because they believe this policy requires (or at least authorizes) them to do so. They usually go on to explain that they hope the removal will be an effective means of "motivating" other editors to provide an inline citation. WhatamIdoing (talk) 21:38, 19 December 2012 (UTC)

• Oppose - as worded - The sentiment behind the statement is okay, but its wording ("...where you have no real concern...") asks us to read the mind of the editors. Any change to BURDEN should be based on tests that are more objective. I think the intention of this proposal could be achieved by improving the definition of what "likely to be challenged" means ... if that could be made more specific, that could help address the underlying issue. --Noleander (talk) 15:48, 19 December 2012 (UTC)

• Oppose The burden should remain on the person [re]adding the text no ifs or buts. Blueboar summed it up for me. -- PBS (talk) 00:06, 22 December 2012 (UTC)

• Oppose Removing unsourced content is very, very, very far from disruptive. — ΛΧΣ²¹ 00:08, 23 December 2012 (UTC)

Note

I feel it should be noted that SpinningSpark, the proposer of this change, is currently involved in an RFC that could be directly impacted by whether or not this change is made. Doniago (talk) 15:39, 18 December 2012 (UTC)

Its "burden of evidence, not "burden of proof"

By the way.......it is not burden of proof. It is "burden of evidence". There is a difference and this is a red flag that the proposer is not quite aware of what the burden policy is referring to. Many editors will argue that "burden" is not met even with a RS provided believing that it is not meeting a "burden of proof" that the information is accurate. We don't prove the information. We simply reference it with relaible sources. If you have not provided the source...you have not met the burden. Period.--Amadscientist (talk) 10:44, 18 December 2012 (UTC)

Continuing the responses to the original proposal

• Strongly oppose Indeed, I oppose the sentence that is there already. Unsupported material should be subject to being removed, especially (but not only) if it has been {{fact}}-tagged or similarly tagged for a reasonable period of time. We are doing our readers an injustice and not fulfilling our primary objectives by presenting material which is not verified. The material may be, as Spinningspark suggests, "entirely uncontroversial and factual" but how is our reader to know that? Finally, we judge edits, not editors, and this opens up inquiry into motivations. The existing "make it clear that you have a concern" already opens that inquiry up more than it should, creates nothing more than a knee-jerk bureaucratic burden, and ought to be removed. We need to return to the simplicity described by Blueboar in this edit. Regards, TransporterMan (TALK) 17:28, 18 December 2012 (UTC)

@North8000: I'm a latecomer to this party. Doing DR I've certainly seen the one-two punch you frequently mention in this discussion (perhaps best described by you here). This may have been discussed, but could there perhaps be a solution to that particular problem via a timing requirement that unsourced material be {{fact}}-tagged for some non-insubstantial period of time — perhaps a month — before it can be removed? Or has that idea already been thoroughly trashed here? Could there be some other more direct solution to that particular problem rather than attacking it through making WP:BURDEN more restrictive? Just asking... Best regards, TransporterMan (TALK) 18:30, 18 December 2012 (UTC)

Yes, AFAIK the use of explicit time limits to remove {{fact}}-tagged facts have been discussed to death without consensus. You say that we disservice readers by including unsourced assertions; but it is also a disservice to delete valid ideas and articles before they have had a change to be properly recorded. Wikipedia was possible because it wasn't published in a finished state but it was allowed to grow, and that core idea is still valid today. That is the reason why the "citation needed" compromise was created (and why it's the single text most associated with Wikipedia, even more than Jimbo's fundraising request). This conversation has inspired me an alternate approach that I hope may satisfy the concerns of both sides, and it might even replace the current "make it clear that you have a concern" (actually I'm predicting my approach will raise the ire of both sides, which is a good start for discussion). Diego (talk) 19:03, 18 December 2012 (UTC)

Your answer begs the question, however: Who says that those supposedly valid ideas are valid? And that's my point: the ordinary reader shouldn't be put to making that judgment or having to look up sources, even if they're easily available, in order to make that determination. If the only unsourced things that were here are those things that are verifiable, I might agree with you but that's not the way it is, not even close. Regards, TransporterMan (TALK) 19:19, 18 December 2012 (UTC)

Precisely because no one can instantly assess whether valid ideas belongs into the final article, talk is needed to elucidate that. The main problem I have with unrestricted and unqualified deletion following BURDEN is that it may get facts lost in article's history, hindering the accountability for the one that deleted the unverified content; at the very least the policy should provide a way to avoid that loss, registering when content has been deleted this way. This is usually not a problem for pages with high traffic and reviewers, but is essential for more obscure topics. Diego (talk) 10:21, 19 December 2012 (UTC)

I dare say that a {{citation needed}} on the Barak Obama article wouldn't last out the day. But in the kind of articles I'm interested in (mainly engineering) they can be very rarely visited. I have provided references on numerous occasions for articles that have been tagged for many years until I came across them. It is a great shame that such material is continually lost because because an editor like me who can be bothered to find sources has not come along before a deleter who can't be bothered. SpinningSpark 19:31, 18 December 2012 (UTC)

(added later) Transporterman, the focus of the person who made this proposal is different than my own. (which you correctly identified with that link.) I was just supporting their idea anyway. Briefly, their concern is about quickly deleting unsourced material en masse. My concern is using this provision in tandem with wp:rs "ideal source" criteria to knock out SOURCED material for POV or pissing-war purposes. To date my approach on the latter is to propose requiring a quick perfunctory questioning of the material when tagging or deleting. Now I am starting to think about wording in wp:rs that says that the strength of the sourcing (= degree of meeting the idealized wp:rs standards) should be higher for challenged material, (and implicitly, vica versa)North8000 (talk) 20:52, 18 December 2012 (UTC)

@SpinningSpark: I agree, totally, that there is unsourced information in articles which has been there, tagged or untagged, for years which is exceptionally valuable, easily verifiable, and which would be a real pity to lose, but there is other information which is worthless, impossible to verify, and absolutely misleading. The every day Wikipedia user should not be put to the task of figuring out which is which, and removing the good information does less harm, in my opinion, than retaining the bad and making it harder to remove. I certainly don't oppose strengthening the rules which require preservation on the talk page, but to say that all unsourced information ought to be kept in mainspace because it might be valuable just risks more damage to the public perception of Wikipedia's trustworthiness. Best regards, TransporterMan (TALK) 21:05, 18 December 2012 (UTC)

• Early in this discussion, I supported the proposal, and now, I see all the opposition to it that has emerged. I'm trying to weigh whether or not the existing wording – "When tagging or removing material on these grounds, make it clear that you have a concern that the material itself is unverifiable. If instead you think the material is verifiable, it is better to try to provide an inline citation yourself before considering whether to remove or tag it." – is enough to get the message across to those users who delete stuff mindlessly. Perhaps it is, and perhaps Blueboar is right to point out that it's easier to just go a get a source oneself. I think that if editors actually take to heart the existing language that I just quoted, there is no problem. But the problem, of course, is when an editor fails to understand that language, or deliberately and disruptively choses to ignore it. I know we sometimes get editors like that. But, since revising Burden to include this language, do we have examples of editors who were pointed to this language and, nonetheless, continued to disrupt? I'm leaning towards the opinion that what we have now may be good enough, but I could easily be persuaded by examples that it isn't. --Tryptofish (talk) 20:45, 18 December 2012 (UTC)

If I wanted to reopen some very painful old now 99% settled down situations, I could point to mothers of all examples of mass deleting of even sky-is-blue material (and then hairsplitting the sourcing on sky-is-blue stuff when I sourced them) in obsessive battling type situations (where there is no POV dispute). Sufficeth to say such things happen, even thought it not my own focus which is wikilawyer-using wp:burden in TANDEM with wp:rs ideal criteria to knock out SOURCED materiel which they do not even question. Usually to POV an article, but sometimes to conduct pissing wars. North8000 (talk) 22:30, 18 December 2012 (UTC)

Yes, I have no doubt. I guess the question becomes whether there is anything we can add to Burden to stop that stuff, or whether we are better off leaving things as they are and dealing with DR about the specific behaviors. Leaving aside the separate issue of sourced material debates, here the issue is unsourced stuff. I'm tentatively thinking that, if someone is doing that kind of obsessive removal of material, and if so-fix-it style finding sources (per Blueboar) can't keep up with the volume of removal, one should try to explain to the user the language that I just quoted above, and if they engage in WP:IDIDNTHEARTHAT, it becomes a matter of DR on the behavior, and not a matter of a simple fix in language here. I'm trying to think, though, of something we could say that might be addressed to the number of removals, as opposed to one or a few removals where there might actually be something to discuss on the article talk page. --Tryptofish (talk) 00:56, 19 December 2012 (UTC)

• Comment Either way, I do feel this guideline needs extra clarification, or simplification. We have an RFC at Talk:Synchronous motor with both sides arguing that the guideline supports either removing or retention, depending on their stance in the debate. If you are not going to examine the motivations of the editors, then you have to accept the guideline is going to be applied bureaucratically in some cases, and you need to say in no uncertain terms that removing unsourced material for whatever reason is legitimate. Anything short of that, and you are saying that it is acceptable to examine an editor's motivations for removing such material. You can't have it both ways, since a guideline is not much use if it can fit diametrically opposed interpretations. Betty Logan (talk) 21:35, 18 December 2012 (UTC)

As one of the editors involved with the RFC, I'd like to state that I entirely agree with Betty's points above. I'll add that if policy is to be revised then I feel the "citation needed" templates may need to be revised as well, as at least some of them explicitly state that "unsourced material may be removed". I've found it's very difficult and stressful to get involved in discussions regarding unsourced material precisely because the policies and guidelines, which are understandably designed for flexibility, end up allowing for conflicting interpretations. Doniago (talk) 21:47, 18 December 2012 (UTC)

You could always try common sense instead. That sometimes works. SpinningSpark 22:22, 18 December 2012 (UTC)

I consider this a personal attack, and not the first one you've made against me recently. Please stop. Doniago (talk) 22:58, 18 December 2012 (UTC)

Advising an editor to use common sense is a personal attack now? Do you not think that you should be applying policies with common sense? SpinningSpark 00:17, 19 December 2012 (UTC)

I'm afraid your comment implies he hasn't been using common sense recently, that's why he's taking offense. Diego (talk) 00:28, 19 December 2012 (UTC)

I'm not implying it, I'm categorically stating that he hasn't been. Doniago's style of editing is precisely the kind of disruption that this proposal is aimed at stopping. His recent foray into synchronous motor, a subject he self-admittedly knows nothing about, was the provocation for making this proposal. There he continues to argue for deletion in the face of opposition from a dozen knowledgable editors. But it is nor just that article. His talk page makes it very clear that he is making a career of this kind of thing. He has also provoked enough opposition in the past for an ANI case to be brought against him. Nor is Doniago the only editor doing this, another case at AIV has just been closed involving Epeefleche. It is not common sense to delete stuff just because we can. A more reasoned approach is more helpful and it would be good if policy made that a bit clearer. SpinningSpark 01:14, 19 December 2012 (UTC)

I'm not saying that what Doniago is doing right now does or does not display common sense, I haven't looked at it, but I do say that for the reasons that I've stated above that deletion of material merely because it is unsourced is not devoid of common sense. Regards, TransporterMan (TALK) 01:44, 19 December 2012 (UTC)

How about this approach, instead?

Maybe leave the main text of Burden as is, but add a footnote at the end of the sentence, "When tagging or removing material on these grounds, make it clear that you have a concern that the material itself is unverifiable." The footnote would be:

"When tagging or removing such material, please keep in mind that other editors may perceive your edits as disruptive if you seem to be paying attention selectively to material of a particular POV, or if you appear to be removing large amounts of material without really having taken the time to assess whether or not it is verifiable. For this reason, it is advisable to communicate clearly that you have a considered reason to believe the material in question is not verifiable."

Instead of saying that policy requires a particular behavior, that then could be wikilawyered, a footnote of this sort approaches the issue from the point of view of advice about how to avoid conflict. It doesn't mandate anything. It doesn't water down the principle that unsourced material can be removed. But if an editor acts in a manner contrary to what it says, it makes possible a DR discussion focusing on whether the tagging/removal is disruptive or not. --Tryptofish (talk) 01:09, 19 December 2012 (UTC)

I haven't read all of the above yet since last night, but this caught my eye and I had to say how much I agree with this. Whether this is what happens or not...bravo for the suggestion!--Amadscientist (talk) 01:28, 19 December 2012 (UTC)

I can certainly buy into that. I suggest that the word "really" is superfluous. Arguing against my own position slightly, I think that adding "with at least a cursory online search" at the end of the first sentence would be helpful. We don't want editors to be criticised for not spending a fortnight doing research at the Library of Congress or the British Musuem. SpinningSpark 01:26, 19 December 2012 (UTC)

I would support the proposal of Tryptofish without any further alterations. I don't agree with telling editors to do the research the contributer refused to do.--Amadscientist (talk) 01:30, 19 December 2012 (UTC)

If you don't agree with that, then surely you would take issue with the imputation in Tryptofish's proposal that removing unsourced material needs a "considered reason"? If I'm copyediting an article, do I always need to leave a mass of talkpage messages about unsourced stuff I may have removed or tagged for a variety of reasons? Is believing something to be untrue or harmful if untrue a "considered reason", or do I need to go further? What's my considered reason for adding a fact tag, beyond "the statement was uncited"?

Overall, I agree with the proposal in spirit, but I don't think it requires so many words and I don't think we need to say what is advisable. How about:

It may be considered disruptive to delete or tag material in a way that is not balanced, taking into account WP:NPOV.

or something like that? Formerip (talk) 01:51, 19 December 2012 (UTC)

Or taking into account WP:V itself. 'It must be possible to verify this material, using some published reliable source' (our actual policy) is not the same thing as 'All material not followed by an inline citation is unverifiable'. WhatamIdoing (talk) 01:57, 19 December 2012 (UTC)

(ec) It's not always a case of "the contributor refused to" provide sources. Sometimes the original person did fully source the material, but the refs were removed (e.g., by a person who refused to follow the directions at WP:DEADREF), or the text may have been changed so many times that the refs got detached and nobody remembers that the citation in the next paragraph covers this material, too, or someone inserted text in between the original material and the original citation, so that the original material appears unsourced and the new material improperly appears to be sourced. Or, for that matter, I've encountered editors who believe that citations at the end of a paragraph can't possibly support the whole paragraph. We don't have a one-citation-per-sentence rule, but some people use that claim as a means of "challenging" fully sourced material. There are many reasons why a citation doesn't happen to appear in a given location that have nothing do with the original contributor refusing to cite the material. WhatamIdoing (talk) 01:57, 19 December 2012 (UTC)

Now you are suggesting that the editor is responsible for every other editors improper actions. 'Quantity over quality' is a bad premise for your reputation, as Toyota found out. -- TRPoD aka The Red Pen of Doom 02:11, 19 December 2012 (UTC)

No, I'm only saying that it's neither accurate nor kind to assume bad actions by the person who originally added the material. If you add good, sourced material, and someone else screws it up later, then we should blame whoever screwed it up later, not go around saying that you "refused to" provide citations. WhatamIdoing (talk) 21:41, 19 December 2012 (UTC)

I oppose the "or if you appear to be removing large amounts of material without really having taken the time to assess whether or not it is verifiable," when added to the "editors may perceive your edits to be disruptive". Those combined at least imply that the perception could be valid. All that is required by the policy right now is that one claim that one have concerns about verifiability, which is wrong and misleading because there's nothing which actually requires you to have those concerns. Indeed, what the policy says is, "Any material lacking a reliable source directly supporting it may be removed." As it is right now, while it may be distasteful and may not be the preferred thing to do, and while some may be so upset about it to file unfounded and inappropriate complaints about people who do it, removing material merely because it is unsourced is perfectly acceptable unless it violates some policy other than WP:V such as POV. The ongoing attempts to change WP:BURDEN are an effort to slip a restriction on that in through the back door or by pecking away at the concept because that result cannot be achieved directly, as was tried and failed in the recent RFC. Regards, TransporterMan (TALK) 02:22, 19 December 2012 (UTC) Supplement: I withdraw and apologize for the struck-through material. Making those kinds of accusations is not my style and I've felt bad about it ever since I said it. The excellent editors on both sides of this discussion have the true welfare of the encyclopedia at heart and absolutely did not deserve this kind of thing from me. Mea culpa, mea culpa, mea maxima culpa. Ashamedly, TransporterMan (TALK) 17:56, 20 December 2012 (UTC)

Just noting what I already said on my talk page: no offense taken by me, at all! --Tryptofish (talk) 22:02, 20 December 2012 (UTC)

TransporterMan, how do you interpret the part that says "Whether and how quickly this should happen depends on the material and the overall state of the article", and that is also part of policy right now? Diego (talk) 10:25, 19 December 2012 (UTC)

Transporterman, I wish (and have been advocating) that what you said ("required by the policy right now is that one claim that one have concerns about verifiability") was true but it isn't. It is certainly contained in the spirit of "challenged or likely to be challenged", but the operative/ nuts and bolts of the policy says the opposite. North8000 (talk) 12:33, 19 December 2012 (UTC)

@Diego: I see that phrase as being so open-ended and subject to interpretation in the eye of the beholder as to render it effectively meaningless as policy. Depends how and in what way? Is unsourced information about My Little Pony pony colors more subject or less subject to quick removal than unsourced information about whether Serbs were systematic victims of The Holocaust? Is unsourced information more subject or less subject to quick deletion in GA's than it is in stubs? I'm sure that we both have opinions about such questions, but that phrase gives little guidance as to which of our opinions is right. And the use of "should" arguably makes the answer to the whole question discretionary. Thus, read in the entire context of WP:BURDEN the phrase appears to be more advice than a rule, and that's how I've always interpreted it and taken it. If I understand what North8000 is saying in his last comment, just above (and I'm not sure that I do and I apologize to him in advance if I'm misinterpreting what he is saying), he may be saying that everything in that paragraph after the initial first sentence, "[a]ny material lacking a reliable source directly supporting it may be removed," may only be advisory, and I think that he may be right. Best regards, TransporterMan (TALK) 14:39, 19 December 2012 (UTC)

Re ""Whether and how quickly this should happen depends on the material and the overall state of the article"" — I interpret this to mean how likely the material is verifiable, how long it has been in the article, and what effect its removal would have on the article. For example, if the material seems verifiable, been in the article for a long time, and constitutes the main part of the article, then efforts should made to find reliable sources rather than to delete. Would changes to the sentence or an additional sentence along these lines be worth considering? --Bob K31416 (talk) 14:59, 19 December 2012 (UTC)

• (edit conflict)Is unsourced information more subject or less subject to quick deletion in GA's than it is in stubs? We have an answer for at least that part. The sentence is to be seen in the light of WP:IMPERFECT and WP:PRESERVE, "the editing process tends to guide articles through ever-higher levels of quality over time", meaning that low quality stubs have more leeway for including "citation needed" during longer periods. Precisely at those stubs is where it's more helpful to tag content instead of removing it. Could we concur to make the requirement to tag-or-move-to-talk-instead-of-remove a strong suggestion for less-developed articles, and a weaker one for good quality ones? After all, WP:PRESERVE already makes it a strong suggestion; I think it would also clarify BURDEN if we explained that sentence. Diego (talk) 15:05, 19 December 2012 (UTC) P.S. (Of course this wouldn't apply to vandalism nor BLP-infringing content). 15:32, 19 December 2012 (UTC)~

Probably the simplest way to state my idea is to bring the operative wording in compliance with (instead of conflicting with) the spirit of "challenged or likely to be challenged", noting that "challenged or likely to be challenged" is a characterization of the material itself,....not it's sourcing status, but sourcing is what it then calls for in those conditions. North8000 (talk) 15:25, 19 December 2012 (UTC)

(I've not read the posts since my last posting above, so this is not intended to comment upon them.) General comment: I'd like to say that I've always seen a difference between an editor who merely occasionally deletes unsourced information and an editor who makes a regular practice of going around deleting unsourced information from a number of different articles without making an effort to see if the information is or is not verifiable. This policy makes it clear that the best practice is to either find sources for the information and add them or to at least make sure that the information is not verifiable before deleting it. It is, however, acceptable to delete it without checking. I do believe the second kind of editor is making himself vulnerable to being sanctioned through blocks and bans, whereas the first kind is not. Though it's not set out anywhere that I've ever been able to find (though I must admit that I haven't looked very hard), community members can and have been blocked or banned simply because of chronic conduct unbecoming a Wikipedian. Over the course of one's Wiki-career and as an everyday matter of behavior, upstanding — not outstanding, just upstanding — members of the community do things the "best practice" way or they don't do them at all and folks who chronically do it the "merely acceptable" way are asking for trouble, not for violating the rules per se but simply for not being responsible citizens. That can happen quicker if their chronic mass deletions can be tied to violation of some other policy or more blatant irresponsiblity such as POV-pushing, edit warring, or perhaps COI, but it can also happen just because of not doing it in the socially-responsible way. The unpleasant thing about that is that they can arguably do quite a bit of damage (although there is an argument to be made, which I have indeed made in the main part of this discussion, above, that it's not really damage) and cause a lot of drama and disruption, before their behavioral pattern becomes sufficiently defined and the community becomes sufficiently fed up. In my opinion, we need to preserve the acceptable conduct level to provide for the situation in which good, upstanding, fully-and-regularly-responsible members of the community spot marginal information and want to delete it without having to search to see if sources are there and without becoming involved in huge amounts of drama when some fanboy or fringe-flogger starts screaming about there being huge amounts of reliable sources about bigfoots coming to Earth on ancient Mayan spacecraft. I'm not opposed to tightening down the standards to let us deal with the irresponsible mass deleters more quickly, but not by throwing out the acceptable level of conduct for use by responsible, but busy, editors. Maybe one place to start would be to simply say make the community-responsibilty standard a bit more explicit by this addition to WP:BURDEN:

"While it is acceptable to occasionally delete unsourced material merely because it is unsourced, all editors have a responsibility to avoid chronic, continual, or regular deletion of unsourced information without either attempting to find a reliable source for the information or making a good faith determination that no such sources exist. If an editor's editing history shows a pattern of chronic, continual, or regular deletion of unsourced information but does not also show frequent additions of reliable, inline sources to unsourced information, the editor may be subject to blocking or banning if the information which has been deleted is not of clearly doubtful verifiability on its face or if the information being deleted indicates that a particular point of view is being pursued or that a conflict of interest is involved."

(There is, incidentally, a third situation which I find much more difficult to approach conceptually: the deletion of a massive amount of unsourced information in a single article or group of closely related articles in a single act or spree.) Best regards, TransporterMan (TALK) 15:33, 19 December 2012 (UTC)

Revised footnote

• I've examined carefully the issues that TransporterMan, SpinningSpark, and WhatamIdoing have – correctly, I think – pointed out as flaws in my original suggestion, and I think it may be possible to revise the footnote to correct those problems. (I, for one, am certainly not trying to slip anything in through the back door!) Revised footnote:

"When tagging or removing such material, please keep in mind that other editors may object to your edits if you seem to be paying attention selectively to material of a particular POV, or if your edit history shows a pattern of chronic, continual, or regular deletion of unsourced information without any effort to check for the existence of sources. Sometimes, material may have been supported by an inline citation in the past, but the citation was removed by subsequent edits. For these reasons, it is advisable to communicate clearly that you have a considered reason to believe that the material in question is not verifiable."

• Better? --Tryptofish (talk) 18:32, 19 December 2012 (UTC)

I really like it. Actually I really like the beginning and the end, and stuff in between is necessary filler. North8000 (talk) 18:38, 19 December 2012 (UTC)

• Seems appropriate for an essay rather than policy. --Bob K31416 (talk) 20:39, 19 December 2012 (UTC)

• Comment I've been wanting to comment on this for a while, but haven't had a chance yet. And now that I take a look, I see that it has turned into a wall of text. :( So let me just say agree that we have a real problem here, and I support efforts to improve the policy. A Quest For Knowledge (talk) 20:42, 19 December 2012 (UTC)

As for an essay and a wall of text, yes, the discussion is a wall, but the proposal is just a short footnote to the policy. And we could delete the sentence about "Sometimes, material may have been supported by an inline citation in the past, but the citation was removed by subsequent edits." Someone suggested the idea, but I don't think it's really necessary here, and there seems to be some sense that the proposal should lose some weight in its middle regions. --Tryptofish (talk) 20:47, 19 December 2012 (UTC)

Essays are useful. Suggest putting it there with more similar thoughts, and when it matures sufficiently, put a link to it from this policy. Also, it might become a guideline, partly a behavioral guideline. --Bob K31416 (talk) 21:20, 19 December 2012 (UTC)

Understood. First, let's see what people think about a footnote here, since there seems to be some interest in covering these issues here on the policy page. But I agree, if there isn't consensus, then it won't be here. --Tryptofish (talk) 21:24, 19 December 2012 (UTC)

We're not talking about giving guidance to well intentioned people, which is what an essay would do. We're talking about stopping people from deliberately mis-using wp:ver. North8000 (talk) 21:29, 19 December 2012 (UTC)

I think something stronger is required than an essay. We already have WP:PRESERVE and editors ignore it. A Quest For Knowledge (talk) 21:33, 19 December 2012 (UTC)

I really like the idea of a footnote and not an essay (although why not create an essay that elaborates on the footnote) and think Tryptofish has the right idea.--Amadscientist (talk) 21:37, 19 December 2012 (UTC)

That works for me. (I suspect that you mean "frequent" rather than "regular".) WhatamIdoing (talk) 21:45, 19 December 2012 (UTC)

Good, thanks all. I'm fine with changing "chronic, continual, or regular deletion" to "chronic, continual, or frequent deletion", or even "chronic or continual deletion" or "frequent or continual deletion". I simply copied that phrase from what TransporterMan had written in a comment. I'd also be fine with removing the sentence about "Sometimes, material may have been supported by an inline citation in the past...". What do we think about that? --Tryptofish (talk) 21:54, 19 December 2012 (UTC)

I agree you need to take out the bit about past citations. That is going off at a tangent from the point we are trying to make and is a situation that is easily dealt with. If one suspects that to be the case one just needs to find the old citation in the history and restore it. SpinningSpark 22:41, 19 December 2012 (UTC)

I like it. You should update the proposed change to provide easy clarity on what it is. North8000 (talk) 10:47, 20 December 2012 (UTC)

Thanks, I agree that a clearer update is needed, and seems to be moving in the right direction. But I'm going to let it marinate another day or two before I do, to see how #Footnote suggestion revisted, below, does, how we feel about the edit summary proposal by Diego, and whether there are any opinions about "chronic, continual, or regular deletion". As that comes into focus, I'll revise it accordingly. --Tryptofish (talk) 22:11, 20 December 2012 (UTC)

I'd want to include the following addition, after "For these reasons, it is advisable to communicate clearly that you have a considered reason to believe that the material in question is not verifiable; placing the word BURDEN or leaving a link to this section, either at the edit summary or article's talk page, is enough to satisfy this need", for the reasons I state in the section below. By making it a suggestion and not mandatory, I hope this can satisfy Amadscientist's concerns that it adds too much instruction WP:CREEP, while retaining my hoped benefits - by suggesting a sign to look for that can be commonly used. Diego (talk) 12:22, 20 December 2012 (UTC)

I find myself with a bit of a dilemma here. I'm basically okay with Tryptofish's 18:32, 19 December 2012 (UTC), version of the footnote, either with or without the second or third sentence, but I'm opposed to the continued inclusion of the sentence to which it will be a footnote, so while I support that version of the footnote, I'm opposed to the whole thing. Best regards, TransporterMan (TALK) 14:50, 20 December 2012 (UTC)

On the other hand, I weakly oppose Diego's 12:22, 20 December 2012 (UTC) suggestion (just above) and oppose Diego's 23:13, 19 December 2012 (UTC) "Another suggested addition" below. I only weakly oppose the one above because it is in a sentence conditioned with "it is advisable" whereas the one below simply adds to the burden I generally oppose, i.e. having to say anything at all. (I also oppose it because it would keep me from universally making clear my intent by putting a userbox on my user page which says, "I never delete material unless I have concerns it is unverifiable." Just kidding.) Best regards, TransporterMan (TALK) 15:01, 20 December 2012 (UTC)

Would you mind explaining why your opposition to the weaker form, in which the sentence creates no burden and is included only as a suggested way to express one's concerns? Diego (talk) 16:55, 20 December 2012 (UTC)

My reason for both oppositions is really the same: that I'm opposed to having to say anything at all. My objection to the one below is stronger because it more directly increases that burden (which, per the discussion above, may not be mandatory at all), while my objection to the version above is weak because it still imposes a standard, and thus more firmly entrenches the notion that something should be said in the first place, even if it is just a suggested standard. Best regards, TransporterMan (TALK) 17:45, 20 December 2012 (UTC)

TransporterMan, I'm sure I'm telling you what you already know, but I don't think we are likely to undo the sentence in the main text (to which the proposed footnote would apply). Diego, I understand how having a clear edit summary is a good thing, but I keep feeling like it's WP:CREEP to recommend what the edit summary should say. If we tell editors to "communicate clearly that you have a considered reason to believe that the material in question is not verifiable", that covers it. How they should "communicate" it "clearly" should be up to them. --Tryptofish (talk) 22:16, 20 December 2012 (UTC)

I'm not so sure. When Mr. Stradivarius closed that RFC he was careful to note that there were only 12 editors in the discussion and that "this close is not the final word on this subject; editors are encouraged to open up new discussions about further changes to the proposal, whether those changes have already been considered in this thread or whether they are completely new ideas." (Emphasis added.) I'm not prepared to do that at this time, it's too soon, but I don't want the idea to become any more entrenched than it is already since it was a weak consensus at best. Best regards, TransporterMan (TALK) 22:26, 20 December 2012 (UTC)

Yes, that's fair, WP:CCC. But I'll put it this way: I'll oppose removing that sentence! --Tryptofish (talk) 22:29, 20 December 2012 (UTC)

Actually, I just noticed (it appears I'm slow today!) that Bob K revised that sentence, and in my opinion, it's an improvement. --Tryptofish (talk) 22:42, 20 December 2012 (UTC)

Another suggested addition

I propose another small addition for the case when content is finally deleted, to make it easier to find it in the future:

"When tagging or removing material on these grounds, make it clear that you have a concern that the material itself is unverifiable by including the word BURDEN in the edit summary."

We're assuming that editors will be able to find the removed content through article's history. This was true in the early years when history lists were sort, but nowadays content edited more than a few months ago will usually be out of reach. Tools like WikiBlame can help locate particular text inside the article, when one knows that some content was removed and has a reasonable idea of how it was worded; but reading the edit summary is the only way to know the intent of editors.

If we include in policy the specific sign to be used for removing unsourced content, all editors removing content will tend to use the same word to signal it, and those removals will be easier to find by someone trying to recover and source removed content. Many editors are already including WP:BURDEN in the edit summary as the reason to remove content; this change would increase the numbers of edits that are marked this way. I know that this proposal is not perfect as not all editors will know or remember to always include the signal, but it will help even if it's used only in 80% of removals of this type.

It also has the benefit that, once content has been removed including this signal, this policy applies in full force and the content can't be restored without inline citation; it's a way for the editor to signal that they know and understand the policy and that they are following its requirement to identify their concern. This avoids the "need to read other editors' minds" problem that was raised above. Diego (talk) 23:13, 19 December 2012 (UTC)

• TL;DR version: always put BURDEN in the edit summary when deleting unsourced content. This way you've complied with your requirement to claim it unverifiable, and I can find your removals and try to source them. Diego (talk) 23:18, 19 December 2012 (UTC)

• Oh good... More instruction creep! (sigh) We can not mandate that editors put certain words in their edit summaries. In fact, we can not even mandate that they even leave an edit summary (I often forget to do so). Blueboar (talk) 23:31, 19 December 2012 (UTC)

No problem then; the removal in that case wouldn't be protected by WP:BURDEN and could be restored by anyone. Note that it doesn't add more rules, as you're already required to state your concerns about unverifiable content; it only adds consistency to the current rule. Placing the signal would end all possibility of wikilawyering about the intent of the editor removing content. And it would really help support wiki process, providing accountability for removed content. Diego (talk) 23:35, 19 December 2012 (UTC)

I had kind of the same reaction as Blueboar did. It's a perfectly good idea for an edit summary, but I think that enforcing this would be impossible. In a way, it's self-evident that an editor who reads BURDEN and bases an edit on it would be, well, basing their edit on it. --Tryptofish (talk) 23:41, 19 December 2012 (UTC)

I understand that an editor is concerned with content being "lost" in the edit history when it is deleted. However I would remind the editor that the edit history is there precisely for retaining the content in some form to be retrieved at a later date. Think about this a minute....if "verifiable" content is being left unsourced simply because there is a source out there somewhere and we are speaking about this as a reason not to delete it, is the same not true of the content in the history? It is still there and just because one editor cannot find it, doesn't mean another editor can't.--Amadscientist (talk) 23:46, 19 December 2012 (UTC)

By that logic we wouldn't have cleanup templates, deletion sorting, nor tags in edit summaries to search edits made by bots. Diego (talk) 10:28, 20 December 2012 (UTC)

I'm not making this up as I go along. You know very well that the history is there for a reason. If you can't take the time to look through it then please don't create issues that are simply not there.--Amadscientist (talk) 10:46, 20 December 2012 (UTC)

Let's make this clear: your suggestion for someone wanting to review this for removed information is to review the diff for each single edit and try to find the reasons why something was removed? Diego (talk) 11:32, 20 December 2012 (UTC)

This procedure reinforces the bright line that WP:BURDEN is supposed to be by defining an unambiguous test to apply the policy:

• Content removed without a BURDEN comment is not challenged and any editor can restore it.
• Content removed with a BURDEN comment is challenged. Editors restoring it without inline sources are breaching policy. No exceptions.

This exact test clarifies the current wording that requires content to be "challenged" and editors to state their "concerns". It would also silence the claims that the deleter didn't properly try to source the content. An editor knowledgeable enough to include the BURDEN comment in the edit summary is making an explicit claim that, yes, he regards the content as unverifiable. This makes that editor responsible for those claims in case that they're misused or lightly made over a prolonged period.

In short, I believe this clear test would help reduce the edit warring and discussion about the meaning of BURDEN that is usual nowadays. Diego (talk) 10:28, 20 December 2012 (UTC)

No. This just does not make any sense. You think an edit summary should qualify a deletion and when not using the term doesn't. LOL! That is really not going to work. That is true instruction creep and just trying way to hard to put up an obvious brick wall to keep unsourced content. We are now beginning down an absurd road. I suggest we stick with the footnote suggestion as it is, by far, the best suggestion being made.--Amadscientist (talk) 10:46, 20 December 2012 (UTC)

"[D]eleter didn't properly try to source the content". This clearly shows you are simply trying to place "blame" (edit) responsibility on the backs of editors that remove the content and that really is the exact opposit of Burden. Deleters don't have to attempt to source it, just verify it...and even that is not a requirement. Policy states that the burden is on those wishing to return it. You now want the burden to be on those removing it. It really should just be the burden of those adding any information. Then we wouldn't have this issue. Since that isn't going to happen why would we then demand others do the research for those not capable of it to begin with?--Amadscientist (talk) 10:54, 20 December 2012 (UTC)

Yeah...blame is too strong.--Amadscientist (talk) 11:56, 20 December 2012 (UTC)

Ok. Well, I'm not the one adding that responsibility; it's already there in the policy your duty to communicate your concerns, and if you fail to do it you're breaching your part of the deal. The only thing my proposal changes is that now, when you fulfill that already existing responsibility, you're shielded from others trying to attack you. Diego (talk) 12:02, 20 December 2012 (UTC)

You kinda are with a requirement to add a specific word. And it has no effect on any shielding...it actually creates issues as this would not be something people will know or even understand and may well just be ignored creating loads of drama. As I and many others have continued to say, there is not real "obligation" or "requirement" to communicate in this manner. As I have said before elsewhere, a deletion itself is a challenge to the content.--Amadscientist (talk) 12:15, 20 December 2012 (UTC)

What would you think of this addition if it was not an obligation nor a requirement? Diego (talk) 13:00, 20 December 2012 (UTC)

This clearly shows you are simply trying to place "blame" on the backs of editors that remove the content Now I think you didn't understand the proposal. This would achieve exactly the opposite, by protecting you from those other editors that "might object if you remove material without giving them time to provide references" and that the policy warns you about. Diego (talk) 11:27, 20 December 2012 (UTC)

Also see that I wrote "claims that the deleter didn't properly try to source the content". I'm not the one making those claims; I want to avoid them being made. Diego (talk) 11:51, 20 December 2012 (UTC)

Footnote suggestion revisited

I suggest brevity might be a good idea. Inspired by North800's suggestion:

After this the prose in "Burden: "When tagging or removing material on these grounds, make it clear that you have a concern that the material itself is unverifiable." add the following shortened version of Tryptofish's suggested footnote:

When tagging or removing such material, other editors may have various objections. For this reason, it is advisable to communicate clearly that you have a considered reason to believe that the material in question is not verifiable.

--Amadscientist (talk) 11:29, 20 December 2012 (UTC)

I also think the "challenged or likely to be challenged" should have a footnote explaining what is meant by this to clarify it.--Amadscientist (talk) 11:31, 20 December 2012 (UTC)

After the term "challenged or likely to be challenged" place a footnote similar to:

A challenge could be a talkpage post questioning the validity of verifiability or an edit, such as deletion of content for no sources or a tag requesting sources. "Likely to be challenged" is any content that is not unquestionable fact.

--Amadscientist (talk) 11:39, 20 December 2012 (UTC)

Nice try, but I think it confuses matters more. North8000 (talk) 12:29, 20 December 2012 (UTC)

I guess I'd rather have it clearer what those "various objections" would be, and why. It also isn't clear to me that we really need to clarify what "challenged" means. (Is it the posting of a challenge, or is it skepticism about the material?) --Tryptofish (talk) 22:06, 20 December 2012 (UTC)

Is there a way it could be adjusted to be less confusing to you North? Typto, it is both. if you have a skepticism it is liely to be challenged and the act of challenging is what I listed. I see confision in the very question which leads me to believe it needs clarification. Maybe not mine...but it needs clarification.--Amadscientist (talk) 23:02, 20 December 2012 (UTC)

This bit: "Likely to be challenged" is any content that is not unquestionable fact. is wrong. We're using the dictionary definition of likely in this policy. You should supply an inline citation (at the time you originally add the material) if, using your very best judgment, it's likely or probable that the claim will actually get challenged (assuming the article is read a typical number of times, etc.) If it's unlikely that the claim will get challenged, then you are permitted to wait for an actual challenge.

The "any content that is not unquestionable fact" standard is the one we use for actual challenges. That is, if it's truly an unquestionable fact (e.g., the normal number of fingers on the human hand), and some kid decides it would be funny to challenge it, then you're permitted to revert the fact tag or deletion on the unquestionable material even though the material has supposedly been "challenged". WhatamIdoing (talk) 16:43, 23 December 2012 (UTC)

Unverifiable as a standard

I think the reason all the above proposals are being rejected is that including unverifiable in the policy was poor choice. The discussion that led to it was poorly attended and was a small part of a long drawn out discussion. While it wasn't a big deal at the time, now editors are grasping on the wording to shift policy substantially in a way that was never intended by the original discussion. To form a belief that something is unverifiable could mean any of a variety of things. Does it mean that I think the claim is wrong and my impression looking at the article, but not doing any research, is that I think it is unlikely that a search will easily verify the claim? Is it that I want to remove the claim and have done a rudimentary online search for sources (as in googling for a few minutes) and found no acceptable sources for verification. Or is it that I have extensively searched to such an extent that I can say with reasonable confidence that it actually cannot be verified. Prior to the October discussion, the policy was clearly in line with the first of the versions, after the RFC, maybe its the 2nd. But even if you maintain that the discussion close intended to move further then that middle ground, it certainly did not receive sufficient participation to support such a dramatic shift in a core policy.

To resolve this, I suggest we amend the unverifiable language as follows:

When tagging or removing material for not having an inline citation, make it clear that you have a concern that the material is unverifiable not easily verifiable.

This would be a much more workable standard, and would target those indiscriminately removing content without trying to source it, but would preserve the long standing BURDEN design by keeping it on those who add or restore content. Monty₈₄₅ 18:18, 20 December 2012 (UTC)

That wouldn't work because it wouldn't discourage deletion of material if the editor thinks the material is not easily verifiable yet thinks that it is verifiable if one worked hard enough to find a source. Another way of looking at it, is that it implicitly says it is OK to delete verifiable material that is not easily verifiable. --Bob K31416 (talk) 19:58, 20 December 2012 (UTC)

How is an editor really supposed to know how hard or impossible it is to verify something without having tried? At best we are talking about a guess, and where different editors will make different guesses, its a bad idea to have policy turn on them. Monty₈₄₅ 21:28, 20 December 2012 (UTC)

One thing that might help to crystallize and focus the discussions is to identify the types of situations that most merit a policy tweak to reduce the problem. North8000 (talk) 21:46, 20 December 2012 (UTC)

I think the best approach is to recognize that BURDEN really has two separate requirements - two situations, if you will - one for removing content, a different one for bringing it back. (Sorry to those arguing that you can freely remove any unsourced content at any time; there's simply no consensus for that). The policy should clarify (without changing the current requirements) the expected behavior of both the people trying to keep and remove the content.

Nobody denies that the burden to cite and keep material lies with those in the keep side, and without sources it will be ultimately removed; but those challenging the content also have a burden to avoid disruption and losing the facts because of a careless removal. This is in line with the consensus that was developed at the latest RfC. Any revision should be in line with helping both sides to better develop consenus in a situation like this. Diego (talk) 21:55, 20 December 2012 (UTC)

I guess I don't agree with a couple of the underlying premises here. First, I don't think everything being discussed is being rejected! Beyond that, I feel that the page does explain what "verifiable" means. (If it doesn't, we've got a big problem!) Given that, I'm not sold that it's unclear what "unverifiable" means. Once we get into distinguishing between easy and not-easy verifiability, we open ourselves up to wiki-lawyering about where the line between them is. --Tryptofish (talk) 22:22, 20 December 2012 (UTC)

I think that "unverifiable" is the safe word for the no-no-fifth rail phrase "looks doubtful". I.E. going around tagging lots of "sky is blue" statements is more discouraged than tagging lots of "sky is green and purple polka dots" statements. In some circles they figure its verboten for editors to know and decide which color statement should and shouldn't experience a full whack from wp:burden. North8000 (talk) 22:25, 20 December 2012 (UTC)

True, but the flip side of that coin comes if we attempt to legislate a safe word for "looks beyond doubt". --Tryptofish (talk) 22:31, 20 December 2012 (UTC)

True. Continuing, there are lots of POV warriors who would knock out "sky is blue" if they just have to say "not a wp:rs" but won't if they have to look stupid and say "I question that the sky is blue" North8000 (talk) 22:37, 20 December 2012 (UTC)

I personally have no desire to go on content removal crusades. But my concern is that when there is a dubious fact, WP:BURDEN is a tool to stop the proponent of the statement from dragging it out into a really unproductive argument other whether it could be verified. Such arguments are particularly pointless because the proponent of the material could just provide the verification. My concern is that expanding the unverifiable requirement will return us to incredibly pointless arguments over whether the claim really is unverifiable or just hard to verify, and that it should therefor stay despite the lack of verification. Monty₈₄₅ 22:53, 20 December 2012 (UTC)

Perhaps I'm confused, but it seems to me that it's better, in that regard, not to differentiate between easily and not-easily verifiable. If someone wants to challenge a dubious statement, they can tell the proponent of the statement that they question its verifiability and want to see a reliable source. If instead they question whether it's easily verifiable, then the proponent can argue that they would have found sourcing if they had tried harder. I'd rather not draw such discussions into the question of ease of finding sources. --Tryptofish (talk) 23:06, 20 December 2012 (UTC)

Before I read through the replies I have to admit I am a little surprised by the recent change in wording. I gave up on the discussion very early whn this first began to develope and I am a little disapointed bu the outcome of the recent change. I am not sure that what is there now in the lede is appropriate at all and seems confusing still. The point is supposed to be that content ve verifiable, but..as monty points out...to what extent? I am very concerned about this issue as not understandable and the discussion clearly shows that. Look, I think have a grip on the spirit of the issue, and I think North, Monty and Typto have a true sense of the spirit of the guideline(there are more editor here that I feel that way about as well but those are the more recent replies). I truly believe some editors are attempting to push a policy forward that simply makes it a wikilawyering excuse to arm warriors who want to keep unsourced content, Not fake or false content but unsourced content based solely on the fact that they "know" a source exists that can verify it. There seems to be a devide. On one side those who believe that (even with one source) all content that is likely to be challenged by anyone should be referenced with an inline citation. Lets face a fact here. NO article can exist without at least one source as an overarching reference. Some editors will simply palce the single source at the bottum and not make any inline citations, but that is not correct. If you have only one source and you are still making claims from that source you are indded supposed to place a citation from that source to the page the claim is made.

This needs some attention from a WMF fellowship participant in my view. This policy page is just confusing.--Amadscientist (talk) 23:20, 20 December 2012 (UTC)

• Monty, Just for the sake of discussion, would you have any problem with the following sentence?

"When tagging or removing material for not having an inline citation, make it clear that you have a concern that the material may not have been published in a reliable source."

--Bob K31416 (talk) 03:50, 21 December 2012 (UTC)

I have no problem with that language. Monty₈₄₅ 04:25, 21 December 2012 (UTC)

Ok. Then how about this modification of the above version?

"When tagging or removing material for not having an inline citation, make it clear that you have a concern that the material may not have been published in a reliable source, i.e. may not be verifiable."

--Bob K31416 (talk) 04:52, 21 December 2012 (UTC)

Again, no objection, the use of may gives sufficient flexibility. Monty₈₄₅ 16:48, 21 December 2012 (UTC)

Bob... lets take a look at what occurs if we do add this... what happens if someone does not make their concern clear? More importantly, what if a wikilawyering editor feels that the removing editor has not made their concern clear enough? Blueboar (talk) 16:54, 21 December 2012 (UTC)

I don't see how that is any less of an issue with the current language. Monty₈₄₅ 17:00, 21 December 2012 (UTC)

I agree with Blueboar's concern. Though I don't think that anything after the first sentence of that paragraph is binding, as now written, an editor grinding a point (I, just for the record, object to the offensive term "wikilawyer" even though, as I point out there, I use it on occasion) can argue that the other way. I'm afraid that it creates a Catch-22 for editors who remove uncited material: if they don't make the declaration, they'll be harassed for that, if they do make the declaration and it proves that it would have been pretty easy to determine that the material was verifiable then they'll be accused of lying (even though the sentence only requires a concern and even though a version which would have required to to declare that the material was, in fact, unverifiable was rejected in the discussion which resulted in this sentence). More drama, not less. Best regards, TransporterMan (TALK) 17:09, 21 December 2012 (UTC) PS: And, by the way, even if I'm wrong and the declaration is mandatory, I see nothing which would suggest that failing to include it in any way invalidates the removal. It may cause the remover to be subject to sanctions for not including the declaration, but the removal is still valid and WP:BURDEN applies to any restoration of the material. Wikilawyering? I'm thinking we need to bring back the AMA to defend people unjustly accused of violating that sentence. (Just kidding.) — TM 17:18, 21 December 2012 (UTC)

Agree with Monty. Strikeouts and underlining are used in the following to show the proposed change in the current version in policy.

"When tagging or removing material for not having an inline citation, make it clear that you have a concern that the material is unverifiable may not have been published in a reliable source, i.e. may not be verifiable."

Note that the proposed change does not introduce the problem that Blueboar is concerned about. --Bob K31416 (talk) 17:32, 21 December 2012 (UTC)

True, I can't speak for Boar but there certainly should have been the sound of an axe being ground behind my last comment. Here, let me make it more relevant with this offer: I will support your revision if the words "you should" are also inserted before the words "make it clear." Best regards, TransporterMan (TALK) 17:40, 21 December 2012 (UTC)

Your suggested addition addresses another issue with the current version of policy, so I would prefer not to include it now because it might make it more difficult to reach a decision. However, I would prefer to consider it after a decision on the proposed version is reached. --Bob K31416 (talk) 18:05, 21 December 2012 (UTC)

This discussion has got bogged down in a number of tangents, so I think it is worth restating what the problem is (at least as I see it). This is not an issue to do with POV pushers and edit warriors (behaviour for which there are already policies and guidelines to deal with) but rather it is an issue with editors who are systematically going through the wiki tagging and/or deleting all uncited material without stopping to consider whether the may actually be problematic or if the deletions are making a nonsense of the articles. Such editors openly admit what they are doing and believe it is a useful task. They are not pretending that they object to the material to fulfil some ulterior goal, it is just a spree of pointless deletions justified by WP:V. This is bad from the point of view of collaboratively building an encyclopaedia. Why can we not just say that it is bad? I don't want to stop editors deleting uncited material they think is wrong, or even just don't like the look of, and I don't want to force them to do research they don't want to, I just want to stop the gratuitous deletions. SpinningSpark 21:14, 21 December 2012 (UTC)

Why can we not just say such deletions are bad?... because there is no consensus that they actually are bad. There are a lot of editors who honestly believe that removing potentially unverifiable material is in the best interest of the project (I am one of them... because I see such deletions as preventing the inadvertent addition of original research). The best way to stop "gratuitous deletions" is to find sources to support the material. Blueboar (talk) 22:03, 21 December 2012 (UTC)

They are bad if they are deleting good material and no one who cares is any longer watchlisting the article. They are bad if the editor persists in proceeding with the deletion, even after being provided with sources or having it pointed out that sources already in the article verify the material, merely because he cannot see a little blue number. Such deletions are pointy and disruptive. SpinningSpark 22:24, 21 December 2012 (UTC)

Um... hold on... you have suddenly shifted from objecting to the removal of unsourced material to objecting to the removal of sourced material. These are totally different situations. (it's not always bad to remove sourced information, but in that situation the burden is on the remover to explain the removal). Blueboar (talk) 23:11, 21 December 2012 (UTC)

Spinner, did you see the suggestion that I made buried towards the end of this tl;dr posting? Doesn't that deal with what you're trying to fix? I was opposed to your original suggestion because I thought the baby was being thrown out with the bathwater, but I don't think that we disagree about the bathwater. Best regards, TransporterMan (TALK) 22:02, 21 December 2012 (UTC)

Yes, that is exactly what I am opposed to. If accepted, that would deal with the issue I raised. SpinningSpark 22:24, 21 December 2012 (UTC)

There's also the problem that sometimes the "gratuitous deletions" go unnoticed, so there's none that can provide the available sources for verifiability even if those exists and they're willing to find them. I liked the version by TransporterMan that talked about the shared responsibility of all editors (a direct application of WP:CONSENSUS) to do what's best for the article. This includes not only trying to do the actions that you think are best, but also stepping into the shoes of editors with a different perspective and trying to facilitate their work, even if you don't fully agree with that view. In particular that, rather than trying to place blame and burden unto people, it would be great to speak of traceability and accountability. As long as one is acting in good faith to edit as they think is best, policy should make it clear that none should be blamed for it. But that doesn't mean that others shouldn't be able to review those actions and comment them. Diego (talk) 22:37, 21 December 2012 (UTC)

Perhaps we ought to have a RFP on the general principle: Is it, or should it be, a violation of Wikipedia policy subject to blocking or banning after appropriate warnings for an editor to chronically, continually, or regularly delete unsourced information over a range of articles, if the editor does not also frequently add inline sources to unsourced information rather than merely deleting it? Best regards, TransporterMan (TALK) 22:53, 21 December 2012 (UTC)

That is also problematic, as it focuses entirely on the removal, and not on the content being removed. I've done a fair amount of RC patrolling (though not as much recently), and you often run into a situation where an unsourced claim is added to an article by an IP editor that is not blatant vandalism, but that seems rather suspicious. The simple solution is often to remove the claim and demand a source, this doesn't require a direct suspension of AGF, and cleanly deals with the problem, if the claim is true, it should be verifiable, and if the claim is re-added without a citation, we can quickly move from a good faith WP:BURDEN revert, to treating the edits as disruption for ignoring BURDEN and or vandalism for adding false claims. The problem with your most recent idea is that RC patrollers are going to be mostly removing content, and at most occasionally citing things. Yet they are not removing content indiscriminately, and are really not a problem to be dealt with. Its tricky to find a way to target those who use BURDEN just because it exists, without also including constructive editors who use BURDEN as a way to deal with various types of suspicious or problematic edits. Monty₈₄₅ 23:02, 21 December 2012 (UTC)

Perhaps recently added material should be made an exception. SpinningSpark 23:25, 21 December 2012 (UTC)

I suspect different editors might define "recently" in different ways. Doniago (talk) 00:14, 22 December 2012 (UTC)

Yes, they might, but we don't want precise, hard and fast legislation here. As always, the rules need to be applied with common sense and in context. SpinningSpark 00:54, 22 December 2012 (UTC)

I am not sure why exceptions need to be made ... unsourced (and thus potentially unverifiable) is unsourced (and thus potentially unverifiable), no matter when it was added. Blueboar (talk) 01:03, 22 December 2012 (UTC)

Then lets say "chronic, continual, and indiscriminate removal of content" is the target problem. This will not include RC patrollers as they are not acting indiscriminately and "regular" has been droppeed, and it does not prevent anyone else removing anything fishy or dubious. SpinningSpark 01:20, 22 December 2012 (UTC)

Nope... For one thing, if some bit of material is unsourced (and thus potentially unverifiable), then removing it isn't indiscriminate. There is a very definite reason for removal. Blueboar (talk) 03:57, 22 December 2012 (UTC)

If we want policy to sanction removal of all unsourced material, no matter how unhelpful or foolish that action might be, then there will always be some editors willing to go round doing it. If that's the case then let's be done with it and write in policy that it is not permitted to add unsourced material in the first place. But that is not what we have in policy currently and it was clearly not the intention of those that drafted the policy to sanction such behaviour. We just need some clarity here. SpinningSpark 13:45, 22 December 2012 (UTC)

You know, I actually agree simultaneously with both SpinningSpark and Blueboar, as paradoxical as that may sound. And I think I have an idea why. We may be talking about two different kinds of editing behaviors here:

1. Sometimes, an editor will closely examine a page, or a group of related pages, and realize that there's a lot of content that might be questionable. Now there may be other editors at those pages who think that the content is just fine, but this one editor, rightly or wrongly, has concerns. That editor repeatedly tags or removes content, citing BURDEN.
2. Other times, we have editors who like to make large numbers of edits, edits that they perceive as housekeeping, and they can be extremely prolific. Often, they use automated tools. They make a couple hundred edits in a single day. Often, these are edits about categories or other gnomish things. Occasionally, such editors get interested in BURDEN. They go around making hundreds of deletions, based entirely on not seeing inline citations, and leave behind pages where other editors are annoyed, and other pages where some encyclopedic content was deleted because no one else was watching.

I think that (1) is a proper use of BURDEN, and it's what Blueboar is quite reasonably seeking to protect. I think that (2) can be a potentially disruptive misunderstanding of BURDEN, and it's what SpinningSpark is trying to make less easy to misunderstand. Am I correct about that? If so, we may be better able to come to agreement if we recognize the basis of what I think is a misunderstanding. --Tryptofish (talk) 23:11, 22 December 2012 (UTC)

I entirely agree with that assessment. SpinningSpark 00:12, 23 December 2012 (UTC)

I agree in part with (2) as it is essentialy bad manners to delete wholsale without taking some time to colaborate and that is the key to Wikipedia. However, I don't wish to see an absolute in the guideline here as that would be instruction creep and we need to make suggestions that will be held as a good standard but not as the only way to do things. The is true issue is not the deletion itself but the content. By our standards information should have an inline citation, however there are some smaller articles that use one or two over arching sources. When a whole sale deletion is made, it isn't really wrong to do so as there should still be an inline citation showing the page from the source that they claim is being taken from. The issue here is not so much about deleting, but more about the content being verifiable and how it is done. Soem larger articles do not use an overarching source. They are just a mish mash of different editors writing from memory. I think we need to emphasize the use of references more to combat the deletions. Not just trying to rework burdem as a policy. Although clearly we could use better wording.--Amadscientist (talk) 00:31, 23 December 2012 (UTC)

That's very good, thanks, both of you. If it should be the case that Blueboar also agrees, then I think we've made some progress. I think that Amadscientist is basically right, that we might be able to improve the wording a bit, but we need to be careful not to create an instruction creep that might be construed as prohibiting edits that ought not to be prohibited. --Tryptofish (talk) 15:14, 23 December 2012 (UTC)

I like the "may not have been published in a reliable source" language. "May not be easily verifiable" leads to people saying "Of course it's not easily verifiable: there's no inline citation to a FUTON bias source already right there! It's not easily verifiable if I have to do a web search all by myself, or get a book from the library!" WhatamIdoing (talk) 16:47, 23 December 2012 (UTC)

A reasonable challenge

The RfC that changed the wording of Burden (among other things such as repositioning sentences) was, in my opinion, closed before all the implications of the additional wording had been though through. The addition of ", and is satisfied by providing a reliable source that directly supports the material." to the sentence "The burden of evidence lies with the editor who adds or restores material." can be misunderstood if it is coupled to the next sentence "Attribute all quotations and any material challenged or likely to be challenged to a reliable, published source using an inline citation."

It seems to me that in this section and its subsections people are making the mistake of coupling the sentences together and drawing the conclusion that to meet "Burden" one always has to add inline citations, which is not what Burden says.

"Challenge" like all other part of policy must be qualified with "what is reasonable to the man on the Clapham omnibus?), and while it is satisfied by providing a citation to reliable source that directly supports the material, depending on the position or type of text is not the only way it can be satisfied. See my posting on at 09:38 on 14 September for more details of where it could be considered unreasonable to demand an inline citation for material.

To stop the current inference being drawn, one possible solution is to move Burden and Challenge into separate paragraphs -- as they were previously (both before and immidiatly after they were placed in separate sections (I thought the separate sections to be mistake)). -- PBS (talk) 15:31, 22 December 2012 (UTC)

How else would you meet your burden? Imagine that I type a somewhat surprising (to some people) fact, like "Some breast cancer organizations still promote breast self-examinations (BSEs) as an inexpensive means of advertising themselves and increasing breast cancer awareness, even though massive randomized clinical trials have demonstrated that BSEs do not save lives and overall cause medical harm to average-risk women." You challenge it. I could (easily, in this case) supply you with a stack of inline citations, but for some reason I don't want to. How else can I meet the BURDEN? WhatamIdoing (talk) 16:54, 23 December 2012 (UTC)

If the material has been challenged, and neither you nor any other editor can and is willing to provide inline citations, then I think you've just made the case for the material being kept out of the article...though perhaps it should be retained on the Talk page in the hopes that an editor will pick it up later. Doniago (talk) 17:01, 23 December 2012 (UTC)

WhatamIdoing you need to read the link to my previous post for other examples, but here is one from that posting: If the lead is constructed properly it is a summary of the information in the body of the article. Assuming that it is only a summary of the information in the body of the article and the body of the article is rigorously supported with in-line citations to reliable sources, then the summary in the lead meets Burden and usually challenging the text in the lead is not considered a reasonable challenge. There is nothing unusual about this example because it is what the MOS guidelines say (see Wikipedia:Lead section#Citations), and what is done in practice as usually only those facts in a lead that do not appear in the body of the article are supplied with citations (occasionally a particularly controversial statement in the lead my be given an in-line citation even if it is cited elsewhere but that an exception to the rule that the lead does not need citations). -- PBS (talk) 19:03, 23 December 2012 (UTC)

I agree that we've always accepted the sufficiency of providing one inline citation for a given bit of material in an article, even if the material appears dozens of times in an article (see Wikipedia:Citation overkill#Needless repetition of citations). But you seemed to be saying that there are ways of meeting BURDEN that involve no inline citations anywhere for the material, which IMO is wrong. WhatamIdoing (talk) 00:26, 24 December 2012 (UTC)

Removal of sourced material by claiming the source does not meet wp:reliable source criteria

[I refactored by creating this section after North8000's comment below of 16:42, 21 December 2012 (UTC). --Bob K31416 (talk) 17:45, 21 December 2012 (UTC)]

I don't know about what others think, but IMHO the most common problematic situation which merits a policy tweak is mis-using wp:burden in tandem with wp:rs to knock out unchallenged SOURCED material (e.g. the deletion edit summary just says "not a wp:rs")....to pursue POV'ing or pissing war objectives. I could give an IMMENSE number of examples of this, but each points a finger at someone, so I am just summarizing. I don't see where the above helps that. North8000 (talk) 11:12, 21 December 2012 (UTC)

AFAICT all of the very long discussion before your comment has been about the issue of uncited material. Perhaps we should stay the course and finish settling the uncited material issue before considering the unreliably cited material issue? --Bob K31416 (talk) 16:29, 21 December 2012 (UTC)

Well, North's comment does fit in to the overall topic... in that he is talking about situations where a) material is removed for lack of citation... b) returned with a citation (per BURDEN)... and c) removed again on the grounds that the citation was not wp:RS ... thus, arguably, not fulfilling the requirements of BURDEN (to return the material with a reliable source). I think his concern does need to be addressed... but as a second step... we first need to reach a consensus on parts a and b before we move on to addressing part c. Blueboar (talk) 16:54, 21 December 2012 (UTC)

Also where the unchallenged material is initially sourced and knocked out. North8000 (talk) 17:02, 21 December 2012 (UTC)

I agree, too, and definitely think it's worth talking about, but later. Best regards, TransporterMan (TALK) 16:57, 21 December 2012 (UTC)

I guess you're right. North8000 (talk) 16:42, 21 December 2012 (UTC)

One approach is to use templates such as {{self-published source}} {{Verify credibility}} (with the parameter failed=y set) and {{better source}}, in a similar manner to how {{citation needed}} is recommended in this policy.

I think that this approach has been quite successful in getting {{rayment}} references replaced with better alternative sources. There is a whole plethora of medieval biography articles based on various unreliable genealogy web sites, sometimes because the information is not available elsewhere, but also because it is easier to cut and past gather information for an article from these sources, than try to access books that are not commonly available. -- PBS (talk) 00:35, 22 December 2012 (UTC)

Moved from #Replacement of "is unverified" with "may not have been published in a reliable source ..." by BobK. I do, however, have a substantive concern. In the discussions above, North points out the problem of editors who try to POV-push by questioning whether or not the source is really reliable. I think this change in the wording risks making that problem worse, because someone could say, "I'm removing this content because it's sourced to an unreliable source, per WP:BURDEN." It's true that the sentence begins by specifying that it's about removing or tagging unsourced content, but I think the end of the sentence muddles the issue. WP:V is not WP:RS. I'd prefer to change the wording to: "make it clear that you have a concern that the material may not have been published in a verifiable source". --Tryptofish (talk) 23:26, 21 December 2012 (UTC)

Are you suggesting, though, that it is not acceptable to remove material that is cited only to an unreliable source? Formerip (talk) 01:00, 22 December 2012 (UTC)

It may be acceptable, but that case is definitely not under the scope of BURDEN, which is for content without a verifiable source at all (and thus to protect against original research). Removing such content requires discussing the reliability of the available source first. Diego (talk) 12:49, 22 December 2012 (UTC)

That doesn't seem very clear at all. The first sentence of BURDEN says "... is satisfied by providing a reliable source...". Formerip (talk) 14:20, 22 December 2012 (UTC)

Given that "reliable" is subjective, you can't perform a simple test for validity once the source is in place. So the "bright line" that BURDEN is supposed to be doesn't exist any more, once you have a source that verifiably backs up the text in the article. WP:BURDEN is satisfied at that point, at most you can have a case of WP:NOTRELIABLE; if the source is removed after talk then you have a case of BURDEN again. Diego (talk) 15:03, 22 December 2012 (UTC)

What makes you think that BURDEN is supposed to be a bright line? It doesn't claim to provide a bright line, and I can't think of a reason it ought to. Formerip (talk) 15:16, 22 December 2012 (UTC)

I didn't invent the term. This is how it was discussed while creating it and how it's being described by people using it at articles. We could say its the consensus on how to apply it. Diego (talk) 08:56, 23 December 2012 (UTC)

So maybe it's being misdescribed, given that the description would be inconsistent with the way it is actually worded? Formerip (talk) 16:08, 23 December 2012 (UTC)

Replying to FormerIP's original question to me, no, not really. For exactly the reasons that you and Diego just discussed, that opens a can of worms if we raise the issue in the sentence that I was discussing in the section below, before BobK split up and moved my comment. What I'm saying is that we used to (up until a day or two ago) have a sentence that did not get into the business of defining the word "unverifiable" in terms of whether or not the material had been published in a reliable source. I was not commenting in this discussion thread, but instead expressing a concern about BobK's edit, discussed below. I think it can be appropriate to remove material that is unsourced, and appropriate to remove material that is sourced to an unreliable source. I also recognize that there can be problems when people cite BURDEN to remove reliably sourced material because they claim that the source actually is unreliable. --Tryptofish (talk) 22:57, 22 December 2012 (UTC)

With your suggested change, are you trying to address the issue of removal of sourced material by claiming the source does not meet wp:reliable source criteria? --Bob K31416 (talk) 14:55, 24 December 2012 (UTC)

Yes. In other words, we know that BURDEN is satisfied by sourcing material to a reliable source. Unfortunately, some editors cite BURDEN incorrectly and unhelpfully by claiming that BURDEN allows them to remove material that is sourced, and that most editors think is reliably sourced, but the removing editor is claiming is a flawed source. There's nothing wrong with arguing about such an issue based on WP:RS. But drawing a line in the sand and saying that the material can never be added back without the approval of that one editor, the assumption of a veto power by that one editor, is not what BURDEN was intended for. That's the issue. Myself, I'm not looking for any kind of major revision to BURDEN based on this concern. Instead, I have a specific concern about the wording that was added by you and Amadscientist in the last few days, in that it unintentionally can make this problem worse. Because the thread below is where I wanted to discuss it, and is where it makes sense to discuss it, I'll explain, again, what specifically I mean about that, there. --Tryptofish (talk) 15:24, 24 December 2012 (UTC)

Replacement of "is unverifiable" with "may not have been published in a reliable source ..."

I implemented the following change.[22]

"When tagging or removing material for not having an inline citation, make it clear that you have a concern that the material is unverifiable may not have been published in a reliable source, i.e. may not be verifiable."

The discussion about it seems to have ended without any objections remaining, and gone off on other topics. BTW, if I thought there was anything contentious about the change, I wouldn't have implemented it. Also, this change doesn't preclude any other changes that are being discussed and is probably compatible with them. Thanks. --Bob K31416 (talk) 22:59, 21 December 2012 (UTC)

I just tweaked the wording, mainly to get away from the "ie". I hope that doesn't change anything substantive. [Moved the rest of this message and FormerIP's response, to the previous section above. --Bob K31416 (talk) 01:45, 22 December 2012 (UTC)]    --Tryptofish (talk) 23:26, 21 December 2012 (UTC) Noting that what was moved was an objection to the more substantive aspects of this change. --Tryptofish (talk) 22:29, 22 December 2012 (UTC)

I edited that a tad more for brevity and clarity. Did not change meaning.--Amadscientist (talk) 02:53, 22 December 2012 (UTC)

Here are the three versions for reference:

(1 2) "When tagging or removing material for not having an inline citation, make it clear that you have a concern that the material may not have been published in a reliable source, i.e. may not be verifiable."

(2 3) "When tagging or removing material for not having an inline citation, make it clear that you have a concern that the material may not have been published in a reliable source, and thus may not be verifiable."

(3 4) "When tagging or removing material not having an inline citation, make your concerns clear that the material might not be published in a reliable source, and therefore, may not be verifiable."

• I think the "i.e." is preferred because it says that the two phrases re being published in an RS and being verifiable, have the same meaning. Whereas "thus" and "therefore" only say that the first phrase implies the second.

• Also, the "for" in "for not having an inline citation" is necessary because it says that this is the reason for deleting.

So, I would like to restore version (1). Thoughts? --Bob K31416 (talk) 04:08, 22 December 2012 (UTC)

I have no objection to returning the "for", however I doubt the need. I do object to returning "i.e." as not needed.--Amadscientist (talk) 04:12, 22 December 2012 (UTC)

Do you recognize the difference in meaning between "i.e." and "therefore" that I mentioned? --Bob K31416 (talk) 04:23, 22 December 2012 (UTC)

"i.e.: "that is". Therefore: "for that reason".--Amadscientist (talk) 04:36, 22 December 2012 (UTC)

It is just my opinion that the use of an obscure Latin acronym is not entirely inline with the sentence meaning and may not be understood by the average editor.--Amadscientist (talk) 04:47, 22 December 2012 (UTC)

In other words, "i.e" would be acceptable if the next part of the sentence simply read "not verifiable", but since it reads "may not be verifiable" is not inline with the meaning of the term. We don't wish to make a claim here that is absolute and are attempting to state that something "might not be available....but could possibly be" Not that it is "absolutely" unavailable. The use of "i.e." is not inline with the sentence structure and takes the chance that another editor would wish the meaning of the sentence changed and not the acronym.--Amadscientist (talk) 05:05, 22 December 2012 (UTC)

Thanks, I'll give it a rest for now, except I restored "for". --Bob K31416 (talk) 05:49, 22 December 2012 (UTC)

OK.--Amadscientist (talk) 05:57, 22 December 2012 (UTC)

I wont change the text for now, but wanted to know if you thought that "When tagging or removing material for having no inline citation...." reads better?--Amadscientist (talk) 05:59, 22 December 2012 (UTC)

Although that is 3 letters shorter, it seems a little less readable, but I think I need to get away from this for awhile to recharge my judgement capacity. : ) --Bob K31416 (talk) 06:12, 22 December 2012 (UTC)

I think that is a great suggestion. I will do the same. Have a happy Holiday!--Amadscientist (talk) 06:26, 22 December 2012 (UTC)

I want to note that there is also the language that existed until this newest round of revisions began:

(1) "When tagging or removing material for not having an inline citation, make it clear that you have a concern that the material is unverifiable."

For the time being, I've reverted it back to that, because I don't really think that we've settled on anything new, and it's better to settle on something through careful discussion in this case. And frankly, no one has convinced me that any of the three proposed changes are improvements. First, I agree with BobK that "for" serves an important role. Second, I agree with Amadscientist that "ie" sounds clunky. Third, I agree with myself that we might not really want to put an "ie", "thus", or "therefore" into the sentence at all, per my comment that got moved into the section above. --Tryptofish (talk) 22:43, 22 December 2012 (UTC)

First, that really wasn't in the spirit of BRD. NEVER revert after such revisions and use BRD as your reasoning, that is just edit warring. Second, you wiped out all of the work we all have done and disrespected the changes that were in good faith. I will not add them back at this point. I will be looking at the suggestions below to see where the discussion is, but do not add you version in. There is no consensus. I am going to seek an admin to decide the closing of the above discussion and see what their thoughts are on what the consensus is for this discussion. You still may be right in returning it to the old version, just not in the way you did it. You should have addressed your concern here and allow a discussion if you felt there was something wrong.--Amadscientist (talk) 23:18, 22 December 2012 (UTC)

Please take a deep breath. That's simply not true. This is a core policy, and there was never consensus one way or the other. Nothing has been lost. --Tryptofish (talk) 23:25, 22 December 2012 (UTC)

And yet, oddly enough, you were fine until the other refinements. So which is it? No consensus or consensus for what you "like". Look, I am not battling over this. I just don't think you should have reverted after all those goodfaith attempts to refine the prose, even after you took part in it. It does "look" like you simply don't like it. Not that that is the case. We need someone who is experianced with these types of discussion becuase once it looks like a conensus was formed, the text is bodly added and a few refinements take place and noone objects, to simply revert does look like edit warring.--Amadscientist (talk) 23:42, 22 December 2012 (UTC)

I think we've worked this out on your talk page, and Jc37's talk page, but for the benefit of anyone else reading here I want to point out that I raised a concern in talk about the original edit, while collegially working with it by making what I hoped was a helpful edit to improve it slightly. My concern got buried by a refactoring, but that doesn't mean that I didn't express it. It would be a mistake to conflate a consensus on the fly between the two editors making the edits – you and BobK – with consensus amongst all the editors on this talk page, and it's generally a good idea to have clear consensus in talk before changing core policies, where the impact of changes can be much greater than that of edits to most other pages. --Tryptofish (talk) 15:08, 23 December 2012 (UTC)

Actually, I did express my original concerns a day ago, but I found today that my comments had been refactored, and not really addressed. --Tryptofish (talk) 23:39, 22 December 2012 (UTC)

Tryptofish, I was going to have a discussion with you regarding your new version and other matters, but I think that you are acting too disruptive for any such discussion to be worthwhile. You are reminding me of SlimVirgin. [Note added 15:43, 23 December 2012 (UTC): ‎Tryptofish moved his new version from above this comment of mine to below.] [--Bob K31416 (talk) 01:35, 23 December 2012 (UTC)

Bob K, that's not right and uncalled for. Regarding the series of events here, if I ever get a bunch of time to figure what the gist of all of the above is, I would weigh in and possibly I would disagree with Tryptofish. But to say such things about someone who has always been a calm and civil voice and makes an effort to respect process, IMHO those comments regarding Tryptofish are way over the top. Sincerely, North8000 (talk) 11:56, 23 December 2012 (UTC)

Actually, it was BobK's refactoring of my comment, a comment that pointed out a problem with BobK's edit, but when refactored made it sound like I had never raised any issues about Bob's edit, that set off the confusion to begin with. And Bob clearly is not familiar of my history with that other editor, nor understands that it's incivil to say that about her when she isn't here. --Tryptofish (talk) 15:08, 23 December 2012 (UTC)

Comment Re refactoring, I don't see it. Comment re SV not here, I made strikeout. --Bob K31416 (talk) 15:59, 23 December 2012 (UTC)

Here, very specifically, are my concerns about the edits that have been made to that particular sentence. After my revert yesterday, Bob made this edit: [23]. His edit summary claims that I "approved" an earlier version. Actually, I did make this edit: [24], in a good faith effort to work with what we had, but, very importantly, I also said this on talk: [25]. I stated very clearly that I had a substantive concern about the way that was going (even though I didn't revert any of it and tried to improve it slightly). Unfortunately, Bob refactored what I had written here: [26]. I trust the intent was good faith, but the effect was to make it sound like I had no concerns about Bob's own edit, and to relegate my expression of concern to a thread that otherwise had been about other language than that. I saw it many hours later, when discussion had continued and it would have been a mess to try to revert the refactoring.

If you look through the edit history on the policy page and the talk page, all of the "consensus" about changing that particular sentence was between Bob K and Amadscientist. Aside from me, no one else commented directly on those changes, although there were lots of comments expressing other concerns about the paragraph and sentence in question. So there was never a real consensus about those particular changes.

Since my revert, here are the further edits that have been made. Bob's edit, [27], also reverted an earlier edit, never discussed, that had changed the next sentence from "considering whether to remove or tag it" to the much simpler "removing or tagging it". Although not discussed, I think the simpler language was an improvement, and I'd rather see it come back now. Then WhatamIdoing made an edit that I think might have been good, [28], only to have it reverted [29]. I'm not clear what the rationale is for that revert to be OK, but my one revert to have caused so many negative comments directed at me. Then there was a further restoration of the discussed language, [30], whose amount of support is unclear: [31]. Again, I had raised this concern: [32]. I don't think that the wording that has been put back has engaged with the concern that I expressed. --Tryptofish (talk) 16:02, 23 December 2012 (UTC)

I'm concerned that underlying this comment of Tryptofish's is misunderstanding and misjudgement. I'll discuss each of Tryptofish's points one at a time, and wait for a reply to each before moving on to the next point and hopefully there can result in agreement. I'll start with one of the simpler points and wait for a response.

(1) Tryptofish wrote, "Then there was a further restoration of the discussed language, [33], whose amount of support is unclear: [34]." — This involves whether to use "thus" or "therefore". Tryptofish, do you have a preference between the two? Either is OK by me. Amadscientist stated in his Talk page comment that you could change "therefore" back to "thus" if you objected to the edit.[35] Tryptofish, What is your concern here? --Bob K31416 (talk) 16:58, 23 December 2012 (UTC)

Bob, if that's the best response you can come up with, I think you are in WP:IDIDNTHEARTHAT territory. My concern, one that you refactored, was not with whether we should say "thus" or say "therefore". It isn't a matter of one being better or worse than the other. It's a matter of where we have gotten to, once we are even considering such a sentence construction. --Tryptofish (talk) 20:23, 23 December 2012 (UTC)

You had a lot of points there with 10 diffs, but we can focus on just the refactored material if you like. So for reference, here's the original proposal along with your original message, which includes the refactored part that was after the first two sentences.

• "When tagging or removing material for not having an inline citation, make it clear that you have a concern that the material is unverifiable may not have been published in a reliable source, i.e. may not be verifiable."
• Tryptofish's comment — "I just tweaked the wording, mainly to get away from the 'ie'. I hope that doesn't change anything substantive. I do, however, have a substantive concern. In the discussions above, North points out the problem of editors who try to POV-push by questioning whether or not the source is really reliable. I think this change in the wording risks making that problem worse, because someone could say, 'I'm removing this content because it's sourced to an unreliable source, per WP:BURDEN.' It's true that the sentence begins by specifying that it's about removing or tagging unsourced content, but I think the end of the sentence muddles the issue. WP:V is not WP:RS. I'd prefer to change the wording to: 'make it clear that you have a concern that the material may not have been published in a verifiable source'."

• The proposal was an improvement over the phrase "is unverifiable". The improvement came from trying to satisfy Monty, who brought up the issue when he started the Talk page section. He was satisfied by the replacement of "is unverified" with "may not have been published in a reliable source, i.e. may not be verifiable". I implemented the change and wrote in the first message of this section that the implementation didn't preclude any other changes that are being discussed. For example the changes that are being discussed in North8000's section.

If you can find a further improvement by considering North8000's ideas, fine. So I put that part of your message in North8000's section where discussion regarding that could be kept together. BTW, it seems to me that there is a problem with your term "verifiable source", which I may go to North8000's section to discuss. But I don't think it is appropriate to discuss it here, instead of in North8000's section. Regards, --Bob K31416 (talk) 03:12, 24 December 2012 (UTC)

Indeed, that was what I wrote in this talk section. Obviously, you disagree with me. But I wrote it here because my concern was specifically that adding the language that you added – "may not have been published in a reliable source, i.e. may not be verifiable" – is language that can be wikilawyered too easily by an editor who is engaged in the kind of conduct first described by North, above. I'm not complaining about BURDEN in general. I'm concerned specifically about the phrase, "may not have been published in a reliable source, i.e. may not be verifiable". My concern remains if one changes "i.e" to either "thus" or "therefore" (although I consider either of those words to read a little more smoothly than "i.e." does). It opens up a situation where an editor, faced with material that has an inline citation, and the inline citation leads to a probably-reliable source that in fact does support the cited material, might nonetheless cite this sentence in BURDEN to say: "I can remove this material, and exert a unilateral veto power over ever restoring it, because you haven't persuaded me that the source is entirely reliable, and BURDEN allows me to do that, because I've expressed my concern that the source may not have been published in a reliable source, and as long as I've raised that doubt, BURDEN has defined that material as "therefore, not verifiable". All of the other language in BURDEN does not go to that extreme. This new language doesn't either, if one reads it in the sense that I believe you intended. However, the way the new language is worded, it can be construed in a manner that I don't think you intended, the way that I just described. I'm looking for policy language that does not lend itself to being misconstrued. --Tryptofish (talk) 15:39, 24 December 2012 (UTC)

For communication between us, maybe we first need to see if we have the same understanding of the meaning of the term "verifiable material". It is my understanding that "verifiable material" means that the material that has been published in a reliable source. --Bob K31416 (talk) 16:01, 24 December 2012 (UTC)

That's my understanding too, more or less. But here's the rub: you and I can agree about that and it still does us no good in dealing with that editor whose behavior I just described, because that editor, instead, is going to say that "verifiable material" is material where the source in the inline citation meets that editor's definition of "reliable". I'm not saying that the language you added is wrong. I'm saying that it has just enough ambiguity that it can be deliberately misconstrued. --Tryptofish (talk) 16:10, 24 December 2012 (UTC)

The change was from "a concern that the material is unverifiable" to "a concern that the material may not have been published in a reliable source, i.e. may not be verifiable." I don't see a change in the meaning. I think the original version was causing problems for some because they didn't have the same understanding of "verifiable material" that you and I have. So I simply clarified that. Are you suggesting that it is less subject to being deliberately misconstrued if it is less clear, as in the original version? That's the only difference between the two versions, clarity of the concept of verifiability. --Bob K31416 (talk) 16:23, 24 December 2012 (UTC)

No. --Tryptofish (talk) 16:28, 24 December 2012 (UTC)

Here, I just figured out specifically where the ambiguity is. The language "may not have been published in a reliable source" is intended, by you and by me, to mean the reliable source may not have been published. But it can be twisted to mean that it has been published in a source, but the source may not be reliable. Changing "not" to "never" removes that ambiguity, because no one will misconstrue it to mean that the source was "never" reliable. --Tryptofish (talk) 16:16, 24 December 2012 (UTC)

Re "The language "may not have been published in a reliable source" is intended, by you and by me, to mean the reliable source may not have been published. " — Is this what you wanted to write or did you accidentally put in some wrong words? --Bob K31416 (talk) 16:37, 24 December 2012 (UTC)

Hmmm. Unless I'm missing something, yes, it's what I meant to say. In other words, I think the intended meaning of the phrase is to mean that the material cannot be cited to a reliable source because such a source does not exist. That's a situation to which BURDEN is properly intended to apply. --Tryptofish (talk) 18:05, 24 December 2012 (UTC)

Your last message clears up what you understood for the intended meaning of "may not have been published in a reliable source". Although I don't see how the phrase can be misconstrued in the way you mentioned, I think we can still discuss your solution in a sense. Your solution appears to be to replace "may not" by "might never". Does that mean that the current version in policy would be acceptable to you if that replacement were made? --Bob K31416 (talk) 18:36, 24 December 2012 (UTC)

What I wrote here would solve it for me, yes. --Tryptofish (talk) 18:50, 24 December 2012 (UTC)

What would you like to do now? --Bob K31416 (talk) 22:03, 24 December 2012 (UTC)

 Done I did it. Peace! --Tryptofish (talk) 16:12, 25 December 2012 (UTC)

And now I'll put in (I meant put here in talk, not put on the policy page, sorry if that led to a misunderstanding. --Tryptofish (talk) 23:39, 22 December 2012 (UTC) a fifth option:

(5) "When tagging or removing material for not having an inline citation, make it clear that you have a concern that the material may not have been published in a verifiable source."

In my opinion, that approach addresses all of the concerns that have been raised about each of the other versions. --Tryptofish (talk) 22:48, 22 December 2012 (UTC)

Another approach would be:

(6) "When tagging or removing material for not having an inline citation, make it clear that you have a concern that the material might never have been published in a reliable source, and therefore may not be verifiable."

That would be less of a change from the recent edits (to which I still object) than #5 would be. But by changing "may not" to "might never have been", it places the emphasis, correctly, on whether sourcing might exist, instead of, misleadingly, on whether sourcing provided by other editors is or is not reliable. --Tryptofish (talk) 20:33, 23 December 2012 (UTC)

Your approach (6) consists of simply changing "may not" to "might never". Specifically here is the change you are suggesting in the current version of policy.

"When tagging or removing material for not having an inline citation, make it clear that you have a concern that the material may not might never have been published in a reliable source, and therefore may not be verifiable."

Either way is OK by me, if you accept this version for the policy, although my preference is for "may not" because it is a little shorter and a little simpler and there is a parallel construction with the phrase after it, "may not be verifiable". So what do you want to do? --Bob K31416 (talk) 22:21, 23 December 2012 (UTC)

I would prefer the language "may not" remain as is. The change seems to almost suggest "never" as having some importance. It seems to conflate or suggest that a source will never be written. It really seems les than contructive IMHO.--Amadscientist (talk) 03:25, 24 December 2012 (UTC)

"Less than constructive": I trust you mean "not an improvement", rather than "bad faith", right? Anyway, "never" actually does have some importance for WP:V. Verifiability is met once a supporting source is found to exist. If there's a sentence on a page without an inline cite, and there's a reliable source out there that supports it, the sentence is verifiable, just not cited. You can see what I said above, about how an editor misusing BURDEN can misconstrue the "may not have been published in a reliable source" language. The editor says, "the source may not be reliable". Now run that scenario again, with the "might never have been published in a reliable source" language. Now, when the editor argues about WP:RS, they have to keep it to WP:RS, because they can no longer hide behind BURDEN. It's not about whether one particular source is, or is not, reliable. It's about whether sourcing that satisfies WP:V exists. At all. If it exists, BURDEN's call to delete on sight and keep deleting on sight does not apply. --Tryptofish (talk) 15:53, 24 December 2012 (UTC)

Please see my message here. I thought there was an accidental mistake in writing your message there, but after reading the above, it looks like there may be some misunderstanding that needs to be cleared up. --Bob K31416 (talk) 17:38, 24 December 2012 (UTC)

The discussion above is about why there is a problem with saying "may not be published" or something like that, instead of "never". About Bob's edit reverting the next sentence back to "before considering removing or tagging", it's not a big deal to me either way; I just thought that it was simpler with the shorter wording, but I don't care that much. --Tryptofish (talk) 15:31, 26 December 2012 (UTC)

There is no consensus here for the "might never" inclusion. Even Bob stated "preference is for "may not". As you also that you don't care that much, I am returning the previous wording.--Amadscientist (talk) 23:51, 26 December 2012 (UTC)

You have completely misrepresented what I said, and frankly, you are showing some WP:OWNership issues with this policy page. I've been more than patient about this, but I have my limits. I said very clearly that I have substantive issues with this choice of words. How you went from that to my allegedly having said that it is no big deal, well, I can only conclude that you cannot be bothered to listen.

And what's even more inept about your revert is that the discussion has moved on to a different way of saying it, if you just bother to read what is below. In that context, your rationale becomes a moot point. --Tryptofish (talk) 15:37, 27 December 2012 (UTC)

Burden of evidence sentence "When tagging or removing material..."

"When tagging or removing material for not having an inline citation, make it clear that you have a concern that the material might never have been published in a reliable source, and therefore may not be verifiable." Sentence version in the Burden of evidence section when PBS made comment of 18:36, 25 December 2012 (UTC). --Bob K31416 (talk) 05:55, 26 December 2012 (UTC)

I think that this whole sentence, (from the time it was first discussed and then introduced into the policy) is instruction creep. I think the wording is too strident and if it is to remain on the page it ought to be qualified with "...inline citation, you ought to make it clear..." (italics to indicate my change or it could be "you may" or "it is considered best practice" or something similar) or we will have editors taking other editors to ANI because they consider other editors to be in breach of a policy point that is mandated and not advisory. -- PBS (talk) 18:36, 25 December 2012 (UTC)

Although I obviously don't consider it to be creep, I think that you are right about the tone of it. In fact, I think someone made such an edit, and it got lost amidst the other things that were going back and forth. I'm going to add "you should", and we can of course discuss it further if there are any concerns. --Tryptofish (talk) 19:38, 25 December 2012 (UTC)

I did not propose "you should" because over at WP:AT it became clear that should means different things in different dialects of English. Why do you prefer "you should" instead of "you ought to"? -- PBS (talk) 00:51, 26 December 2012 (UTC)

I'm not familiar with the discussion at AT, and I'm not aware of it being pointed out here. I actually didn't realize that there was an issue about it, and I seem to remember it having been in an earlier edit by someone else that was never really discussed in this talk. Of the various options, it seems to me to be simpler, but I'm not married to it. Of the other possibilities: "you ought to" seems kind of wordy without gaining anything from the extra wordiness; "you may" seems to me too tentative, and it's really not acceptable to me to have language that says that it's just something a few editors might want to do; the "best practice" formulations seem to me to have the same problems with tentativeness that "you may" does. An editor who habitually fails to engage in this so-called "best practice" isn't merely exerting editorial judgment; they are being disruptive. --Tryptofish (talk) 15:28, 26 December 2012 (UTC)

And if the choice of a phrase is an issue, I'm fine with going back to the much simpler wording that we had before, with no phrase about what "you" should do. --Tryptofish (talk) 15:33, 26 December 2012 (UTC)

And... some of this can be made clearer if we add that footnote at the end of the sentence. --Tryptofish (talk) 16:38, 26 December 2012 (UTC)

How about this: simply insert the word "please" where those other phrases would have gone. That way, we soften the tone, without opening up insoluble debates about whether one "should" or "may", etc. Thus: "..., please make it clear that you have a concern that..." --Tryptofish (talk) 18:18, 26 December 2012 (UTC)

For "should" see the ANI that stated it and the very long discussion starting at Wikipedia talk:Article titles/Archive 38#RM not required.

I do not think that "please" is as useful as "you ought to". -- PBS (talk) 21:45, 26 December 2012 (UTC)

Really? It's such a big deal to you that "please" is unacceptable? It boggles the mind! --Tryptofish (talk) 15:40, 27 December 2012 (UTC)

As I've said previously, I believe that the current context probably indicates that this is a "best practice" rather than a mandate, but I'm with PBS in feeling that ought to be made more explicit. I think "please" helps the construction that it's not mandatory, but is still ambiguous, and I don't know about the discussion at WP:AT but as a lawyer who speaks Texas (ahem) American English, at least, I consider "should" to be clearly permissive while "ought to" to be more ambiguous. Any of the three is better, however, than the current version. The better thing to do, as PBS suggests, is to just rid ourselves of the sentence. If we are going to consider the regular practice of removing uncited materials — i.e. the chronic failure to engage in best, rather than minimal, practices — a sanctionable offense then we ought to just say that rather than beating around the bush; if we're not, then we ought to just forget about it. As I suggested back at 22:53, 21 December 2012, maybe we ought to have an RFP to decide that principle, rather than just keep going around in circles about this language. Best regards, TransporterMan (TALK) 14:56, 28 December 2012 (UTC)

Further information, in light of those concerns: #Concern about appearing to deprecate "Burden". --Tryptofish (talk) 16:41, 1 January 2013 (UTC)

(Wasn't Texas planning to secede from the union??) Anyway, what the page says now (subject to whatever happens next...) is "you ought to", and I'm the one who put it there, on the theory that I might as well go along with PBS and now you, rather than make a big deal over it. Pending any proposal to delete the whole thing (which I will strongly oppose), I hope that works for the time being. Please! As for WP:RFP, believe me, that's a temptation! But I think you really meant WP:RFC, and I think that, or even another mediated process, may be what we end up with, and I'm OK with that. --Tryptofish (talk) 15:11, 28 December 2012 (UTC)

self-whack! #1 for not realizing the text had changed & #2 for the "RFP" (I'm batting 1000 this morning, rassin' frassin' ...). I'm fine for the nonce with "ought to". Sheepishly, TransporterMan (TALK) 15:28, 28 December 2012 (UTC)

From one fish to another, I'm glad this satisfies you (at least for now). --Tryptofish (talk) 15:35, 28 December 2012 (UTC)

I agree that (1) we are going round in circles, and (2) an RFC is a good way out. It will establish a firm consensus. I think the RFC should be aimed at establishing the principle rather than some precise wording, otherwise the RFC will get bogged down too. The exact wording can be thrashed out afterwards. I also think we should discuss the form of the RFC before launching it to avoid controversy during the course of it. SpinningSpark 21:29, 28 December 2012 (UTC)

How about this?

Hey boys and girls, how about this? Given the disagreements over "not published" and "never published", how about a different approach that avoids the mix-up entirely?

Change:

"When tagging or removing material for not having an inline citation, make it clear that you have a concern that the material might never have been published in a reliable source, and therefore may not be verifiable."

To:

"When tagging or removing material for not having an inline citation, make it clear that you have a concern that there may not be a reliable source for the material, and therefore the material may not be verifiable."

--Tryptofish (talk) 16:33, 26 December 2012 (UTC)

Slight change in what you suggest that adds "published":

"When tagging or removing material for not having an inline citation, make it clear that you have a concern that there may not be a published reliable source for the material, and therefore the material may not be verifiable."

--Bob K31416 (talk) 17:23, 26 December 2012 (UTC)

Yes, I think that's good. --Tryptofish (talk) 18:05, 26 December 2012 (UTC)

What would you like to do now? : ) --Bob K31416 (talk) 18:35, 26 December 2012 (UTC)

I suppose the honest answer is take a nap. (joke) Let's let it sit for a while, in case anyone wants to raise an objection, and if there are no screaming fits over this talk thread and the one just above it, we could change the sentence to:

"When tagging or removing material for not having an inline citation, please make it clear that you have a concern that there may not be a published reliable source for the material, and therefore the material may not be verifiable."

--Tryptofish (talk) 20:13, 26 December 2012 (UTC)

For now, it might be better to leave out "please", since you are currently having a discussion with PBS regarding that addition. Adding "please" could be a separate next proposal. --Bob K31416 (talk) 14:24, 27 December 2012 (UTC)

Yes, it comes as quite a surprise to me that "please" would even be an issue. I'm fine with just going along with "you ought to" instead, but I think we might as well go ahead with the rest of it now. --Tryptofish (talk) 15:42, 27 December 2012 (UTC)

Just so we understand what's happening

There is a lot of intense complex work going on here. So much that only about 4 devoted souls are really spending the time to fully keep up on what's happening. And then there are some folks who aren't that deep in but very cautions about a major policy change coming about from a "fast one" and tend to "oppose" to be safe to avoid that. May I suggest that the "4 folks" come up with something pretty agreeable to them and then cast a slightly wider net by extracting, distilling and identifying a proposal as having been vetted by such? I emphasized slightly to avoid making a huge burdensome (no pun intended :-)) deal out of it. North8000 (talk) 23:59, 26 December 2012 (UTC)

• Like--Amadscientist (talk) 02:36, 27 December 2012 (UTC)
• If you are suggesting that instead of editing policy with incremental changes, we edit it with one large change, no thanks. The present editing is beginning to move in an incremental cooperative way that should minimize concerns. --Bob K31416 (talk) 04:44, 27 December 2012 (UTC)
  • I am not at all sure that is what he meant. I thought it was more a suggestion to work together more on the above changes before implimentation. I for one don't mind that the burden sentence has been moved down, but thought it could use a little more discussion before it got done. I can't talk for North so, I'll let him make it clear if he wishes.--Amadscientist (talk) 05:49, 27 December 2012 (UTC)

Then we'll see if that's what he meant.

Re Burden sentence being moved down, it seemed premature since discussion had just started with someone that opposed it, and thus it is puzzling that it was implemented. --Bob K31416 (talk) 06:17, 27 December 2012 (UTC)

It really did seem premature, but I didn't feel a need to change it back. The section is called Burden of evidence...so I would prefer it at the top....but am frankly not all that outraged that it was moved. I guess you could say I am neutral on it.--Amadscientist (talk) 06:22, 27 December 2012 (UTC)

Well, that raises a question... whether we leave the BURDEN sentence at the bottom of the section, or return it to the top, the section (as a whole) isn't really about the "Burden of evidence"... its about "making and responding to challenges" (and specifically to challenges over unsourced information). So perhaps we should rename the section? Blueboar (talk) 13:21, 27 December 2012 (UTC)

Suggest leaving it at the bottom for now. The only opposer's last response was short and vague, so it's not clear if there is a meaningful reason for that editor's opposition. Let that recent change sit for a few days to see if there is any other opposition. If not, then suggest making a next proposal to rename the section. --Bob K31416 (talk) 14:04, 27 December 2012 (UTC)

Roughly I basically meant / suggested that the "4" full-timers work out a proposal and then clarify it and clearly ask the semi-active watchers here what they think, and give that a couple of days. North8000 (talk) 13:36, 27 December 2012 (UTC)

It sounds like you are asking for us to combine the various concerns that have been presented recently and draft a single, inclusive proposal... rather than work on them separately and incrementally. Is this correct? If so, I would be concerned that objections to one part of the "inclusive proposal" would mean the whole thing gets shot down. Thus none of the concerns would get addressed. Blueboar (talk) 13:54, 27 December 2012 (UTC)

I'm not sure this is a good idea, but perhaps it could be stipulated that if someone opposes part but not all of the proposal then they should draft a counterproposal that editors can weigh in on, to keep things moving forward? Doniago (talk) 14:15, 27 December 2012 (UTC)

In principle, this is a good idea, but it simply doesn't reflect reality. I've been trying to work with the other editors who are paying the closest attention, but I feel like I keep running up against a wall of either IDIDNTHEARTHAT or my-way-or-the-highway. I try to work with what others want, but the response is that I have to agree with everything or someone accuses me of disruption if I say that I disagree. I suggest "third way" solutions, and others engage in pissing wars instead of even reading what I wrote. There won't be an agreed-upon proposal from four editors. Maybe four different proposals, or maybe just me walking away in disgust. One good solution would be for those editors who have not been paying close attention, to start paying close attention. Another, although I hate to have to say it, is to decide that we are where we were with VnT, and we need to have a mediated process to resolve this. --Tryptofish (talk) 15:51, 27 December 2012 (UTC)

One reason that, for myself anyway, I prefer incremental changes is that while I'm certainly interested in what changes may occur here, I also feel like I'm looking at a wall of text where what the final result will be is very much unclear. I'm the sort of person who isn't especially comfortable talking with people he doesn't know especially well, so I tend to sit back and read what others are saying and consider their points, and only chime in if I feel I have something especially useful to say. Otherwise it seems to make more sense to wait until the group is zeroing in on final changes, and offer my opinion on those if I end up having an especially strong one. I guess I also hate the idea that I'll chime in on the conversation on day 3 of a 30 day discussion, and by the end nothing that I said will end up having mattered in any case. In that sense, kudos to the editors who are staying actively involved with the discussion for having more staying power for and investment in this sort of debate than I do. Doniago (talk) 16:13, 27 December 2012 (UTC)

I think that folks are giving me waaay too much credit for having/proposing a specific plan. I guess it was more as an observation. Taking myself for an example, lately I have been spending about 20 minutes a week (which is a lot for/on trying to watch one discussion) reading the happenings here and there is such a huge amount of proposals, edits, discussions that that isn't enough to "keep up" or understand what is going on well enough to weigh in on anything and so I haven't done so. North8000 (talk) 18:19, 27 December 2012 (UTC)

I don't know what to say. You are right that this has devolved into tl;dr. But that presents a Catch-22: if I try to engage in discussion, I cannot count upon a rational response, and if I make an edit, I get yelled at by people who didn't read what I already said in talk. I think I have a track record as someone who is pretty good at working with complex discussions of this nature, but this time, I'm about to just take the whole thing off my watchlist. I don't see any solution unless, either, more editors decide to do the heavy lifting of reading all the tl;dr, or, if we go to some sort of organized mediation. Failing that, the page is just going to drift. --Tryptofish (talk) 18:57, 27 December 2012 (UTC)

I'd almost say that what we need here is a firm "one proposal at a time" policy. Or at least something like "Here's the three proposals we're considering to address concern X. Please comment on them and/or add your own counter-proposal". The current discussion feels to me like a huge overlap of proposals that aren't entirely clear presented in a manner that makes it difficult to discuss, or even figure out roughly what level of support/opposition each has. I certainly understand where your frustration is coming from...nobody should be yelling at you though, and if that's the case then those people really need to review civility policies or chill out before they talk. I'd be happy to see us avoid mediation, but I think more organization and clarify would be beneficial. Could we just adopt a system of "(Clearly-written) Proposal X, Discussion of Proposal X, Proposal Y, Discussion of Proposal Y, etc....Straw Poll, Resolution"? Just trying to come up with ways to ease the frustration....— Preceding unsigned comment added by Doniago (talk • contribs)

I have to agree with Doniago here, the discussion on this page has been hijacked multiple times with particularly axes to grind. Let's grind one axe at a time. I would also like to thank Tryptofish for the effort he has put into consistently trying to find resolutions acceptable to all throughout the last week or two. SpinningSpark 20:11, 27 December 2012 (UTC)

Thanks, I really appreciate that! --Tryptofish (talk) 20:17, 27 December 2012 (UTC)

Attempted summary

In that spirit, let me do this, to (sort of) do what Doniago suggested. Here is a short summation of what I understand the current issues to be – with the understanding that other people might not agree with my summary.

1. Blueboar has proposed a reordering of the BURDEN section, independent of any wording changes. The discussion is here. The most recent edits to the policy page are here and here. I think the most recent talk question about it is here.

   • I have been bold and implemented the reordering. Blueboar (talk) 23:05, 27 December 2012 (UTC)

   Yes. What do you (and other people) think about this? --Tryptofish (talk) 14:38, 28 December 2012 (UTC)

2. There is a lot of discussion about one sentence in BURDENCHALLENGE. The most recent discussion of it is here. The corresponding edit is here.
3. About that sentence, there is still some discussion about this, mainly centering on whether to say "please" or "you ought to". (Not a big deal for me.)
4. About the next sentence, there's a bit of differing opinion about this. (Another not a big deal for me.)
5. Related to the much-discussed sentence, there was earlier discussion about a footnote to come at the end of it, discussion here. I'd prefer to let that sit for a while, and after dealing with the other stuff, I'll bring it up again in revised form.

--Tryptofish (talk) 20:34, 27 December 2012 (UTC)

@North8000 who wrote "May I suggest that the '4 folks' come up with something pretty agreeable to them and then cast a slightly wider net" if you lived thought the great WP:ATT incident then you will know that a small group of dedicated editors working together on a sub page is a dicey way of progressing. In the case of WP:ATT it ended in a train smash. In the case of WP:WTW a merge of several guidelines was constructed by an group working in isolation and then the change was forced through by edit warring it in. The problem is that that both groups work hard in good faith and as is human nature having delivered what they thought was an improvement, the members of the group were defensive to others who had not participated in the discussion and criticised their achievements. Those outside the development thought that the group were not heeding their concerns. So the process was divisive, and I do not think it is a suitable development method for this project. -- PBS (talk) 21:25, 27 December 2012 (UTC)

WP:att is probably was an immense undertaking and immense amount of a changes....something that is very difficult to do in Wikipedia. Probably not a good parallel/example to what I was discussing. Sincerely, North8000 (talk) 00:30, 30 December 2012 (UTC)

Looking for the cite

On occasion, I come across a claim that the info is not cited when what is actually true is that the cite is not where that challenging editor expects (it is inline before or after in the text). Because our "style" permits that not every sentence or part of a sentence be cited this seems unduly disruptive (same, actually when the reference is listed at the end of the article). If you are redoing the burden stuff, it seems this should be addressed. Alanscottwalker (talk) 16:47, 29 December 2012 (UTC)

Yes, that's a good point. I'm inclined to think that the best place to address it is when we go back to discussing #Revised footnote. --Tryptofish (talk) 19:16, 29 December 2012 (UTC)

It is a good point, but I think in order to implement it would require dropping the insistence that the person wishing to keep the material has to provide an inline cite. There is probably not going to be support for doing that, but in my opinion it is far more important to demand that enough information is provided to verify the fact (ie a reliable source, accessible in some public place, article title, page number etc). The inline cite business is a bit of a distraction; it really does not not matter how the source has been advertised, inline, end of paragraph, end of article, talk page, or even somewhere else altogether. The important point is that the fact is proven to be verifiable. Inline might be ideal, but verification is the important thing and what this policy should be concerning itself with. How cites are marked in the text is more a matter for guidelines than for here. SpinningSpark 21:49, 29 December 2012 (UTC)

Actually, in a format where anyone can edit anything at any time, the inline presentation of the source is practically essential to enable verifiability. if there is a source at the end of a paragraph at point A in time that verifies all the content in the paragraph, and then Editors Kim, Lee and Sam add additional content to the paragraph without providing sources, the citation at the end of the paragraph means bugger all to being able to verify the new content when time B rolls around. -- TRPoD aka The Red Pen of Doom 01:32, 2 January 2013 (UTC)

Back to the sentence at hand...

I have reverted Tryptofish's reversion as clunky, not suggesting a best route or practice but using the word "Ought" which simply means "must" or need", as well as redundant usage of "material" without need. If this is a requirement then just use that word....but it isn't. It is a suggestion as what is the best practice. Overuse of the term "material" reads very badly.

There was no consensus for the changes that were made and If there was any new discussion I could not find it in the massive wall of text. As suggested, let us deal with these issues one at a time.

The way it was written was changed against the consensus of the editors working on that section unilaterally even after editors expressed concerns about the changes against the consensus that was formed. Before we can begin deciding on even more changes we really should stabilize the page with a consensus on the last alterations.

Tryptofish's version:

"When tagging or removing material for not having an inline citation, you ought to make it clear that you have a concern that there may not be a published reliable source for the material, and therefore the material may not be verifiable."

What I reverted back to is:

"When tagging or removing material for not having an inline citation, it is always best to make it clear that you have a concern that there may not be a published, reliable source for the content, and therefore the information may not be verifiable."

Shouldwe use completely different wording or is there a consensus that can be formed for one of the versions above?--Amadscientist (talk) 08:54, 1 January 2013 (UTC)

I am getting a little tired of the policy wording constantly shifting while it is being debated. This is not good and I have reverted. This wording is very recent and just because you think it is better is not a good reason to try and edit war it in. Admittedly, the version I have reverted to also includes other recent changes. If that is a concern then revert back further to a stable version before this discusssion started such as this, but all this to-and-fro is not acceptable. For the record, I do think editors damn well ought to have a concern before deleting anything. SpinningSpark 11:26, 1 January 2013 (UTC)

I don't think you meant "stable", since the fact that it changed means that it was not stable. Perhaps the criterion should be the last version that had consensus before changes were considered. This does not mean any past version that had consensus, but rather the latest one. We wouldn't want to remove improvements that had consensus.

Regarding the present version with "you ought", it was discussed on this talk page and seemed to be acceptable among those editors who were discussing it. If there is a change made, I don't think it should be a change to a version that hasn't been discussed at all on the Talk page. Any such new version needs to be first discussed on this talk page to get consensus or not encounter objections. We should try to make the process as orderly as possible. --Bob K31416 (talk) 13:08, 1 January 2013 (UTC)

I don't know if this will help or not, but perhaps we are at the point where there is nothing to be gained by arguing about what is or isn't the current consensus, or arguing about process. Those things have pretty much spun out of control, sadly. What's most important is what is best for the policy page to say, on the merits. Not who thought of it, and not how the sausage was made.

Amadscientist expressed concern that my edit summary here was unclear, so I will try to explain it in more detail.

(1) The latter part of the sentence – "there may not be a published reliable source for the material, and therefore the material may not be verifiable." – seems to me to draw a parallel. The part after the comma explains the implications of the part before the comma. Consequently, repeating the word "material" (although I can see how there's an argument that the repetition might be "clunky") helps the reader understand the parallel.

(2) There was a considerable amount of discussion of "ought" versus alternative wordings here. For the moment, it seems to me that the discussion had settled on "ought" in preference to other possible wordings. As I said, I was surprised that some other editors feel as strongly as they do about which wording to use, but it seems to me that any change from "ought" is going to be controversial.

--Tryptofish (talk) 16:18, 1 January 2013 (UTC)

The last version with consensus was:

"When tagging or removing material for not having an inline citation, make it clear that you have a concern that there may not be a published reliable source for the material, and therefore the material may not be verifiable."

--Amadscientist (talk) 22:52, 1 January 2013 (UTC)

The addition of ought does not actually have a full consensus and this is a challenge to to that wording.--Amadscientist (talk) 23:02, 1 January 2013 (UTC)

(edit conflict) Yes, and I liked that version (although I no longer have any idea what did or did not really have consensus). But there were a lot of complaints that "make it clear" was too bossy, or something like that. And "please make it clear" wasn't bossy enough, or something like that. And now that you've insisted on moving it out of the discussion where it followed from, I trust you can figure out for yourself where those objections were. --Tryptofish (talk) 23:04, 1 January 2013 (UTC)

Yes, and I am sure those with any standing disagreements can find their way to this discussion. But to be fair to the "ought" wording and any other concerns an RFC should begin with a proposal written as neutrally as possible. Perhaps we can simply add the varying versions with A, B, C and allow a straw poll to determine this. It is imprtant that the wording be written as precisely and clearly as possible.--Amadscientist (talk) 00:39, 2 January 2013 (UTC)

I think if we are going to hold an RfC it would be better to hold an RfC on the principle involved. From my POV, this is whether or not mechanical tagging and deletion is considered desirable or disruptive (note: not a suggested wording). An RfC on a specific wording is liable to generate a wall of text, multiple alternative suggestions, and will get bogged down in argument. It will be next to impossible for anyone to close with consensus. On the other hand, establishing the view of the community on the principle first will make it a lot easier to arrive at an exact form of words, and will be probably be far less controversial. SpinningSpark 23:34, 1 January 2013 (UTC)

(ec) Last version with consensus Really? Perhaps you could provide a link to where that consensus was reached. Are you sure you don't mean the last version you agreed with? Also please provide a diff to that version. I have searched back as far as September with Wikiblame and cannot find a version with that exact text (although I am sure you are right that there was such version, but what you have pasted can't be quite exact.) SpinningSpark 23:17, 1 January 2013 (UTC)

Sorry, but if you want to wade through the discussion have at it. Clearly there is NO consensus now, but the wording of Ought was not agreed on by everyone and is not a true consensus. As I said an RFC is needed. If there was a consenus there would be no need for an RFC.--Amadscientist (talk) 00:43, 2 January 2013 (UTC)

I am not opposing an RfC (but see my comment above), nor do I think I was making a claim there was consensus. You were the one who used the c word and if you wish to stand by that statement it is down to you to PROVEIT. SpinningSpark 00:53, 2 January 2013 (UTC)

No one need prove anything on Wikipedia, but simply demonstrate what is necessary. That really isn't necessay at this point. A consensus did exist for a short period except for precise wording. Now we have no consensus and at least one comment suggested we lose the line altogether. When looking through the past year, the entire first two section have been very unstable but that line began to get some agreemant and now it looks a little "Frankensteined" -all stitched together with little form.--Amadscientist (talk) 01:15, 2 January 2013 (UTC)

Re Amadscientist's comment "Frankensteined" — It's alive! I suppose some could say Frankensteined for Wikipedia in general, where the contributions of different editors are stitched together. --Bob K31416 (talk) 02:45, 2 January 2013 (UTC)

True, but generally we try to make it look like purposefully written text. Redudant wording and structure are usually addressed. Give me a bit...I'm working in my lab on something. Be back in a short while. *Get back to work Igor. what?...no, I don't care if your name is "Fritz"...get to work you hunched back assistant!*--Amadscientist (talk) 03:05, 2 January 2013 (UTC)

Then Dr. Frankenstein's mistake was not collaborating with a cosmetic surgeon. It would be nice if there were in Wikipedia, board certified cosmetic surgeons for writing style that could be called upon for help. I would ask such a doctor if anything could be done with the phrase, "make it clear that you have a concern that there may not be". --Bob K31416 (talk) 03:22, 2 January 2013 (UTC)

Funny thing...we do. They are called "Copy editors" We have just such doctors with their own Guild.--Amadscientist (talk) 04:20, 2 January 2013 (UTC)

They don't look "board certified" but maybe they would be useful anyhow. --Bob K31416 (talk) 16:46, 2 January 2013 (UTC)

I discussed it with someone. Would this work?

"When tagging or removing material for not having an inline citation, consider mentioning that there may not be a published reliable source for the material, and therefore the material may not be verifiable."

--Bob K31416 (talk) 04:18, 2 January 2013 (UTC)

Still doesn't need the redundant wording. I will add this to my RFC, with a slight copy edit replacing the first and last "material" with "content" and "information". If you don't approve of the edit, feel free to adjust your suggestion once it goes up. Igor is almost finished securing the kites at the top of the tower.--Amadscientist (talk) 05:04, 2 January 2013 (UTC)

I don't approve of the edit, but I also don't approve of the form of the RfC. So I'll try to continue the discussion with you here. Regarding repetition of "material", would this work?

"When tagging or removing material for not having an inline citation, consider mentioning that there may not be a published reliable source for the content, and therefore it may not be verifiable."

--Bob K31416 (talk) 15:03, 2 January 2013 (UTC)

First sentence of lead

The first sentence of the lead,

"In Wikipedia, verifiability means that people reading and editing the encyclopedia can check that the information comes from a reliable source"

does not express the meaning of "verifiability" in Wikipedia, i.e. that the material has been published in a reliable source. Instead it expresses an effect of verifiability. Suggest changing it to,

"In Wikipedia, verifiability means that the information in the encyclopedia has been previously published in a reliable source."

--Bob K31416 (talk) 07:26, 6 January 2013 (UTC)

Oh god.......not again. We just changed that portion and now you want to re-word it yet again. I beg you to do one thing. Check the archives and notify all those involved in the last discussion before implementing any change to the lede. Please.--Amadscientist (talk) 11:33, 6 January 2013 (UTC)

My motivation for the suggestion is that the term "verifiability" is not defined in the first paragraph, which only gives the desired effects of verifiabilty without defining it. Because of that, editors may tend to fill in the vacuum by each having their own definition of verifiability, which can be the cause of not understanding each other when discussing verifiability. I've seen that happen on this talk page, for example.

How many of you know that in Wikipedia, verifiability means that the information in the encyclopedia has been previously published in a reliable source, and that this term means no more and no less than this. The present first sentence is not a definition, it is an effect of verifiability. --Bob K31416 (talk) 15:46, 6 January 2013 (UTC)

You are correct that the first sentence does not define "Verifiability"... but I don't think it should. That's what the rest of the policy page is for. Our concept of "Verifiability" is very complex and nuanced. I don't think it can or should be summed up in one sentence. It needs an entire page to define and explain it. We don't want editors to read one sentence of the policy... we want them to read all of it. Blueboar (talk) 16:51, 6 January 2013 (UTC)

If you don't think that what I gave is the definition of verifiability, please give an example of verifiability that the definition doesn't cover. A hypothetical example would be fine. Thanks. --Bob K31416 (talk) 21:50, 6 January 2013 (UTC)

Please add me to the list of people who have low enthusiasm for reopening discussion of the lead. I like the way that the sentence draws attention to what the policy means in terms of what it helps readers to do. --Tryptofish (talk) 23:38, 6 January 2013 (UTC)

Another danger

The general argument being put here is that any content which is unsourced reduces the reliability of WP and that we should therefore consider removing it. The problem with doing this too enthusiastically is that it promotes another form of unreliability, which is what I would describe as a cut-and-paste mentality. Good but unsourced material that is deleted may be replaced with bad but sourced material.

In many subject areas it is easy for a person with no background knowledge of the subject to Google the subject and then cut and past snippets from online sources into a a WP page. To a non-expert in the subject this may give the illusion of reliable content but to an expert it reads as word salad. Sometimes this is done in good faith at other times with some ulterior motive, often by people who have some gripe or objection to an organisation.

I am not, by any means, objecting to our requirement for material to be verifiable, I am just pointing out that, even with the most worthy principles, it is possible to go too far. Martin Hogbin (talk) 10:11, 6 January 2013 (UTC)

Remember that Wikipedia is a work in progress, and anyone can fix problems... when you come across an article that is simply a word salad of reliably sourced snippets... fix the problem by re-writing it. Blueboar (talk) 15:20, 6 January 2013 (UTC)

I do, when I can. But I may run into the original editors, who have deleted the well written and correct version, because it has no sources and object to their, sourced, version being changed. Martin Hogbin (talk) 19:29, 6 January 2013 (UTC)

How should the "Tagging or Removing" prose in Burden be written?

A dispute has arisen in regards to the wording in burden regarding the tagging or removing of content. This request for comment is to establish the specific wording for just that part of the "Burden of evidence" section of our "Verifiability" policy.--Amadscientist (talk) 05:44, 2 January 2013 (UTC)

(Disclaimer:There is a small portion of the prose that has survived through the various changes. For clarity I am retaining such prose, in colored green text. Blue text, as always, is a wikilink. Versions are bolded for ease of reading (but might be changed to italics if others disapprove).

Background

Versions of the text over the last year, from January 5TH to December 17TH

The prose for this line has undergone a number of changes over the last year. January[36]: "Whether and how quickly this should happen depends on the material and the overall state of the article. Editors might object if you remove material without giving them time to provide references. It has always been good practice to try to find and cite supporting sources yourself." By the end of Febuary[37] additional prose was added: Whether and how quickly removal should happen depends on the material and the overall state of the article; consider adding a citation needed tag as an interim step. Editors might object if you remove material without giving them time to provide references. It has always been good practice to try to find and cite supporting sources yourself. This appears to have remained stable through the month of March[38], through April, May and June[39]. The text in Agugust had a single alteration that added the word "but" to between to exsiting parts of the text creating a longer sentence but not effecting the section in question[40]. From the last day of August to the end of September the only change was the removal of the word "but" and seperating senstences again[41]. In the month of October a more substantial change was made[42]: Whether and how quickly this should happen depends on the material and the overall state of the article. Editors might object if you remove material without giving them time to provide references; consider adding a citation needed tag as an interim step. When tagging or removing material for being uncited or insufficiently sourced, please indicate that you have a concern that the material itself is unverifiable. If you believe the material to be verifiable, then it is better to provide sourcing yourself, instead of removing the material. It has always been good practice to make reasonable efforts to find sources yourself that support such material, and cite them. It appears that the month of November is when the section became contentious. A succession of changes began on this particluar part of the Burden policy. From the 1ST to the 7TH alteration ended with this temporary version[43]: Whether and how quickly this should happen depends on the material and the overall state of the article. Editors might object if you remove material without giving them time to provide references; consider adding a citation needed tag as an interim step. When tagging or removing material for being uncited or insufficiently sourced, please indicate that you have a concern that the material itself is unverifiable. If you believe the material to be verifiable, try to provide sourcing yourself. If unsuccessful, it is best practice to leave the material with a tag (which alerts other editors to the fact that the material needs a citation) rather than to remove it. The last portion was tweaked to[44]: "If unsuccessful, it may be either tagged or removed." This was a challenge to the validity of "Best practice" in regards to tagging. This was reverted, reverted back then reverted again. Then the entire portion was returned to the October version before tagging was introduced as "Best practice"[45]. After that, a restoration was made of all the content not in dispute[46]. After that a small amount of miner changes were made leading to: Whether and how quickly this should happen depends on the material and the overall state of the article. Editors might object if you remove material without giving them time to provide references; consider adding a citation needed tag as an interim step. When tagging or removing material for being uncited or insufficiently sourced, please indicate that you have a concern that the material itself is unverifiable. If you think the material is verifiable, then it is better to provide sourcing yourself than to remove the material. It has always been good practice to make reasonable efforts to find sources yourself that support such material, and cite them. Further tweaks were made leading to[47]: Whether and how quickly this should happen depends on the material and the overall state of the article. Editors might object if you remove material without giving them time to provide references; consider adding a citation needed tag as an interim step. When tagging or removing material on these grounds, make it clear that you have a concern that the material itself is unverifiable. If instead you think the material is verifiable, then it is better to provide sourcing yourself than to remove the material. It has always been good practice to try to find sources yourself that support such material, and cite them Which saw a flurry of further revisions resulting in[48]: Whether and how quickly this should happen depends on the material and the overall state of the article. Editors might object if you remove material without giving them time to provide references; consider adding a citation needed tag as an interim step. When tagging or removing material on these grounds, make it clear that you have a concern that the material itself is unverifiable. If instead you think the material is verifiable, it is better to try to provide an inline citation yourself than to just remove the material. If unsuccessful, then consider whether or not to remove or tag it. By the end of November a further change to the last line was made [49]: If instead you think the material is verifiable, it is better to try to provide an inline citation yourself before considering whether to remove or tag it. The line has continued to be reverted, revised and tweaked. On Dec 17 a dicussion began to address concerns on the unstable prose but continued to produce revisons and reverts.

From the discussion beginning on December 17TH, the line began to see some agreement but stability of the prose and a true consensus has eluded the page.

The revisions being considered:

• (A)

  Whether and how quickly this should happen depends on the material and the overall state of the article. Editors might object if you remove material without giving them time to provide references; consider adding a citation needed tag as an interim step. When tagging or removing material for not having an inline citation, you ought to make it clear that you have a concern that there may not be a published reliable source for the material, and therefore the material may not be verifiable. If instead you think the material is verifiable, it is better to try to provide an inline citation yourself before considering whether to remove or tag it.

• (B)

  Whether and how quickly this should happen depends on the material and the overall state of the article. Editors might object if you remove material without giving them time to provide references; consider adding a citation needed tag as an interim step. When tagging or removing content for not having an inline citation, editors should try to make it clear that they have a concern that there may not be a published, reliable source for the claim, and therefore the information may not be verifiable. If instead you think it is possible to be verified, it is better to try to provide an inline citation yourself before considering whether to remove or tag such material.

• (C)

  Whether and how quickly this should happen depends on the material and the overall state of the article. Editors might object if you remove material without giving them time to provide references; consider adding a citation needed tag as an interim step. When tagging or removing content for not having an inline citation, consider mentioning that there may not be a published reliable source for the claim, and therefore the information may not be verifiable. If instead you think it is possible to be verified, it is better to try to provide an inline citation yourself before considering whether to remove or tag such material.

• (D) (added from discussion by User:Qwyrxian)

  Whether and how quickly this should happen depends on the material and the overall state of the article. In some cases, adding a citation needed tag to give others time to provide references may be a good interim step, but information which is in any way contentious or likely to be challenged can be removed. If unsourced information is removed, be sure to provide a clear explanation in either an edit summary or on the article's talk page.

Editors should feel free to add their own revisions if they feel inclined in the discussion section or !Vote for one of the above examples. This is an attempt to find a consensus on this portion of the Burden policy only.

Survey

• I object to the RfC as written. The history and selection of options above do not actually reflect what has been discussed on this page. Those three options include wordings, presented as co-equal options, that have only appeared on this talk page in the last 12 hours or so. --Tryptofish (talk) 15:35, 2 January 2013 (UTC)
• Oppose all three options shown above. There are other options being discussed that I think would be superior to any of those three. --Tryptofish (talk) 15:38, 2 January 2013 (UTC)
• I prefer Option D:

  Any material lacking a reliable source directly supporting it may be removed. Whether and how quickly this should happen depends on the material and the overall state of the article. Editors might object if you remove material without giving them time to provide references; consider adding a citation needed tag as an interim step.[2] When tagging or removing material for not having an inline citation, please make it clear that you have a concern that there may not be a published reliable source for the material, and therefore the material may not be verifiable. If instead you think the material is verifiable, it is better to try to provide an inline citation yourself before removing or tagging it.

  --Tryptofish (talk) 15:54, 2 January 2013 (UTC)

• Oppose all three versions, none of them truly address the substantive issue (disruptive deletions), and none of them bear any resemblance to the suggested text I made at the beginning of this discussion. And oppose the whole concept of this ill-conceived RfC. SpinningSpark 19:39, 2 January 2013 (UTC)
• Weak D. I don't think any of these is good enough to use, but D is the closest and I'm in favor of using it as a starting point for further improvements. Andrew³²⁷ 18:53, 3 January 2013 (UTC)

Threaded discussion

• I don't see any real substantive difference between A and B. C, by contrast, minimizes the principle that we do not normally remove unsourced content from articles entirely unless we believe it to be unverifiable. In other words, it is not appropriate to simply remove any and all content for no other reason than that it currently lacks citation within the article, and when we do remove content we should be prepared to explain that we thought it was unverifiable and why. This is represented in policy elsewhere at WP:PRESERVE, which states that we fix what is fixable rather than deleting it, and even WP:BLPDEL limits outright deletion (of entire articles or article portions) to "contentious" material, obviously less than simply anything regarding a living person), urging improvement unless that's not possible (see also WP:BEFORE). And all of this is not simply just a bunch of words, but a deep necessity of this being a work in progress, open always to new editors who obviously aren't going to be familiar with citation templates, manuals of style, etc., and they and their contributions should be valued and treated with the assumption of good faith absent evidence to the contrary. I haven't followed the recent edit disputes on this page so maybe I'm overthinking this RFC, but that to me seems like what the language differences are about. postdlf (talk) 06:10, 2 January 2013 (UTC)

The differences between A and B are: "you ought to make it clear" changed to "editors should try to make it clear", as well as changing the redundant over usage of the word "material" (used 4 times) to "content", "claim" and "information" with "material" used once at the end. Also "If instead you think the material is verifiable," was changed to "If instead you think it is possible to be verified,". Version C has this last change in that version as well but could easily be mixed and matched where consensus agrees.--Amadscientist (talk) 06:53, 2 January 2013 (UTC)

• B has a grammatical error in it ("editors should to make it clear"), maybe that's supposed to be "try to"? I'd tend toward B myself, as it's a good idea in general to state why you're doing something, but it's never been a requirement per se. Seraphimblade ^{Talk to me} 06:16, 2 January 2013 (UTC)

Typo fixed. Thanks. Good eyes. Mine were getting a tad tired after reading through the last year of revisions.--Amadscientist (talk) 06:42, 2 January 2013 (UTC)

But there is a requirement to explain your actions for controversial edits, it's the essence of WP:CONSENSUS and it's explicitly listed as a requirement at WP:Discussion (at least twice, in "Follow the normal protocol" and "Discuss with the other party"). Any removal that is not for illegal content, vandalism or nonsense is controversial (since at least one editor thought it belongs at Wikipedia) and should be either tagged for some time before removing, or removed with an explanation. Diego (talk) 10:50, 2 January 2013 (UTC)

There is no requirement to discuss when removing or tagging unsourced or poorly sourced information. In fact the link for WP:Discussion merely goes to Wikipedia:Dispute Resolution where the issue is simply about discussing when there is an active dispute. A challenge itself (removing or tagging) is not a dispute. A dispute arises when someone disagrees with those choices. Even then, what that says is: "Talking to other parties is not a formality; it's imperative to the smooth running of any community.". What that means is that it is not to be avoided and is necessary. But that still is not an actual requirement. As a DR/N volunteer I have seen a number of disputes go un-discussed by participants for many reasons. We can't force them. But, I could live with just saying "Should make it clear" though that really isn't stating that doing so is a requirement. A purposeful wording that was intended.--Amadscientist (talk) 11:53, 2 January 2013 (UTC)

I don't see the distinction you're making between it's imperative and not to be avoided and it's a requirement; for me those are the same things. For me, the point to clarify is that one can't just silently remove unsourced content; tagging it is OK as it's enough to explain the reasons, but a plain removal without comment is not. Diego (talk) 12:08, 2 January 2013 (UTC)

As is said to me time and time again, "the wording is poor" (that is what this RFC is about) but it simply does not say it is a requirement or something we can force on editors. Heck...if I could, I would make every editor listed at DR/N discuss the issues, but we simply cannot do it and there is no policy or guideline that makes it so.--Amadscientist (talk) 12:21, 2 January 2013 (UTC)

Are you talking now about the wording of WP:BURDEN or the wording of WP:DR? Because I definitely read WP:DR as requiring editors to explain their actions. Even if BURDEN didn't include any explicit requirement to discuss, DR does. We cannot require any level of commitment to the project from editors, but we can definitely expect them to explain the actions they've already performed if they want them to be part of consensus building. Diego (talk) 14:49, 2 January 2013 (UTC)

As I said, it is a DR policy, not an editing policy. There is a difference. And it does not require discussion. There are very few brightline rules on Wikipedia and discussion is not one of them when simply editing, adding tages or removing content. One is not even required to make an edit summary as some tools do not give that option. One man's vandal is anothers confused editor.--Amadscientist (talk) 23:07, 2 January 2013 (UTC)

• I think I prefer the wording of B ("have a concern that there may not be a published, reliable source for the claim" better than "the material may not be verifiable"), minus the "try to make it clear...". We should not imply that you can skip explaining the reasons for removing unsourced content (instead of tagging or sourcing it as preferred per WP:PRESERVE), because explaining one's controversial actions is not optional (at most it may be deferred until someone else disputes that action). Diego (talk) 10:57, 2 January 2013 (UTC)
• Possible suggestion What do others think of adding something similar to this?

If the uncited material in question is easily verifiable through a simple search of reliable sources Google news, Google books, etc., it should not be removed.

Would this be helpful or not? 64.40.54.118 (talk) 13:08, 2 January 2013 (UTC)

Google News and Google Books are not themselves reliable sources, but they might help you find reliable sources. Jc3s5h (talk) 13:27, 2 January 2013 (UTC)

Poor wording choice on my part. I meant the searching action was of reliable sources, as opposed to a general search of the web, which may or may not find reliable sources. 64.40.54.118 (talk) 13:53, 2 January 2013 (UTC)

Was this ever published anywhere besides this page prior to the most recent note at WP:VPP? If not, then this is very saddening and the entire discussion should be considered suspect; if it was, I'm sad I missed it. I don't have time to read all of the above page, but what I've read so far doesn't even seem to support any of the versions above. The proposal mentions other suggestions, so here's one:

Whether and how quickly this should happen depends on the material and the overall state of the article. In some cases, adding a citation needed tag to give others time to provide references may be a good interim step, but information which is in any way contentious or likely to be challenged can be removed. If unsourced information is removed, be sure to provide a clear explanation in either an edit summary or on the article's talk page.

In other words, this policy should say what it's always said: if you add something, provide a source, and if you don't, your information can and should be removed if any other editor objects. Anything less than this supports puffery ("City X has more doctorates than any other city in Country Y"), promotional editing ("X is the leading brand of soft drinks in Country Y"), flat out lies added in bad faith ("Hey, assume good faith, it's true, I read it somewhere, just give me a few months to remember where that was"), and other crap. Don't give bad faith editors and editors who simply don't understand our underlying principles a wordy tool they can use to keep pages in poor condition, even temporarily. Qwyrxian (talk) 14:00, 2 January 2013 (UTC)

FWIW I like that wording, provided not a single comma is changed from it. It states very clearly the responsibilities of both sides of a discussion with respect to both sourcing and communication to build consensus. Diego (talk) 14:55, 2 January 2013 (UTC)

Qwyrxian, I'm confused. Your draft text seems to support the "have to have a reason to remove uncited information (or, at least, have to be willing to say that you have a reason)" position but your discussion of it seems to support the "uncited information can be removed merely because it is uncited" position. Drafts (A)-(C) all assume the latter position (and, indeed, I was a bit surprised that Amadscientist did not include a version which more directly supported the former, must-have-or-claim-a-reason position more clearly). As for me, I support version (C) followed by (B) (though I don't care much for including "try"), then (A). Best regards, TransporterMan (TALK) 15:43, 2 January 2013 (UTC)

To Qwyrxian. I started the RFC and followed the instruction to be sure and place notifications in the areas as suggested. I cannot speak for the editor who originally began the Dec 17 discussion. I have no idea if they made mention of that discussion anywhere else. Transportationman, the first version (A) is the version reverted back into the policy page by the editor who began the Dec 17 discussion with additons by a different editor. (B) is my copyedit version that I believe is better wording and (C) is a last minute suggestion (that I also like). I haven't decided which one to support yet as I was waiting to see if there was a better version that would be proposed by someone else. I kinda like Qwyrxian's version as well. Not entirely clear on what you are disapointing with but perhaps you could clarify or make a proposal?--Amadscientist (talk) 01:27, 3 January 2013 (UTC)

Well, perhaps it's a difference in perspective. To me, my version implies that any challengeable information may be removed. Maybe I haven't phrased mine correctly, but it implies that "rv: unsourced, dubious information" is sufficient. Similarly, going to an established article with unsourced info, one could remove it by saying "this info has been unsourced for a year, time to remove" is sufficient. This is because, for me, anything which rises in complexity above common sense is "challengable", and for me, what is "common sense" is an extremely narrow category of knowledge, and only refers to those things which would be common knowledge among those with only a primary level education across the English speaking world. I consider everything else as needing a source. To me, the earlier versions seem to make "tagging" the more likely choice, while my version makes "removal" the more likely choice. Furthermore, part of the thrust on my version is that if I remove unsourced info and someone else adds it back in, that when I re-revert to remove the info and start a talk page discussion, the default position is to remove the info while discussion is ongoing. Perhaps, though, my version didn't get that point across as clearly as it is in my mind. Qwyrxian (talk) 02:53, 3 January 2013 (UTC)

Some thoughts with an example

I am a physicist. If I see a statement in a physics article in WP with no supporting reference that I think is a bit dodgy I may tag or even deleted it. If I see something I think is wrong, which has no supporting reference, I will definitely delete it. So far so good.

When I look at the Category of sets page though things are a bit different (sorry page if I have sent you a plague of deletion locusts). It says:

'The epimorphisms in Set are the surjective maps, the monomorphisms are the injective maps, and the isomorphisms are the bijective maps.

The empty set serves as the initial object in Set with empty functions as morphisms. Every singleton is a terminal object, with the functions mapping all elements of the source sets to the single target element as morphisms. There are thus no zero objects in Set.'

There are no references. On the other hand I have no clue (well not much) what it means. Should I delete these statements because they have no supporting references? I think not. For all I know, the statements may be quite obvious to anyone with a significant grounding in the subject. I know that I know nothing about the subject and that I am therefore unqualified to say even whether these statements are even questionable.

Is there anyone here who thinks that I should delete these sentences, and in fact most of the article? Martin Hogbin (talk) 19:14, 3 January 2013 (UTC)

"Should delete"? Probably not, but not because you're not an expert. Any editor should be able to make any change to an article without being an expert in that topic, and lack of sourcing is in particular one of those reasons that any editor ought to be able to do that. "Can delete without fear of being blocked or banned for it?" The basic answer is yes, but it also depends on how you go about it (if you're going to do wholesale deletion then WP:PRESERVE, at least, kicks in). We're not really dealing with the "should" question here: I don't think anyone involved in this discussion would dispute that what should be done is to either find RS for the material, determine that RS probably don't exist and only then delete it, or fact-tag it (or the whole-article equivalent of fact-tagging) and leave it in place for someone else to work on. What we're in dispute about is what you "can" do or "must" do: Can you just delete it and walk away? Can you delete it, so long as you're willing to make a declaration that you have some doubt about it being verifiABLE? Best regards, TransporterMan (TALK) 21:11, 3 January 2013 (UTC)

That is all fine enough but it seem there are some editors who want to take a harder line by suggesting that statements without references should be deleted, even by people who do not understand what they mean or who have no real grasp of the background subject. Martin Hogbin (talk) 18:40, 4 January 2013 (UTC)

Re: "...even by people who ... have no real grasp of the background subject." first, we are not writing for those with a grasp of the background subject... we are writing for everyone and anyone. So, we should be paying attention to what those with no grasp of the background tell us. If they remove material for lack of citation, we know that the material needs a citation. More importantly, those of us who do have a grasp of the background should know where to find such citations. BURDEN should not be all that difficult to comply with if you actually do know the topic. Blueboar (talk) 18:26, 6 January 2013 (UTC)

This has nothing to do with people with no background knowledge trying to understand what is written. I agree, we should write for all levels of reader. What I am objecting to is someone who deletes a section just because it has no references and even though they have to reason to suspect that it is wrong. Do you support this? Martin Hogbin (talk) 19:32, 6 January 2013 (UTC)

Sometimes yes, sometimes no... it depends on the specific section, the specific articles, etc. etc. What I do support is trusting our editors to use their best editorial judgement to decide whether to remove or not. They might make a different decision than I would make, but that's OK.

After all, if removed material really is verifiable, it should take the rest of us very little time to find a citation for it (a citation that we probably should have added in the first place)... and since it will take far more time to whine and complain about the removal than to do this, my attitude is: why bother complaining... let those who wish to remove, remove freely... after all, returning what they remove, now supported by a citation, simply improves the article. Blueboar (talk) 20:37, 6 January 2013 (UTC)

What I ought to be writing here is just, "I agree with Blueboar," but let me gild the lily: I don't support the deletion of text merely because it has no refs if the deleting editor has no reason to suspect that there is nothing wrong with the text, but neither do I support making any presumptions about what a deleting editor has or does not have any reason to suspect. Instead, I support assuming in good faith that an editor who deletes text because it is unsourced believes that there is a problem with it, unless perhaps (and only perhaps; this is a matter that needs to be worked out here) the editor has demonstrated by his or her behavior that there is a very strong evidence that such assumption is incorrect. Best regards, TransporterMan (TALK) 15:44, 7 January 2013 (UTC)