Wikipedia talk:Deprecated sources
![]() | Discuss sources on the reliable sources noticeboard To discuss the reliability of specific sources, please start or join a discussion on the reliable sources noticeboard (WP:RSN). |
This is the talk page for discussing improvements to the Deprecated sources page. |
|
![]() | This project page does not require a rating on Wikipedia's content assessment scale. It is of interest to the following WikiProjects: | ||||||||||||||||||||
|
![]() | This project page has been mentioned by multiple media organizations:
|
Behavioral guidance
editIn the currently active RfC for The Sun, Andy Dingley shared some concerns regarding editor behavior when deprecated sources are used. A section on this page that provides behavioral guidance to other editors would be welcome. Specifically, the section should address:
- How to respond to an editor who uses a deprecated source in violation of restrictions set by community consensus
- How to respond to an editor who inappropriately reverts a citation of a deprecated source that is covered under one of the exceptions
New user warnings with appropriate messages may be helpful. — Newslinger talk 12:46, 8 December 2018 (UTC)
Integration of alternate version
editI was actually writing my own version of this essay at the same time. :-) I have added my text in its original form and will do the integration next. Please give me some time to finish editing, and afterwards please feel free to modify/revert/etc. Sunrise (talk) 06:04, 9 December 2018 (UTC)
Done (also pinging Newslinger). I tried my best not to privilege either version when removing duplicated material (although I imagine I may have been biased in favor of myself despite that). I did make a few other adjustments as I edited, the biggest one being that the idea of being more reliable historically was actually specific to the Daily Mail. Using it as a primary source about itself is fairly easily extrapolated to other cases, but historical reliability was a specific counterargument advanced by a few editors in the RfC who commented that the Daily Mail had a better reputation in the past. Anyways, I ended up merging the exceptions into the next section to keep everything together. Sunrise (talk) 07:45, 9 December 2018 (UTC)
- This looks great! I like how your version gives detailed advice for handling deprecated sources in general, instead of relying too heavily on the 2017 Daily Mail RfC. The "Acceptable uses..." section also provides some behavioral guidance, which should be useful. Thanks for merging this in! — Newslinger talk 18:40, 9 December 2018 (UTC)
Implementing edit filter warnings for deprecated sources
editThere is a discussion at Wikipedia talk:Reliable sources/Noticeboard § Implementing edit filter warnings for deprecated sources that relates to this page. Please feel free to participate if you are interested. — Newslinger talk 08:54, 20 January 2019 (UTC)
The Irish Sun
editWould it be alright to add The Irish Sun (without the link) under The Sun (UK) in the same box in the sources section of the project page? The Irish Sun is part of The Sun as can be seen on that page and The Irish Sun links to The Sun (United Kingdom)'s page. Its just that the way it is curently stated does not make it clear from looking at the table alone that this includes the Ireland addition, as the Repbulic of Ireland is a different country to the United Kingdom. Helper201 (talk) 03:11, 6 November 2019 (UTC)
- Let's centralize this discussion at WT:RSP § The Irish Sun. — Newslinger talk 14:11, 6 November 2019 (UTC)
Is deprecation just for "ongoing" (continuously published) news sources?
editThe list seems to indicate as much, but is this just an accident of history? I ask because, over three years before the Daily Mail RFC, I opened a discussion at RSN about the Japan Encyclopedia, and there was an (admittedly weak) consensus that it was sub-optimal, should probably not be used going forward, and should probably be replaced where it is currently in use. The book's usage has apparently declined at a rate of around 100 articles annually since then (in 2013 I said it was used in around 1,000, and now the number is down to 426), despite my not having done much heavy lifting in removing it, so it may not be worth discussing, but I'm wondering if an RFC for deprecation would be worthwhile for future instances of similar sources. But are field-specific works too "niche" to be formally deprecated by RFC? Hijiri 88 (聖やや) 13:25, 4 December 2019 (UTC)
- There isn't any inherent requirement for a particular publication status or source type. I would say that the list's current composition in this respect reflects the fact that continuously published sources are more likely to have new citations added to WP, and therefore naturally tend to be greater concerns for editors to deal with. Similarly, there are fewer field-specific sources included only because they're less likely to be widely cited, making the deprecation of other sources more likely to be a priority. One of the key points here is that deprecated sources aren't "special" in any way that isn't already established by the existing content policies, with the related element that just because a poor source isn't deprecated it doesn't mean that it isn't just as unreliable as those that are. So I'd still want to mention that you don't need deprecation in order to remove sources for being generally unreliable (using the term as defined at WP:RSP), and that could resolved at RSP without needing a full RfC. Otherwise, to decide on whether to hold an RfC, I'd ask about whether you think there's any particular benefit to formalizing a consensus: e.g. if the status of the source is unclear, if it keeps coming up in discussions, if you think it should be edit-filtered, or if you're planning to review a large number of the existing uses. Or, if you're confident that any RfC would lead to a consensus for deprecation, then there is also the option of treating the source as de facto deprecated and therefore cutting out the bureaucracy. Sunrise (talk) 00:48, 5 December 2019 (UTC)
- I'd say it's more that depreciation is generally reserved for sources that are both unreliable and which have people that constantly try to use them regardless. "Dead" sources rarely meet the latter criteria because they don't tend to have a strong audience or fanbase, so it's rarely necessary to take the step of depreciating them. --Aquillion (talk) 22:20, 30 May 2020 (UTC)
Missing information (kind of)
editHello, There are a few other sources that I believe are also deprecated but are not mentioned on the list. Is this list updated or how can we update it? As far as I know, there are two sources in my knowledge that are blocked but are not available here. They include The American Reporter and Thrive Global. Can someone please look into this? Moreover, can someone tell me if there is a place where I can check what sources are deprecated and about whom there are discussions going on? This can be of significant help when working on edits and new articles establishing if a subject meets WP:GNG or not. Thanks. NotJuggerNot (talk) 15:38, 7 May 2020 (UTC)
Can anyone look into this, maybe?NotJuggerNot (talk) 01:36, 13 May 2020 (UTC)
- Hi NotJuggerNot, this page only lists the sources that have undergone a request for comment on the reliable sources noticeboard that found consensus for deprecation. I wasn't able to find any discussions of American Reporter or Thrive Global on the noticeboard, and neither of these sites are on the spam blacklist. Thus, neither of these sites are deprecated or blacklisted. However, Thrive Global is a consulting firm and their company website is almost certainly not considered a reliable source for notability purposes. American Reporter (american-reporter.com) has been defunct since its founder passed away in 2016; archived versions of the site are still usable, although I'm not entirely sure it's considered a reliable source and you may want to double-check on the noticeboard. Please note that theamericanreporter.com is an entirely different site – one looks like a content farm, which is considered self-published. — Newslinger talk 03:07, 13 May 2020 (UTC)
Recent Changes
editSashiRolls, can you explain you recent changes here? More so the most recent change, as it wasn’t clear from the edit summary. As far as the one before that, I wouldn’t mind an explanation, but it’s not necessary. I’m fairly sure that wouldn’t be an acceptable edit for most people. Honestly, I’m just confused about the most recent partial-revert. Symmachus Auxiliarus (talk) 00:46, 24 May 2020 (UTC)
- A block is an automatic procedure governed by regex at the global spamlist, which is indeed different from a ban. Technically speaking, it is physically possible to click through the machine-run edit filter and enter a deprecated source as a source about itself, unless of course it's not considered a reliable source about itself, in which case you really aren't supposed to. ^^ -- SashiRolls 🌿 · 🍥 01:04, 24 May 2020 (UTC)
I see that the word "block" was removed again. I have restored it since the mediawiki page describing blacklisting process uses the term "block" eleven times. I see no reason to avoid straightforward description using the same terms as on the page describing the process. -- SashiRolls 🌿 · 🍥 11:21, 30 May 2020 (UTC)
- On the English Wikipedia, blocks are applied to users, not to sources or domains. Blacklisting refers to disallowing the use of a domain. The term block is also imprecise, because it does not consider the fact that parts of the domain can be whitelisted. This page draws a comparison between deprecation and banning, not blocking, because the term ban was used by media sources. Finally, why does your edit in Special:Diff/959743328 add "deprecated" in scare quotes when this page is called Wikipedia:Deprecated sources? — Newslinger talk 11:33, 30 May 2020 (UTC)
- Hi, @Newslinger:. I think we should be rigorous. You did not comment on the fact that the page describing the blacklisting process calls it a "block" not a "ban", which is correct, as it is based on reg-ex matching, not on human intervention. The use of the specialized term "deprecate" should be explained on this page. To "deprecate" means to criticize (cf. merriam-webster), or to replace an old syntax or function with a new one... e.g. <font> tags being replaced. Reliable sources agree that a prohibition is a ban. The most common "support" vote in the 2017 RFC was "support prohibition". It is for this reason that the use of scarequotes around the term "ban" is dubious. Your argument about whitelisting seems to me a red herring, because white-listing (where it exists -- you would need to provide examples) only allows for exceptions to a more general machine "block". -- SashiRolls 🌿 · 🍥 11:46, 30 May 2020 (UTC)
I did not equate blacklisting to a "ban", andboth blocks (WP:BLOCK) and bans (WP:BAN) are most frequently associated with users (not domains or sources) on the English Wikipedia, regardless of how MediaWiki's manual describes the SpamBlacklist extension. In this context, deprecation is defined in "Effects of deprecation" and has nothing to do with syntax. The "About Us" pages of notable blacklisted sites are frequently whitelisted for use in the articles on the sites. See MediaWiki talk:Spam-whitelist/Common requests § General exceptions for details. I've opened an RfC to invite opinions from other editors. — Newslinger talk 12:50, 30 May 2020 (UTC)
- Hi, @Newslinger:. I think we should be rigorous. You did not comment on the fact that the page describing the blacklisting process calls it a "block" not a "ban", which is correct, as it is based on reg-ex matching, not on human intervention. The use of the specialized term "deprecate" should be explained on this page. To "deprecate" means to criticize (cf. merriam-webster), or to replace an old syntax or function with a new one... e.g. <font> tags being replaced. Reliable sources agree that a prohibition is a ban. The most common "support" vote in the 2017 RFC was "support prohibition". It is for this reason that the use of scarequotes around the term "ban" is dubious. Your argument about whitelisting seems to me a red herring, because white-listing (where it exists -- you would need to provide examples) only allows for exceptions to a more general machine "block". -- SashiRolls 🌿 · 🍥 11:46, 30 May 2020 (UTC)
- The page itself equated blacklisting with banning until 11:28 30 May 2020. -- SashiRolls 🌿 · 🍥 03:47, 31 May 2020 (UTC)
- Fair enough. Now that the {{Citation needed}} tag is removed, I consider this matter settled. — Newslinger talk 05:16, 31 May 2020 (UTC)
- The page itself equated blacklisting with banning until 11:28 30 May 2020. -- SashiRolls 🌿 · 🍥 03:47, 31 May 2020 (UTC)
RfC: Acceptable uses of deprecated sources
editWhich of the following should be used as the first sentence in Wikipedia:Deprecated sources § Acceptable uses of deprecated sources? — Newslinger talk 12:41, 30 May 2020 (UTC)
See #Recent Changes for context. — Newslinger talk 12:41, 30 May 2020 (UTC)
Option 1: (from Special:Permalink/959750864)
Although deprecated sources are considered generally unreliable, there are rare situations in which they can be used.
Option 2: (from Special:Permalink/959751455)
Deprecation is not a ban on using the source,[citation needed] despite having been reported as such in the media.[1]References
- ^
- Kalev Leetaru (2 October 2017). "What Wikipedia's Daily Mail 'Ban' Tells Us About The Future Of Online Censorship". Forbes. Retrieved 25 December 2018.
- Jasper Jackson (8 February 2017). "Wikipedia bans Daily Mail as 'unreliable' source". The Guardian. Retrieved 21 November 2018.
- Jon Sharman (9 February 2017). "Wikipedia bans Daily Mail because it's an 'unreliable source'". The Independent. Retrieved 21 November 2018.
- Sebastian Anthony (10 February 2017). "Wikipedia bans Daily Mail for "poor fact checking, sensationalism, flat-out fabrication"". Ars Technica. Retrieved 21 November 2018.
- Also see Daily Mail § Other criticisms.
- Option 1. The sentence is consistent with the rest of the page. The previous section, "Effects of deprecation", explains that deprecated sources are considered generally unreliable, and the remainder of the "Acceptable uses of deprecated sources" section elaborates on the rare situations in which they can be used (including under the WP:ABOUTSELF policy). Since this page is not in article space, there is no need to include citations to external sources. The {{Citation needed}} tag in option 2, recently added in Special:Diff/959750231, does not improve the page. — Newslinger talk 12:41, 30 May 2020 (UTC)
Since this page is not in article space,
(Sorry, but this is technically not true: cf. Deprecated sources). -- SashiRolls 🌿 · 🍥 15:23, 30 May 2020 (UTC)- My point is that Wikipedia:Deprecated sources is not in the Wikipedia mainspace, and is exempt from WP:V ("All material in Wikipedia mainspace, including everything in articles, lists and captions, must be verifiable"). Wikipedia:Deprecated sources does not need to cite reliable sources, nor does it need to have a {{Citation needed}} tag on claims not cited to reliable sources. Redirects to Wikipedia:Deprecated sources are not relevant. — Newslinger talk 01:24, 31 May 2020 (UTC)
- I disagree that a page which redirects from mainspace into Wikipedia namespace does not need any references. Moreover, it cannot say, in Wikipedia's voice, what this page said prior to my edit. -- SashiRolls 🌿 · 🍥 02:19, 31 May 2020 (UTC)
- I've started a discussion on this matter at WT:V § Applicability to pages in project space. — Newslinger talk 02:33, 31 May 2020 (UTC)
- I disagree that a page which redirects from mainspace into Wikipedia namespace does not need any references. Moreover, it cannot say, in Wikipedia's voice, what this page said prior to my edit. -- SashiRolls 🌿 · 🍥 02:19, 31 May 2020 (UTC)
- As a result of this discussion, the redirect was removed and is now being considered for deletion. I had originally noticed it because it made the page easier to find. -- SashiRolls 🌿 · 🍥 14:41, 31 May 2020 (UTC)
- My point is that Wikipedia:Deprecated sources is not in the Wikipedia mainspace, and is exempt from WP:V ("All material in Wikipedia mainspace, including everything in articles, lists and captions, must be verifiable"). Wikipedia:Deprecated sources does not need to cite reliable sources, nor does it need to have a {{Citation needed}} tag on claims not cited to reliable sources. Redirects to Wikipedia:Deprecated sources are not relevant. — Newslinger talk 01:24, 31 May 2020 (UTC)
- Option 2 without the {{Citation needed}} tag is also fine, in my opinion. — Newslinger talk 13:38, 30 May 2020 (UTC)
- Neither. However, longstanding content (since 9 December 2018) including RS should not be summarily deleted, as was done here. If en.wp wishes to make a distinction between "ban" and "deprecate" which is not simply the distinction between "ban"/"deprecate" and "block" then RS should be cited to support the distinction, since RS overwhelmingly do not accept that distinction between "deprecate" and "ban". Further, the dictionary definition of "deprecate" does not list "ban" as a synonym. Moreover, the most common support vote in the 2017 RfC was "support prohibition". Prohibition is indeed a synonym for ban. Also, creating an RfC which tries to mandate the inclusion of a "citation needed" template is silly. Supplying a reference would be considerably more straightforward.
- tldr;: have the courage of your convictions: call a ban a ban & a prohibition a prohibition. -- SashiRolls 🌿 · 🍥 13:37, 30 May 2020 (UTC)
- Regarding "creating an RfC which tries to mandate the inclusion of a 'citation needed' template", your comment misrepresents my intent. You added the {{Citation needed}} tag in Special:Diff/959750231 (option 2), and when I replaced the sentence with alternative wording in Special:Diff/959750864 (option 1), you reverted my edit in Special:Diff/959751455 which restored the {{Citation needed}} tag. I started this RfC with the intention of eliminating the {{Citation needed}} tag, not to "mandate the inclusion" of the tag. — Newslinger talk 01:37, 31 May 2020 (UTC)
- I eliminated the CN template ten hours before you wrote this. Is your intention to eliminate the references present in the article since December 2018? My apologies, but I don't understand why you created an RfC 21 minutes after my restoration of the references you deleted. -- SashiRolls 🌿 · 🍥 02:19, 31 May 2020 (UTC)
- No, my intention is to eliminate the {{Citation needed}} tag. I am indifferent about the citations. — Newslinger talk 02:34, 31 May 2020 (UTC)
- Done at 16:41, 30 May 2020. -- SashiRolls 🌿 · 🍥 03:02, 31 May 2020 (UTC)
- Thank you. This RfC was started before that edit. Since editors in this discussion have proposed other wordings, this RfC will determine whether there is consensus for any other changes. — Newslinger talk 03:12, 31 May 2020 (UTC)
- Done at 16:41, 30 May 2020. -- SashiRolls 🌿 · 🍥 03:02, 31 May 2020 (UTC)
- No, my intention is to eliminate the {{Citation needed}} tag. I am indifferent about the citations. — Newslinger talk 02:34, 31 May 2020 (UTC)
- I eliminated the CN template ten hours before you wrote this. Is your intention to eliminate the references present in the article since December 2018? My apologies, but I don't understand why you created an RfC 21 minutes after my restoration of the references you deleted. -- SashiRolls 🌿 · 🍥 02:19, 31 May 2020 (UTC)
- Regarding "creating an RfC which tries to mandate the inclusion of a 'citation needed' template", your comment misrepresents my intent. You added the {{Citation needed}} tag in Special:Diff/959750231 (option 2), and when I replaced the sentence with alternative wording in Special:Diff/959750864 (option 1), you reverted my edit in Special:Diff/959751455 which restored the {{Citation needed}} tag. I started this RfC with the intention of eliminating the {{Citation needed}} tag, not to "mandate the inclusion" of the tag. — Newslinger talk 01:37, 31 May 2020 (UTC)
- tldr;: have the courage of your convictions: call a ban a ban & a prohibition a prohibition. -- SashiRolls 🌿 · 🍥 13:37, 30 May 2020 (UTC)
- Neither As 1 seems to just be an enabler of "but its reliable in this instance" style arguments. And option 2 really does not tell us anything (This also have been used as an excuse to use a deprecated source). Slatersteven (talk) 13:38, 30 May 2020 (UTC)
- Option 1 if we have to choose between them; it's better wording in that it integrates this with our other policies by mentioning "generally unreliable" specifically. I dislike the wording of option 2 because there's something a bit passive-aggressive about the way it talks about the media; additionally, the argument that it is a functional ban even if some limited uses exist is reasonable (ie. 'ban' doesn't necessarily mean 'banned in absolutely all cases', which makes the nitpicking argument that that sentence is trying to pick with the media a bit dubious.) I agree with the people above saying that even better wording might be possible, but between these two I definitely prefer 1. As far as the "it encourages people to use them!" thing goes, there have always been (very) limited acceptable usages, which the section in question goes into more detail on. Edit: Option 3 below has all the problems of option 2; I'm strenuously opposed to any version that includes the "contrary to the media" wording or anything to that effect. Additionally, option 1 is preferable in that it establishes that using the source should be rare, while 2 and 3 seem like they're worded more to hedge the concept of depreciation by saying "but it's tots not a ban, guys." Focusing on what depreciation is rather than what it isn't makes far more sense, especially avoiding the unduly confrontational "it's not what the media told you it is" framing. --Aquillion (talk) 14:22, 30 May 2020 (UTC)
- You do not mention "option 3". Could you comment on that option please? -- SashiRolls 🌿 · 🍥 14:25, 30 May 2020 (UTC)
- I'll move it here:
- You do not mention "option 3". Could you comment on that option please? -- SashiRolls 🌿 · 🍥 14:25, 30 May 2020 (UTC)
- because:
- reason (e.g. a source may sometimes be used as a source about itself if it is considered reliable about itself.)
- reason (e.g. a source may have once been considered reliable (start-date, end-date))
- reason (?)
References
- ^
- Kalev Leetaru (2 October 2017). "What Wikipedia's Daily Mail 'Ban' Tells Us About The Future Of Online Censorship". Forbes. Retrieved 25 December 2018.
- Jasper Jackson (8 February 2017). "Wikipedia bans Daily Mail as 'unreliable' source". The Guardian. Retrieved 21 November 2018.
- Jon Sharman (9 February 2017). "Wikipedia bans Daily Mail because it's an 'unreliable source'". The Independent. Retrieved 21 November 2018.
- Sebastian Anthony (10 February 2017). "Wikipedia bans Daily Mail for "poor fact checking, sensationalism, flat-out fabrication"". Ars Technica. Retrieved 21 November 2018.
- Also see Daily Mail § Other criticisms.
ps: I'm not "absolutely" attached to the adverb "absolutely", but it might be best to describe the reality...-- SashiRolls 🌿 · 🍥 14:39, 30 May 2020 (UTC)
- So, the real issue here is that fans of a given deprecated source will use any excuse whatsoever to claim an exception for their particular favourite source or particular favourite usage of it.
- In the course of removing usages of The Sun or the Daily Mail - almost every usage of which is bad or should be presumed bad, by the clear wording of the RFC conclusions - I have frequently been told that the mere fact that a source is "generally prohibited" means there are exceptions, therefore the particular usage in question should be allowed, and that's the whole excuse. I recently had a demand to prove from first principles that the Mail is bad, despite two general RFCs. I have had two editors in the past week claim that removing WP:DAILYMAIL usages violates WP:NOTCENSORED. Last year there was a user (who turned out to be a defrocked ex-admin) literally following my edits around, restoring usages of The Sun. The same user insisted and edit-warred a controversial claim on a BLP sourced to a no-longer-available page on thesun.co.uk back into the article, on the excuse that it was about a sportsman and WP:RSP said some editors thought the Sun was OK for sports. I have a talk page full of similar nonsense, which is a catalogue of bad and weird excuses for using sourcing that absolutely shouldn't be in Wikipedia.
- So - whatever wording is used here will be used as an excuse by the WP:IDIDNTHEARTHAT squad. That's what I'd like to see a new wording use as its threat model.
- I would suggest something that stresses that, although there are exceptions, they are extremely rare. If you're looking for excuses to use a source, you're doing Wikipedia sourcing wrong.
- Any advocate's desired exception almost certainly isn't one - The Sun used to have over 8,000 usages, and now has five. (And I'm very sceptical that The Sun was as involved in unmasking László Csatáry as they claim, but I'd have to research that one a lot more to knock it off.) Even quite a lot of WP:BLP "permissible" WP:SPS or WP:BLPSELFPUB usages shouldn't be there, and if the claims in them are noteworthy for a BLP then they would have been noted elsewhere.
- We don't need an explanation to tell sensible people that they shouldn't use sewage-quality sources - we need a stick for the ones that insist on it anyway - David Gerard (talk) 16:18, 30 May 2020 (UTC)
- All that is fine. What I think is needed is straightforward transparency, not saying things like "it's not really a ban coz' the software doesn't automatically block every use and zap the offending would-be editor's keyboard with however many volts." :) -- SashiRolls 🌿 · 🍥 17:00, 30 May 2020 (UTC)
- "Deprecation is not a blanket ban - although deprecated sources are "generally prohibited", there may be acceptable uses for a deprecated source, e.g., permissible WP:SPS or WP:BLPSELFPUB usages. But, although there are exceptions, they are extremely rare. In almost all cases, the deprecated source should not be used at all, even where technically permissible." How's that? - David Gerard (talk) 17:52, 30 May 2020 (UTC)
- All that is fine. What I think is needed is straightforward transparency, not saying things like "it's not really a ban coz' the software doesn't automatically block every use and zap the offending would-be editor's keyboard with however many volts." :) -- SashiRolls 🌿 · 🍥 17:00, 30 May 2020 (UTC)
- I would change possibly permissible to technically permissible, but otherwise I like it. --Guy Macon (talk) 22:09, 30 May 2020 (UTC)
- You're right - just changed it to "technically" :-D - David Gerard (talk) 00:26, 31 May 2020 (UTC)
- I would change possibly permissible to technically permissible, but otherwise I like it. --Guy Macon (talk) 22:09, 30 May 2020 (UTC)
- SashiRolls, there's an edit filter. The filter warning is as follows:
- An automated filter has detected that you are adding a link to a deprecated source, considered generally unreliable after discussion by the community. References to these sources are generally prohibited, especially when other more reliable sources exist. Please cite a more reliable source instead. If the only source you can find for the claims is one of these deprecated sources then you should not add the content in question.Note: There are limited exceptions to this rule (such as when the source itself is the topic being discussed).
- If you have checked the policy on Wikipedia:Reliable sources and the deprecated source guidance (or checked at Wikipedia:Reliable sources/Noticeboard) and verified that your edit is one of the limited exceptions, then you may resubmit it by clicking "Publish changes" again. Please do not do this unless you have first verified that this specific use of the deprecated source has broad support, especially for biographical articles: deprecated sources are liable to be removed on sight and persistent addition can lead to editing restrictions.
- Your proposals are not in line with this, and also not in line with the multiple RSN discussions that have shown majority or supermajority support for removing deprecated sources. The Daily Mail is in fact an edge case: a more representative onw ould be something like VDARE, which should never have been used in the first place. Guy (help!) 10:29, 1 June 2020 (UTC)
- I think a lot of y'all may be misjudging my intent. I'm listening to what you've said and have incorporated suggestions in the latest edits, which you are more than welcome to improve upon. I think periodically tightening info pages up can never hurt. The wishy-washiest language on the page is legacy text. -- SashiRolls 🌿 · 🍥 10:43, 1 June 2020 (UTC)
Support David Gerard's wording. See above. --Guy Macon (talk) 22:09, 30 May 2020 (UTC)
- David Gerard's wording looks good to me. It is consistent with policy and practice, and does not include an unnecessary {{Citation needed}} tag. — Newslinger talk 01:18, 31 May 2020 (UTC)
- Since you continue referring to the CN tag, I have struck "Option 2" above since nobody is or was arguing for it. It was added at the same time as "ban" in scarequotes was rewritten without scarequotes. It was restored once because you deleted 3 independent references from the page in the next edit. -- SashiRolls 🌿 · 🍥 02:31, 31 May 2020 (UTC)
- (Not serious comment) we could always use my version of [Citation Needed] ... Note: click on the "Citation Needed" in the previous sentence for a surprise!) --Guy Macon (talk) 15:54, 31 May 2020 (UTC)
- glad to see you're feeling peppery again, Guy. :) -- SashiRolls 🌿 · 🍥 02:34, 1 June 2020 (UTC)
- (Not serious comment) we could always use my version of [Citation Needed] ... Note: click on the "Citation Needed" in the previous sentence for a surprise!) --Guy Macon (talk) 15:54, 31 May 2020 (UTC)
Reject Added wording in which WP:SPS or WP:BLPSELFPUB are mentioned. WP:ABOUTSELF (already mentioned in the article) is relevant. The above do not seem to be. (i.e. for RT, Breitbart, The Sun, The Daily Mail, etc. it is silly to talk about self-published sources or BLP self-pub)... The rest is fine... I would also like to see the references to reliable sources maintained. Feel free to edit the entry directly David. This RfC was started to eliminate a CN template. That has been done. Perhaps we could close this as a successful RfC?-- SashiRolls 🌿 · 🍥 02:19, 31 May 2020 (UTC)
- Yeah, WP:ABOUTSELF is fine for what I mean here - David Gerard (talk) 08:29, 31 May 2020 (UTC)
- I ended up adding it as a cf. I'm not very clear on the reasons for mentioning it, though I suppose exclusive press releases type things are possible. In any case, it does no harm to link to them, it boils down to the same thing.-- SashiRolls 🌿 · 🍥 10:59, 1 June 2020 (UTC)
- I mentioned it because something like it is a frequent excuse for DM usages - David Gerard (talk) 08:20, 4 June 2020 (UTC)
- I ended up adding it as a cf. I'm not very clear on the reasons for mentioning it, though I suppose exclusive press releases type things are possible. In any case, it does no harm to link to them, it boils down to the same thing.-- SashiRolls 🌿 · 🍥 10:59, 1 June 2020 (UTC)
- Support something like the David Gerard version, don't have strong opinion about WP:ABOUTSELF vs BLPSEFPUB ... buidhe 01:54, 1 June 2020 (UTC)
- David Gerard's wording is correct. Adamantly oppose Options 2 and 3 as both offer an open invitation to drama and POV-pushing - it could, for example, be used to resist removval of clearly inappropriate fake news sites, things like WorldNetDaily and so on. Guy (help!) 10:23, 1 June 2020 (UTC)
- Support on Option 3 with the wording that I see David has suggested of why. In addition, a case that has been discussed is related to the idea of established experts in their field (as determined by external sourcing and consensus) that publish their recognized opinions on notable topics in the deprecated source - eg the most common being the UK media critics that are regulars and other sources. This is an extension of RSOPINION, which has been discussed at WT:RS and VPP but without unfortunately getting too many external voices to provide input. We don't need to add that now, but I do agree leaving placeholders for exceptions like this. --Masem (t) 00:40, 2 June 2020 (UTC)
- "that has been discussed" is a weasel phrasing here - what you mean is that you've brought it up at WP:RSN and been unable to get consensus to carve out an exception for that specifically. So no, I don't think adding your desired carveout is good - David Gerard (talk) 08:19, 4 June 2020 (UTC)
- Support Option 3, but without the bit about Media reports, which is reeally not relevant here. Option 3 makes mcleaer the limited scope of exceptions when deprecated sources may be used. DES (talk)DESiegel Contribs 16:58, 4 June 2020 (UTC)
Disputed wording
editSashiRolls added the following in Special:Diff/960143466:
Deprecating a source is also a weaker measure than deploying a bot to replace every instance of the source with a citation needed tag.
I am disputing this wording under WP:BRD, as I do not think it improves the page. I am not aware of any bot that does what is suggested here. — Newslinger talk 10:25, 1 June 2020 (UTC)
- That's fine and inessential. I tried to incorporate some of David Gerard's version into the text, divide legacy sourcing from future sourcing and add a bit of etymology... really "pray against" should be "preach against"... ^^ -- SashiRolls 🌿 · 🍥 10:27, 1 June 2020 (UTC)
- Newslinger, deprecation is deprecation. What people do to remove the deprecated source is a matter for discussion ona source by source basis. A fake news site will be nuked faster than a source that is merely crap, after all. Guy (help!) 23:47, 7 July 2020 (UTC)
Lenta.ru, 112 Ukraine and Veterans Today
editIt seems like these sources were deprecated through a larger RfC on sites identified by RS as state-sponsored fake news / disinfo. It might be wise to add an asterisk to these sources on the list and to add a note that state-sponsored fake news / disinfo has been deprecated. Snooganssnoogans (talk) 20:17, 7 July 2020 (UTC)
Certain sources
editI have serious issues with the deprecation of The Grayzone, Mint Press News, Occupy Democrats, RT, TeleSur, and Sputnik. Why not Fox News and Newsmax? I also think anything The New York Times says about the Israel-Palestine situation is untrustworthy (it would take too long to go into details, but somehow I think nothing The New York Times could do would ever get them deprecated. --Scottandrewhutchins (talk) 23:36, 7 July 2020 (UTC)
- Scottandrewhutchins, check the archives of WP:RSN and the linked debates. RSN is where we discuss individual entries. Guy (help!) 23:48, 7 July 2020 (UTC)
- In regards to grayzone -so far, I found one archive that is mostly about Max Blumenthal and his alleged polotical ties. https://en.m.wikipedia.org/wiki/Wikipedia:Reliable_sources/Noticeboard/Archive_287#RfC:_Grayzone The summary - "There is consensus that The Grayzone publishes false or fabricated information." Like what exactly? With what proof? It seems like a very small group of us military hawks busy on wikipedia, can create a consensus to call anything as fake news if they don't like it. But no specific compelling proof were ever given that he was making fake news. Unless writing negatively about america's role in libya, etc are deemed lies by editors?
American media has been caught in many lies and biaaes, yet are called reliable? Hpw does that make sense?https://www.salon.com/2016/09/16/u-k-parliament-report-details-how-natos-2011-war-in-libya-was-based-on-lies/
Granted I am still new to this media but been reading it for a day, and noticed that it has a solid tendency to call out US MEDIA misinformation many months before it is recognised as such. Like with ie. Bolivia - Nyt very recently confessed that its many reports of election rigging were fundamentally flawed. But who had alone criticised them for that months ago, and were proven correct? Not mainstream media but the independant media Grayzone seem to publicly point out how NYT had misled months before NYT finally admitted their reports were flawed and when things are now too late to change. They appear to be the anti-fake news to me, based on bolivia alone. https://thegrayzone.com/2019/11/14/oas-us-coup-bolivia-evo-morales/#more-16684 If measured by number of disinformation, NYT would deserve to be called a depreciated source. Not grayzone who were the ones who had called them out correctedly. https://theintercept.com/2020/06/08/the-nyt-admits-key-falsehoods-that-drove-last-years-coup-in-bolivia-falsehoods-peddled-by-the-u-s-its-media-and-the-nyt/ MangoTareeface9 (talk) 08:30, 10 July 2020 (UTC)
- The WP:DEPSOURCES list is just an index of deprecated sources, and discussions on this talk page are not considered in evaluations of the sources. You are free to discuss specific sources, like The Grayzone, on the reliable sources noticeboard. — Newslinger talk 16:07, 18 July 2020 (UTC)
Mail on Sunday
editThe Mail on Sunday just got deprecated - see WP:MAILONSUNDAY. I also added it to the deprecation warning filter. Anyone want to add it to the table? - David Gerard (talk) 15:57, 16 November 2020 (UTC)
Proposed clarification of deprecated sources guidelines
editWhile deprecation "is not a blanket retroactive "ban" it appears that it's treated as such. I've come across mass deletion of references to Sputnik news agency which included completely uncontroversial ones (such as ([1], [2], [3]) where no new source was supplied. I agree with the decision about the deprecation of Sputnik in general and believe that no references to it should be added to Wikipedia, however I don't think that mass-deletion of existing links improves the encyclopedia when the statement is uncontroversial and there are no other references. Having a reference rather than a [citation needed] tag would made finding a new more reliable source easier.
I'm talking about really uncontroversial facts like the number of visas granted to Muslim pilgrims or adding certain food to the UNESCO intangible heritage list. The problem with Sputnik is that there is a lot of Russia propaganda there, not that that they get random facts wrong.
I saw that this is not an isolated incident but something that generates a lot of similar complaints, which is why I'm proposing to clarify the acceptable uses (new text bold):
Initial version
| |||
---|---|---|---|
|
“ | Deprecation is not a blanket retroactive "ban" on using the source in absolutely every situation, contrary to what has been reported in media headlines. In particular, reliability always depends on the specific content being cited, and all sources are reliable in at least some circumstances and unreliable in at least some others. Citations to deprecated sources should not be removed indiscriminately, and each case should be reviewed separately. Users should make a reasonable effort to replace deprecated sources by reliable ones rather than removing them. | ” |
Alaexis¿question? 20:13, 25 November 2020 (UTC)
- Comment I think the issue here is more that this editor is doing mass removals of sources and making zero effort to find replacements. I came across them when they removed a source from an article on my watchlist and it took about five seconds to find a replacement by Googling it. It's verging on disruption (or possibly simply is disruption) and probably should be raised at ANI to get them to stop (they are almost certainly in violation of the requirement not to remove them indiscriminately given the speed of their edits). Back on topic, I would support an amendment to the wording that editors have to make reasonable efforts to find a replacement before removing a deprecated source, but I would not advocate retaining them for uncontroversial stuff on the basis that this potentially gives them traffic. Number 57 20:28, 25 November 2020 (UTC)
- Sounds good.Alaexis¿question? 20:37, 25 November 2020 (UTC)
- Regarding the user indirectly mentioned above, this mass blanket removal had been raised with them multiple times previously. When it came around to Telesur TV, they responded with an eye-rolling-inducing red herring of WP:AGF, and turning the replacement standard mentioned above on its head: no, the very nature of the mass removal assumes there is no replacement source, or else they would have resorted to {{Deprecated inline}}. CaradhrasAiguo (leave language) 19:56, 26 November 2020 (UTC)
- Some wording that speaks to a reasonable good faith effort to review, seek possible replacement, and consider a possible tag of the source for each use of a deprecated source, while strongly discouraging mass removal without reasonable review, is appropriate. We shouldn't try to dictate what the proper procedure should be for a review, but to simply acknowledge mass removals of deprecate sources should not be done, unless there is specific consensus for such mass removal. This advice to mean seems to be implicit from the line "Citations to deprecated sources should not be removed indiscriminately..." advice that already exists, but clearly some are not taking that that way. --Masem (t) 20:37, 26 November 2020 (UTC)
- This page is a description of consensus. If you want to change how deprecated sources are handled on Wikipedia, you probably need to take the issue to WP:RSN - where this question has come up repeatedly and been rejected repeatedly - David Gerard (talk) 00:02, 26 November 2020 (UTC)
- As far as I understand WP:RSN is the place to discuss whether a source is reliable in a certain context. I don't want to argue about these 3 cases or even this source in general. I want to define more explicitly what the current consensus position (Citations to deprecated sources should not be removed indiscriminately) means. Alaexis¿question? 16:10, 26 November 2020 (UTC)
- You'll excuse me if I find this functionally difficult to distinguish from an attempt to claim WP:LOCALCONSENSUS on an out-of-the-way page, after the idea has been multiply rejected, with citations to policy backing its rejection, on WP:RSN and even on WP:VPP - David Gerard (talk) 19:27, 26 November 2020 (UTC)
- I've flagged this discussion at RSN: Wikipedia:Reliable sources/Noticeboard#Discussion_seeking_to_stop_deprecated_sources_being_removed_from_articles - David Gerard (talk) 19:40, 26 November 2020 (UTC)
- Thanks for mentioning it there. I couldn't find any relevant guidelines about it on WP:RS which is why I decided that a clarification would be useful. Alaexis¿question? 19:53, 26 November 2020 (UTC)
- As far as I understand WP:RSN is the place to discuss whether a source is reliable in a certain context. I don't want to argue about these 3 cases or even this source in general. I want to define more explicitly what the current consensus position (Citations to deprecated sources should not be removed indiscriminately) means. Alaexis¿question? 16:10, 26 November 2020 (UTC)
I have grave misgivings about this proposal. But first, here is what I strongly agree with:
- A reliable source is better than an unreliable source. Everyone should be encouraged to replace unreliable sources.
- When replacing a source the claim in the article should be checked against the new source and modified as needed.
- Whether to remove the claim that is sourced to an unreliable source orb just the source is a judgement call, depending on whether it is controversial and/or likely to be challenged.
- Removing an unreliable source when a reliable replacement is easily found is a Bad Thing. This is the heart of complaints about indiscriminate removals or mass removals, and rightly so. I consider spending less than five minutes on each removal to be evidence that the editor isn't looking hard enough for a reliable replacement.
Here is where I have my misgivings.
- Any course of action that leaves the unreliable source in the article is a Bad Thing. I see zero reason for keeping existing unreliable sources while forbidding new ones.
- If you really can't find a reliable replacement, leaving in the unreliable source is The Wrong Solution. Replace it with a citation needed, and add a comment saying "[URL] removed because it is not a reliable source." That helps future editors to find a better source without keeping bad sources in articles. The argument against replacing unreliable sources with citation needed above is a red herring and easily fixed by using HTML comments. --Guy Macon (talk) 20:39, 26 November 2020 (UTC)
- I would support a special version of {{citation needed}} for deprecated sources that would allow the deprecated reference to be incorporated into that template so that it only shows on a mouse-over or source code review and not in the default page display. (CN's "reason" param does this already) The special template would be something easily found via WP search tools so that we can find where there are deprecated-CN tags and potentially by source as well in the future; we could arguably do this tagging with CN already but its making sure that deprecation is mentioned as the reason is included. --Masem (t) 20:45, 26 November 2020 (UTC)
- Ideally if done right, this would be as easy as slapping something like {{cn-d|<ref>deprecated source</ref>}} around the deprecated ref as to make it as easy as possible to mark and hide the source. --Masem (t) 20:58, 26 November 2020 (UTC)
- I would support Masem's suggestion. (t · c) buidhe 21:03, 26 November 2020 (UTC)
- This would unduly hamper searches for deprecated source usages to remove - they use "insource:", and keeping the link in there would leave them still in the backlog - David Gerard (talk) 21:34, 26 November 2020 (UTC)
- That's why I suggested a new template specific for cn's replacing deprecated sources. The template is then searchable on its own, and then you can search for what sources are used in those templates. --Masem (t) 00:18, 27 November 2020 (UTC)
- I don't oppose leaving a hidden link that is only visible in the edit window, but I do oppose leaving bad sources visible to the reader -- and that includes rollover text. --Guy Macon (talk) 00:37, 27 November 2020 (UTC)
- This problem happens if the link is present anywhere in the wikitext - David Gerard (talk) 00:44, 27 November 2020 (UTC)
- That's why I suggested a new template specific for cn's replacing deprecated sources. The template is then searchable on its own, and then you can search for what sources are used in those templates. --Masem (t) 00:18, 27 November 2020 (UTC)
- This would unduly hamper searches for deprecated source usages to remove - they use "insource:", and keeping the link in there would leave them still in the backlog - David Gerard (talk) 21:34, 26 November 2020 (UTC)
- I would support Masem's suggestion. (t · c) buidhe 21:03, 26 November 2020 (UTC)
- Ideally if done right, this would be as easy as slapping something like {{cn-d|<ref>deprecated source</ref>}} around the deprecated ref as to make it as easy as possible to mark and hide the source. --Masem (t) 20:58, 26 November 2020 (UTC)
- I would support a special version of {{citation needed}} for deprecated sources that would allow the deprecated reference to be incorporated into that template so that it only shows on a mouse-over or source code review and not in the default page display. (CN's "reason" param does this already) The special template would be something easily found via WP search tools so that we can find where there are deprecated-CN tags and potentially by source as well in the future; we could arguably do this tagging with CN already but its making sure that deprecation is mentioned as the reason is included. --Masem (t) 20:45, 26 November 2020 (UTC)
- Oppose - invalid proposal. Deprecated sources are bad sources, and they don't belong in Wikipedia.
- The purpose of deprecation is to allow generally unreliable sources to be rapidly removed in large amounts, without having to argue each and every bloomin' instance with querulous editors who've decided they're fond of this particular usage. It's for sources we've decided are such trash that they need to go.
- There's no reason to believe Sputnik ever belonged in Wikipedia - it just took formal deprecation to finally get rid of it.
- This proposal would place more restriction on removal of deprecated sources than we have on removal of less-bad sources. Keeping the little blue number when it's deceptive to the reader pretends we have sourcing when we don't, and this is bad. Taking out bad sources we literally can't trust does not in any way imply bad faith in the editor who put them there - but they're still bad sources that should be removed forthwith - there is no reason to deliberately leave a bad source in. There is especially no reason to make an assumption of WP:OWNership of the bad source, such that it has the privilege of staying in a month, when a merely "generally unreliable" source wouldn't get that privilege. Bad sourcing is as un-WP:OWNed and editable as any other edit covered in the edit notice.
- I urge you to read the discussion at VPP from December last year.
- In any case, this proposal attempts to override policy and strong guidelines - WP:V and WP:RS (which is included by reference in WP:V) will always allow editors to remove unreliable sources on sight with an edit reason of "unreliable source", even if you try, as here, to claim special protections for deprecated sources. No WP:LOCALCONSENSUS here can change that.
- This proposal is yet another attempt to hamper the removal of deprecated sources, on a claim that they are of value. They are precisely the sources that have been found, by broad general consensus, not to be of value - David Gerard (talk) 21:43, 26 November 2020 (UTC)
- I believe that there is a strong consensus to not simply delete these citations and that the editor doing the removing must make some effort to find a replacement. As I said before, I consider spending less than five minutes on each removal to be evidence that the editor isn't looking hard enough for a reliable replacement. --Guy Macon (talk) 00:37, 27 November 2020 (UTC)
- I would strongly disagree - in my removal of tens of thousands of DM and Sun cites, quite a lot were just trash, and felt like they'd stolen from me the 15 seconds it took to assess and remove them. In my direct practical experience, deprecated sources are bad, and they're usually used badly. It doesn't take 300 seconds to assess vituperative DM bile, and removal rather than replacement is often correct - David Gerard (talk) 00:41, 27 November 2020 (UTC)
- You 'strongly disagree' that there is a strong consensus against the kind of action you take part in..well, why bother with procedure then, every party in every dispute gets to be their own judge and jury, in all cases whatsoever. As for rules and regulations, every policy means whatever anyone personally says it means. For the record, there is nothing in this proposal that changes what is already written. Clarifying is not changing. This user already does precisely what the existing guideline says should not be done. Calling an effort to clarify text to a guideline you're already openly disregarding as being a 'rule change' is ludicrous. AGF is not without reasonable limits.... how much farther does this be allowed to continue until we say enough is enough? Firejuggler86 (talk) 00:19, 3 December 2020 (UTC)
- I would strongly disagree - in my removal of tens of thousands of DM and Sun cites, quite a lot were just trash, and felt like they'd stolen from me the 15 seconds it took to assess and remove them. In my direct practical experience, deprecated sources are bad, and they're usually used badly. It doesn't take 300 seconds to assess vituperative DM bile, and removal rather than replacement is often correct - David Gerard (talk) 00:41, 27 November 2020 (UTC)
- Deprecation determines that sources that were previously used should now be considered bad, but this means that their prior use may have been added in good faith or valid intentions and we should not be readily dismissing their use so flippantly. You are making the assumption that they were added in bad faith in the past retroactively, which simply is not true; no one knew the source was bad until the RFC on deprecation happened. So we have to assume that their use was in good faith and review each use with some reasonable care. There is no deadline to remove them per WP:DEADLINE, and it is clear from the existing wording on this page that removal should be done with care to avoid disruption. Flat-out removal of deprecated sources without any attempt to locate new sources or to judge how the removal affects article quality has been shown to be disruptive. Of course, there is no harm in having CN tags on uncontentious material that needs better sourcing - that's why we have those tags, and there is no issue to leave the deprecated sources as "hidden" within those tags to help editors see what the original bad source was so that a better source can be found, if that's deemed to be helpful. WP is always a work in progress so these do no harm. It would be a problem if they persistent into a FAC or the like, but that would be a sourcing problem even there. --Masem (t) 00:16, 27 November 2020 (UTC)
- And you care whether the person who originally added the source did so in good faith... why? Yes. We should "dismiss their use so flippantly" now that we know that they are bad. It's not like someone is criticizing editors for using bad sources before we figured out that they were bad. You are correct that "flat-out removal of deprecated sources without any attempt to locate new sources is disruptive" but you are wrong in asking that the removing editor "judge how the removal affects article quality". Removing bad sources always improves article quality. There are no exceptions. The only place where the removing editor needs to "judge how the removal affects article quality" is when deciding whether to delete the bad source or delete the bad source along with the claim it supports. --Guy Macon (talk) 00:37, 27 November 2020 (UTC)
- There are exceptions! For example, in an article about a controversial topic I wrote "The tweets were first republished in an article in FrontPage Magazine on July 30, 2014.[ref to FrontPage Magazine article in question]" A user removed that ref (without replacing it). I didn't want to fight so I replaced it with a ref to another news site that contained some quotes from FrontPage Magazine. The only way to claim that the user's edit improved the article is to axiomatically claim that "all deprecated sources are bad so removing them is good." Which misses the point because a reader of that article clearly would have benefited from a link to FrontPage Magazine so that they could have found out for themselves what it wrote about the tweets in question. ImTheIP (talk) 03:52, 27 November 2020 (UTC)
- I've seen edits like this, too. We have editors who seem to believe that RSP overrules what WP:V says ABOUTSELF. WhatamIdoing (talk) 05:22, 16 December 2020 (UTC)
- While removing bad sources (and text that is only attributed to bad sources) is always a good thing, it is as equally important on the WP:5P as avoiding disruption to the work. Sources that have been in place prior to the deprecation determination, as this policy already states, should not be immediately treated as suspect but should be carefully reviewed to determine appropriate action as to avoid disruption. To consider a similar issue, what happens when we find a user with numerous seemingly good edits (say in the thousands) has been a sock of a previously banned user? Do we immediately revert all those edits? No we don't at that scale, but we do make it clear that the edits should be reviewed for appropriateness and can be kept if other editors feel they are okay, but can be removed if they are suspect. Same type of process and approach that we should be applying to deprecated sources. A careful, human review to figure out the lowest-disruptive way to proceed. It's why "deprecated" is used and not "blacklisted" or "banned", which would require the immediate stripping of said sources regardless of what it who do the articles. The attitude that the only option for deprecated sources that have long been used in articles is to remove them with no attempt to seek replace or let editors seek replace is a disruptive option. --Masem (t) 04:07, 27 November 2020 (UTC)
- There are exceptions! For example, in an article about a controversial topic I wrote "The tweets were first republished in an article in FrontPage Magazine on July 30, 2014.[ref to FrontPage Magazine article in question]" A user removed that ref (without replacing it). I didn't want to fight so I replaced it with a ref to another news site that contained some quotes from FrontPage Magazine. The only way to claim that the user's edit improved the article is to axiomatically claim that "all deprecated sources are bad so removing them is good." Which misses the point because a reader of that article clearly would have benefited from a link to FrontPage Magazine so that they could have found out for themselves what it wrote about the tweets in question. ImTheIP (talk) 03:52, 27 November 2020 (UTC)
their prior use may have been added in good faith
This does not mean they were good then either, and is no reason not to remove them - the relevant policy is no ownership of articles.You are making the assumption that they were added in bad faith
No, I have at no point done so. However, they are bad sources; deprecation makes it clear that they are bad sources.There is no deadline
However, that is no reason to hamper their removal, as this proposal tries to.has been shown to be disruptive
You have raised this claim repeatedly, and every example I recall, consensus has failed to agree with you - please point me to a discussion where the consensus has been that removal of a deprecated source was disruptive.- You have completely failed to engage with the policy issues of this proposal - David Gerard (talk) 00:41, 27 November 2020 (UTC)
- Unless you have a time machine to deprecate these sources before they were added, we must assume they were originally added in good faith, by definition of good faith. Editors adding those sources may have had no idea that in future we'd deprecate the source, so there's zero reason to punish those edits by calling them bad now. And the fact that numerous discussions have come up when editors have begun removing deprecated sources en masse shows that other editors consider this disruptive. There's been no discussion to conclude that definitively this is or is not disruptive, but that's why this thread appears to be open, because it still is being considered disruptive. We're trying to find the collaborative solution here, and that's where the stance "nope, gotta get rid of them all" is not helping. --Masem (t) 04:13, 27 November 2020 (UTC)
- And you care whether the person who originally added the source did so in good faith... why? Yes. We should "dismiss their use so flippantly" now that we know that they are bad. It's not like someone is criticizing editors for using bad sources before we figured out that they were bad. You are correct that "flat-out removal of deprecated sources without any attempt to locate new sources is disruptive" but you are wrong in asking that the removing editor "judge how the removal affects article quality". Removing bad sources always improves article quality. There are no exceptions. The only place where the removing editor needs to "judge how the removal affects article quality" is when deciding whether to delete the bad source or delete the bad source along with the claim it supports. --Guy Macon (talk) 00:37, 27 November 2020 (UTC)
- I believe that there is a strong consensus to not simply delete these citations and that the editor doing the removing must make some effort to find a replacement. As I said before, I consider spending less than five minutes on each removal to be evidence that the editor isn't looking hard enough for a reliable replacement. --Guy Macon (talk) 00:37, 27 November 2020 (UTC)
- I don't see how removing a bad source is at all different from any other edit when it comes to the feelings(?) of the original editor. If anything, other people rewording or entirely rewriting/deleting a paragraph is far more "punishing" to such editors since it is explicitly deeming their creative contribution inadequate. If editors are so sensitive to criticism, and place so much ownership over their edits that they would even notice a change to a particular sentence, I think the issue is actually one of competence. Are there really that many (any?) editors who are active enough to have a watchlist, or obsessive/possessive enough to regularly check the reflists of articles they edit, that they even catch a ref removal, let alone feel slighted by it?
- I'm also really puzzled by the claim that replacing a deprecated source with [citation needed] somehow makes it harder to find RS substitutes? Is this in comparison to tagging the ref with [deprecated source?], which also alerts other editors to the need for a replacement ref while preserving the shitty source? Because I find it hard to believe the content provided by the bad source consistently makes it that much easier to track down alternatives. JoelleJay (talk) 08:52, 27 November 2020 (UTC)
- Just a point of order but "The problem with Sputnik is that there is a lot of Russia propaganda there, not that that they get random facts wrong.” isn't accurate, disinformation issues aside Sputnik is a low quality source with very little in the way of real journalistic credentials which regularly gets random facts wrong. I don’t really see the point in making a stand for a source thats both an active publisher of disinformation *and* low quality gutter trash which does almost no real investigative reporting. If this is the best hill you can find to die on this stand won’t last long. At least choose a deprecated source which produced something other than churnalsism, for example CGTN. Horse Eye's Back (talk) 22:26, 26 November 2020 (UTC)
- @Amigao: User:Amigao has deleted the use of Sputnik as a source for the following:
- Russian Prime Minister Dmitry Medvedev was reportedly critical of the TPP, saying that "the WTO is being encroached upon" and this might lead to the "destruction of world trade".[1]
- I accept that Sputnik may be unreliable when reporting on anything the West might be doing. However, I'd be shocked if they would be off base in describing the position of the Russian Prime Minister. The designation of a source as "Deprecated" should not be used as an attack on Russia or on things Russian.
- Beyond that, I'm deeply disturbed by the use of a "Deprecated sources" as a form of censorship. My concern is that Wikipedia has a big problem recruiting volunteer editors, and deleting a source like this without a discussion of what's credible seems like an attack on whomever cited that source.
- Rather than deleting references to outlets like Sputnik, I think we should be entering into discussions about what is and is not credible given the available evidence.
- I wish to discuss an example that has had a big influence on my life: I was raised in the US during the early days of the Cold War. I spent 6 years in the US military during the US war in Vietnam. Former US President Eisenhower said, "I have never [communicated] with a person knowledgeable in Indochinese affairs who did not agree that had elections been held as of the time of the fighting [leading to the defeat of the French in 1954], possibly 80 per cent of the population would have voted for the Communist Ho Chi Minh".[2]
- This was the universal expert consensus that was not even mentionable in the mainstream media of that day.
- That book by Eisenhower was published in 1963. I first heard mention of it in 1969, but I didn't have a sufficient reference to actually read the source until much later. I was skeptical. Finally in 2005 I got the book. Since then I've been able to talk about it. Roughly 57,000 US military personnel and 6 million Southeast Asians died, essentially because the mainstream media in the US could not report honestly about the situation in Southeast Asia. For a comparable discussion of the War on Terror, see "Wikiversity:Winning the War on Terror".
- I think we need more honest and open discussion of the portrayal of different perspectives. Deleting sources claiming they are "Deprecated" to me turn opportunities for dialog and building bridges into barriers to communications and understanding. This seems to me to be the opposite of what Wikipedia could be and should be.
- However, I get the impression from reading some of the above discussion that I perhaps should be making these arguments in some other forum.
- Where do you suggest I try to make these arguments?
- Thanks, DavidMCEddy (talk) 02:57, 27 November 2020 (UTC)
- This sounds like you're disagreeing with the categorization of Sputnik as depreciated, not that you're arguing over how depreciated sources can be handled. A potentially-contentious statement of fact about a position taken by a living person is obviously not something that can be cited to a depreciated source under any circumstances (it couldn't even be sourced to an unreliable source.) Regarding your opinion that you would be
shocked if they would be off base in describing the position of the Russian Prime Minister
, you're welcome to your opinion, but the entire meaning of depreciation was that its history of publishing factual inaccuracies means that no, we do not trust it to accurately portray the past and present positions of the Russian Prime Minister. While I can't speak to Sputnik specifically, plenty of depreciated sources would do things like misrepresent the opinions and positions of people they supported, or people with the power to control their output, in order to make them sound better - this sort of misinformation allows politicians and others to portray the most "favorable" vision of themselves at all times. In the US, comparable outlets will outright misrepresent the positions of politicians they support in order to allow them room to take contradictory actions or ones that are actually unfavorable for their constituents. So no, the political positions held by the Russian Prime Minister in particular are something that should not be cited to a source that was depreciated for publishing deliberate falsehoods in the service of the Russian government. --Aquillion (talk) 10:30, 29 November 2020 (UTC)- @Aquillion:
the entire meaning of depreciation was that its history of publishing factual inaccuracies
. Theoretically, yes, but not in practice. The wording used in RfCs for the deprecation option isOption 4: Publishes false or fabricated information, and should be deprecated as in the 2017 RfC of the Daily Mail
. However, if you look through RfCs that have led to deprecations, the discussions often have little (or nothing) to do with publication of false or fabricated information. They often have to do with concerns about the tone of editorials, selective coverage, etc., but no pattern of publishing flat-out false information is established. It seems that "Option 4" (i.e., deprecation) has come to be a general stand-in for "we don't want this source to be used", regardless of what the text "Option 4" actually says. Later on, the wording of the RfC gets trotted out to claim that these sources publish fabricated information, even though the original deprecation discussion had nothing to do with fabricated information. For many of these sources, the arguments that editors make in the RfC would logically lead one to believe thatOption 2: Unclear or additional considerations apply
is the correct designation, but "Option 4" has somehow come to mean, "I emphatically disagree with Option 1". -Thucydides411 (talk) 15:04, 29 November 2020 (UTC)- Again, it feels like you're arguing against the individual depreciation RFCs by saying that people should not have pushed for depreciation or that the RFCs shouldn't have been closed as depreciation (because the text of the RFC makes the meaning of depreciation clear.) If you feel that way, the route to take is to try and overturn those individual RFCs, not to try and reverse them by changing the definition of depreciation. My perception is that the people who made those arguments in those RFCs generally felt the points they made ultimately showed that the sources were publishing false or fabricated information, and that the RFC closer agreed, notwithstanding your opinion that they were focusing too much on tone, bias, and selective coverage. My personal opinion is that those things, while not sufficient for depreciation on their own, are a vital piece of the argument for depreciation because they are what turns "keeps making sloppy mistakes" into a pattern of false or fabricated information, ie. when a source with a clear POV makes repeated "errors" that favor their POV, that is more serious than a source whose errors are scattershot and don't point in a particular direction, because it suggests that their inaccuracies are the result of overarching intent and institutional policy. I would argue that there is, additionally, a point where sufficiently selective coverage produces a result that is, collectively, false; a source that disregards context entirely or presents facts in an intentionally deceptive fashion is, in fact, producing false or fabricated information (eg. the "Dihydrogen Monoxide" hoax falls under false and fabricated information because it intentionally leads the reader to a false conclusion even if it does so using actual facts.) Since you brought up editorials (and I don't think what people mention about them is actually limited to tone), I also disagree with what I read as your implicit assertion that sources cannot be held to account for errors in editorials or opinion pieces - WP:RSOPINION is not a blank check, and requires a degree of reliability, as well as the basic WP:RS reputation for fact checking and accuracy (you cannot cite an opinion to a Reddit thread, after all.) A source that uses no editorial controls or fact-checking for its opinion pieces at all, and therefore frequently publishes opinion pieces containing patiently false or fabricated information, cannot be cited under WP:RSOPINION and ought to ultimately be depreciated for opinion purposes - that is to say, treated the same as a self-published source, one where publication grants no reliability or weight. --Aquillion (talk) 23:30, 29 November 2020 (UTC)
- @Aquillion:
- This sounds like you're disagreeing with the categorization of Sputnik as depreciated, not that you're arguing over how depreciated sources can be handled. A potentially-contentious statement of fact about a position taken by a living person is obviously not something that can be cited to a depreciated source under any circumstances (it couldn't even be sourced to an unreliable source.) Regarding your opinion that you would be
Let's keep the discussion focused. We know there are serious issues with deprecated sources. My point is that mass removals often do more harm than good. WP:RS#Deprectaed sources does not say that these references must be removed en masse so the proposal does not contradict this policy. The fact is that automated such removal is disruptive, see for example the talk page of a user who is doing it, there are a lot of users who believe that the removals have been unjustified, and these are just things someone cares enough to complain about.
If you believe that all deprecated sources are universally terrible and should never be used on Wikipedia, replacing a link to a deprecated source with a {{cn}} should actually worry you a lot. If just the reference supporting certain statement is removed, then the (potentially inaccurate) statement itself remains, and can live for years with a {{cn}} tag. This is another reason to spend a few minutes on each case and either remove the both the reference and the statement which it supports, or find a reliable source for it.
I like the idea of using a special tag when removing deprecated sources en masse, which would, on one hand, not show deprecated source to the reader but keep them in the code. Alaexis¿question? 09:49, 29 November 2020 (UTC)
- @Alaexis: What automated removal? Can you provide diffs? Nobody else has mentioned those Horse Eye's Back (talk) 14:42, 29 November 2020 (UTC)
- I was referring to removals at the rate of 3 changes per minute which means that the editor could not have possibly reviewed them. You can look at the examples I gave in the beginning and check the contributions of that user. So technically not automated but for all intents and purposes this is equivalent to a bot replacing all such references by cn tags. Alaexis¿question? 19:08, 29 November 2020 (UTC)
- I strenuously oppose any attempt to add red tape or restrictions to the removal of depreciated sources. The purpose of depreciating a source is to ultimately get it out of the encyclopedia, fullstop (outside of the very few cases outlined in WP:ABOUTSELF), and any effort to delay or interfere with that is unacceptable. The whole idea of adding red tape to depreciation removals inverts the concept of deprecation - most of these suggestions would make it harder to remove a depreciated source than an unreliable one (which, after all, has no such red tape; any unreliable source can be replaced by a {{cn}} on sight, and doing so, if it is genuinely unreliable in context, is always an improvement.) To go over a few of the specific points made here:
- I deeply oppose any suggestion that would leave the depreciated source as a comment in the article source - that is an awful idea on several levels. Comments should be kept to a minimum and should be reserved for useful things; a comment pointing to an unusable source is not useful. And how long would such worthless comments clutter article text?
- Since depreciated sources cannot be used (outside of very limited WP:ABOUTSELF exemptions, which do not apply to the examples listed above), replacing a depreciated source outside those situations with a {{cn}} tag is an admirable improvement to the article and ought to be encouraged as much as possible. If you are unaware, a {{cn}} tag not only adds a visible indicator on the page showing that a source needs to be added, it also adds the article to the relevant category; and, of course, the fact that it is visible to the editor alerts all readers that the statement lacks a citation. It's baffling that you would suggest that this makes it more likely that the statement will remain unsourced - since a depreciated source is never a valid source, it was uncited before, but this was not obvious to casual readers and did not place it in the proper category. A {{cn}} tag both vastly increases the chance that the situation will be corrected, and reduces the damage for as long as it goes uncorrected, since it avoids fooling casual readers into thinking that the statement has a valid source (the way a depreciated source does as long as it remains present.)
- All the removals listed in the OP were appropriate and the replacement with a {{cn}} tag was entirely correct; the idea that they were "uncontroversial uses" goes against the whole point of depreciation. The purpose of depreciation (and the thing that separates it from merely being generally unreliable) is that there are essentially no situations in which a depreciated source counts as 'published' for WP:V purposes; they can occasionally be used, but only under the exemptions (such as WP:ABOUTSELF) that would apply to a blogpost or the like, since they have the same status. While you say you support the depreciation of Sputnik, the argument that they can be trusted for simple facts belies this - a source that has repeatedly and intentionally published misleading material cannot be trusted even for things that, at first glance, appear uncontroversial. If they could be trusted for such things, they would merely be generally unreliable and not depreciated. A depreciated source cannot be trusted for anything. Fullstop. That's the basic meaning of depreciation - by default, we extend them no trust whatsoever on any topic. They can be used, in certain limited situations, but generally only in the contexts where trusting them is not required, which absolutely does not include citing them for stuff that an editor feels is "uncontroversial." If you think they can be trusted for those things, argue to overturn their depreciation, don't try to redefine depreciated to just mean another type of generally-unreliable.
- Removing depreciated sources is time-consuming work, and the people who devote their time to it deserve our support, not this sort of suggestion for poorly-considered red tape. Yes, there are many different ways to handle removal, and which one is used should depend on the situation, the source, and so on; but nothing is gained by adding additional rules or restrictions to those removals, because virtually anything is an improvement over leaving a depreciated source in place. Furthermore, none of the people making these perennial suggestions have actually produced anything to support the idea that the mass-removal of depreciated sources are disruptive. I see only a handful of people objecting, mostly on fairly weak grounds - and obviously it's a given that some people will object to the removal of depreciated sources; that's the whole reason we depreciate sources in the first place. If it was completely uncontroversial among all editors that a source is always unusable, we wouldn't need to depreciate it. --Aquillion (talk) 10:30, 29 November 2020 (UTC)
- 100% agree with Aquillion (even though their spelling of "deprecated" suggests source reliability has a fixed initial value that gradually decreases over time... ;)). Deprecated sources should not be on wikipedia–they give readers and editors the impression a statement is sourced when it is not and should always be replaced on-sight with [citation needed] tags if a replacement can't be found immediately. JoelleJay (talk) 18:22, 29 November 2020 (UTC)
- Adding a {{cn}} tag is, in my considerable experience, one of the few things that ever gets a replacement source found. Various other tagging mechanisms observably don't. Note that this is describing what actually happens.
- We don't need more red tape on removal of deprecated sources. They were deprecated because they are unfit for use in Wikipedia - that's what it means - David Gerard (talk) 10:34, 29 November 2020 (UTC)
- oh, looking through those cited talk page removals, they appear to be people who also seem, like you, to think deprecated sources are good actually. The point is that they aren't.
- And I concur: if you're literally accusing other editors of
automated removals
, that's quite an accusation. You need to produce diffs or withdraw the claim - David Gerard (talk) 15:15, 29 November 2020 (UTC)
- People who remove sources need to be careful about it - one problem is where a statement is sourced to multiple sources, one of which is depreciated - people tend to just removes the depreciated statement without checking whether the remaining sources actually back up the whole of the statement or just part of it - in this case you MUST check before simply removing the source - and if necessary use Cn span or tag the depreciated source with [unreliable source?] and flag it up on talk pages and/or appropriate wikiprojects so that somebody who has access to the reliable sources which things like Sputnik will be based upon - which will often be Russian language sources for Sputnik - can fix it.Nigel Ish (talk) 11:32, 29 November 2020 (UTC)
- So you're doing this yourself, yes? - David Gerard (talk) 15:17, 29 November 2020 (UTC)
- @Nigel Ish: "one problem is where a statement is sourced to multiple sources, one of which is depreciated - people tend to just removes the depreciated statement" Now thats interesting and would be an issue, but I’ve never personally seen it happen. Can you provide a few diffs? Horse Eye's Back (talk) 15:23, 29 November 2020 (UTC)
- [4], [5], [6], [7] are a few from the last couple of days - not that removing sources from several articles per minute suggests that the editors aren't closely studying the articles.Nigel Ish (talk) 15:39, 29 November 2020 (UTC)
- @Nigel Ish: I’m confused, in none of those diffs was the statement (or for that matter any text at all besides the citation) removed. Horse Eye's Back (talk) 15:58, 29 November 2020 (UTC)
- The point is that if someone just removes the depreceated source without checking whether the remaining sources back it up, then they may be misrepresenting what the remaining sources actually say. If the sentence say a and b, when a reliable source says a and the bad source says b, then if you just remove the bad source, then the article is then stating that the good sources says b when it doesn't. But as it is clear that the opions of anyone who disagrees with David Gerard is not welcome here, I will withdraw from this discussion.Nigel Ish (talk) 16:17, 29 November 2020 (UTC)
- Thats a very different point from the one we were discussing. Isn't what you’re now describing WP:SYNTH? In most contexts all the sources have to support the entirety of the text they’re being used to support. If removal of one of a number of sources so massively changes whether or not the information is sourced then shouldn't the citations have been put on the information they individually supported in the first place rather than being grouped together? David Gerard might be being a bit aggressive but their point is a good one, we seem to have a lot of doctors writing prescriptions they themselves don't take. Horse Eye's Back (talk) 16:20, 29 November 2020 (UTC)
- The point is that if someone just removes the depreceated source without checking whether the remaining sources back it up, then they may be misrepresenting what the remaining sources actually say. If the sentence say a and b, when a reliable source says a and the bad source says b, then if you just remove the bad source, then the article is then stating that the good sources says b when it doesn't. But as it is clear that the opions of anyone who disagrees with David Gerard is not welcome here, I will withdraw from this discussion.Nigel Ish (talk) 16:17, 29 November 2020 (UTC)
- You haven't replied yet as to whether you yourself are fixing deprecations using the procedure you're prescribing for others - David Gerard (talk) 16:06, 29 November 2020 (UTC)
- @Nigel Ish: I’m confused, in none of those diffs was the statement (or for that matter any text at all besides the citation) removed. Horse Eye's Back (talk) 15:58, 29 November 2020 (UTC)
- [4], [5], [6], [7] are a few from the last couple of days - not that removing sources from several articles per minute suggests that the editors aren't closely studying the articles.Nigel Ish (talk) 15:39, 29 November 2020 (UTC)
- @Nigel Ish: "one problem is where a statement is sourced to multiple sources, one of which is depreciated - people tend to just removes the depreciated statement" Now thats interesting and would be an issue, but I’ve never personally seen it happen. Can you provide a few diffs? Horse Eye's Back (talk) 15:23, 29 November 2020 (UTC)
Arbitrary break (guideline clarification)
edit- An issue at the base of this is actually what "deprecated" means for WP as in most places it is considered a phasing out of something that was once accepted; it does not mean that a sudden transition. If this was a computer system, say if we were programming in Java and a function was said to be deprecated with a given release, it would mean that while the function would still be supported in that version, it may not be supported in the next. A development team would not necessarily to automatically replace that function with a different version but would likely need to make sure each replacement makes sense in context and doesn't break anything else. Myself (and I feel a number of other editors here) as well as this page are treating deprecation in this fashion. On the other hand, we clearly have editors that feel that when sourcing have been labeled deprecated by an RFC, they must be removed ASAP. Yes, this is a Good Thing (TM) at the end of the day, particularly when we're talking something as bad as the DM due to its fabrications, but we're still talking "deprecation" and not a blacklist, and we have to balance the needs of the verifyability pillar with that of the collaborative editing/minimize disruption pillar equally.
- To that point, perhaps the problem is that with some of the sources that have been deprecated, namely like the Daily Mail where we deprecated them due to actually falsifying details, where having that source lingering around is a rather bad thing and there is a strong reason to remove it faster than other deprecated sources that just have bias issues that may not be factually wrong but are questionable for quality sourcing (like RT, Fox News, etc.) That is, simply calling something like DM "deprecated" is too weak a term and really we should have blacklisted it with exceptions as outlined at the RFC when it itself is part of the news story. Similar for any other word known to actually falsify information (and thus where we have a good chance to be reporting false information in Wikivoice). Whereas for most other deprecated sources we have, we're concerned their bias may have given us poor information (particularly on BLPs) and thus we do want to replace with better sources or remove the information if no replacement can be made, but again, going back to the concept of what "deprecated" means and considering the balance needed in the pillars, this should not be a rushed job save for BLP pages where verifyability is a higher priority over most of these other concerns.
- In other words, we may simply want to call sources that falsify details like the DM as simply blacklisted or banned sources with noted exceptions, rather than treating them as deprecated sources. This gives better reason to take more aggressive approaches to their removal (in bulk and without having to worry about leaving CNs behind) which few would debate. The other remaining deprecated sources, which were deprecated for their questionable biases and other factors, then should have a reasonable approach to balance their removal with minimizing disurption as this page otherwise suggests. --Masem (t) 16:24, 29 November 2020 (UTC)
- But even under the focused editing at issue here it still takes months or years to remove all links to a deprecated source. We are not currently talking about sudden transitions. Horse Eye's Back (talk) 16:26, 29 November 2020 (UTC)
- It should take months or years to remove these links, under the notion of deprecation as a phased transition, particularly in a volunteer project, which is why there's concern when deprecated sources are just being removed without attempts to find replacements or leave behind indicators that replacements are needed. When we have a situation like the DM which I fully agree is a dangerous source to leave around with nearly no exceptions, that requires more urgency and why just using the "deprecation" title is probably the wrong term for it; its a different class of problem to deal with. --Masem (t) 19:03, 29 November 2020 (UTC)
- This is a version of "deprecation" that you have made up. It is nowhere to be found in the findings of deprecation RFCs. Where did you get this bizarre idea that it "should" take months or years?
- Your attempt to try to play with words to make "deprecation" mean what you would like it to is not convincing - David Gerard (talk) 19:23, 29 November 2020 (UTC)
- Because that's what the word means in the English language and how deprecation is to be treated via this page. The problem is that when we applied to the Daily Mail, where there is a stronger case for getting rid of DM links far sooner than later, it gives the wrong impression that that also applies to all other sources labels deprecated. --Masem (t) 01:01, 30 November 2020 (UTC)
- This is your idiosyncratic interpretation. The word is used here to mean that they are bad sources and should not be used in Wikipedia at all. This implies their removal. The rest is you making stuff up from nowhere - David Gerard (talk) 12:56, 30 November 2020 (UTC)
- I don't think that's an idiosyncratic interpretation. Dictionary definitions such as "to withdraw official support for or discourage the use of (something, such as a software product) in favor of a newer or better alternative" do not suggest to me that the process needs to follow any particular timeline, and certainly not that it's necessary to remove the discouraged or officially unsupported thing as rapidly as possible.
- To look elsewhere in the Wikimedia movement for what we mean with that word, when the devs deprecate software features, they often take multiple years to move from issuing warnings about it being deprecated to blocking it. WhatamIdoing (talk) 05:31, 16 December 2020 (UTC)
- This is your idiosyncratic interpretation. The word is used here to mean that they are bad sources and should not be used in Wikipedia at all. This implies their removal. The rest is you making stuff up from nowhere - David Gerard (talk) 12:56, 30 November 2020 (UTC)
- Because that's what the word means in the English language and how deprecation is to be treated via this page. The problem is that when we applied to the Daily Mail, where there is a stronger case for getting rid of DM links far sooner than later, it gives the wrong impression that that also applies to all other sources labels deprecated. --Masem (t) 01:01, 30 November 2020 (UTC)
- I don't see any reason why intentionally making it take "months or years" to replace depreciated sources would be a good thing; the faster we can get them out, the better. Taking time to put each broken part of the wiki in the best possible state is commendable from a WP:SOFIXIT perspective, but simply replacing a depreciated source with a {{cn}} tag is almost always an improvement, and I'm not seeing anyone giving any real examples above where it caused problems that are worse than (or even as bad as) the very real and serious issues cause by leaving large numbers of depreciated sources in place. The reality is that a depreciated source harms Wikipedia's reputation while giving casual readers the mistaken impression that a statement is sourced when it is not; a {{cn}} tag is more likely to be fixed and, until it is, more accurately warns a reader that the statement it is attached to should be viewed with skepticism. The people who feel that more time and effort should be spent putting each bad source in the best possible state would be better off spending their time replacing {{cn}} tags with valid sources, rather than throwing red tape in front of people who are putting in the time-consuming work of implementing RFC outcomes and improving the wiki's sourcing. Category:All articles with unsourced statements could use the time and energy that is wasted on this still-terrible perennial suggestion. --Aquillion (talk) 23:47, 29 November 2020 (UTC)
- It should take months or years to remove these links, under the notion of deprecation as a phased transition, particularly in a volunteer project, which is why there's concern when deprecated sources are just being removed without attempts to find replacements or leave behind indicators that replacements are needed. When we have a situation like the DM which I fully agree is a dangerous source to leave around with nearly no exceptions, that requires more urgency and why just using the "deprecation" title is probably the wrong term for it; its a different class of problem to deal with. --Masem (t) 19:03, 29 November 2020 (UTC)
- But even under the focused editing at issue here it still takes months or years to remove all links to a deprecated source. We are not currently talking about sudden transitions. Horse Eye's Back (talk) 16:26, 29 November 2020 (UTC)
- I'll give you reasons for not allowing wholesale deletion of deprecated sources:
- * Wholesale deletion of references and comments supported by deprecated references is a quicker way to get rid of "bad" sources. However, how do you know they are "bad"? Daniel Kahneman won the 2002 Nobel Memorial Prize in Economics in part by establishing that virtually everyone thinks they know more than they actually do.
- * It's laudable to eliminate "bad" sources from Wikipedia. However, I've been drawn to this discussion by the deletion of a reference to Sputnik that was cited for a comment by Medvedev. Sputnik may be generally unreliable. It many publish pure propaganda. However, everyone who works there doubtless knows that if they published something that substantively displeased Medvedev or Putin, they might not see another sunrise. I would naively think it would be very reliable for things like that.
- * I think Wikipedia's most important contributions so far may have been its contributions to conflict resolution. A 2006 article for the Canadian Library Association noted that for controversial topics, "the two sides actually engaged each other and negotiated a version of the article that both can more or less live with. This is a rare sight indeed in today’s polarized political atmosphere, where most online forums are echo chambers for one side or the other."[3] Similarly, Shi et al. concluded that 95 percent of Wikipedia articles could benefit from engaging more editors from more diverse perspectives. Five percent suffered from an excess of conflict. This was based on a content analysis of all the edits to English-language Wikipedia articles relating to politics, social issues and science from its start to December 1, 2016, representing approximately 5 percent of the English Wikipedia.[4] Deleting a source like that without discussion on the associated "Talk" page can do more harm than good to the mission of Wikipedia: It squanders an opportunity for dialog with people with different perspecties in the name of "efficiency".
- * How many moles of the CIA, Mohammed bin Salman, Recep Tayyip Erdoğan or Xi Jinping might it take to completely destroy the integrity of Wikipedia in deleting sources they were able to get deprecated? This may sound silly, but I spent 6 years in the US military during the Vietnam War, essentially because then-president Dwight D. Eisenhower knew that he might not survive politically if he allowed honest elections to occur there in 1956. Similarly, there is ample evidence that the primary recruiters for Islamic terrorism are Saudi Arabia and the United States. That works, because the mainstream media create the stage upon which politician read their lines, and too few people seek alternative sources of information. I don't want to give that much power to a self-selected group of editors who decide what sources are "deprecated".
- Or am I barking up the wrong tree? DavidMCEddy (talk) 00:43, 30 November 2020 (UTC)
- We know they are bad sources because we, Wikipedia, have determined by consensus that they are bad. This (consensus) is a fundamental, non-negotiable aspect of Wikipedia. The alternative is to allow anything and everything into mainspace without moderation -- after all, if a self-selected group of volunteers deciding "what is RS" is too Orwellian or cabalistic, then who are they to say what statements are WP:DUE, or what is NOTABLE, or how NEUTRAL an article is, or what constitutes WP:SPAM or WP:PLAGIARISM or WP:OR?
- As others have mentioned, the issue with citing a deprecated source (e.g Sputnik) for "uncontroversial" information on a subject they are not independent of/are strongly biased toward is just as bad as citing it for info on their opponents. We sure don't trust North Korean media for truthful assessments of Kim Jong-un, for example.
- I do not follow the reasoning behind the latter two paragraphs. Wikipedia shouldn't deprecate sources because CIA moles might be flooding the RFCs? Alt-news should be permitted solely by virtue of not being mainstream (even when it consistently publishes demonstrably false disinformation)? All I will say is this is not the appropriate place for conspiratorial nonsense. JoelleJay (talk) 03:33, 30 November 2020 (UTC)
- Technically, we have determined by consensus that they are generally bad and undesirable, and not that a direct quotation taken from a now-deprecated source cannot be "relied upon" to determine whether those exact words were actually published in that source. WhatamIdoing (talk) 05:35, 16 December 2020 (UTC)
- All the RFCs did was to label the sources as deprecated and implement any appropriate edit filters for that; it did not give any editor the okay to start removing sources beyond the principles outlined on this page (which call for a careful nuanced approach, not the rush job that has been demonstrated in discussion above and before). Deprecation is meant to be a phased transition, and while we take immediate steps to prevent new uses of deprecated sources for good reasons, we should be treating old uses as if they were originally added in good faith and with valid intent but should be implicitly questionable sources that should be replaced -- excluding the case of sources like the DM which I outlined are downright harmful. Because WP is a work in progress and there is no deadline to get it right, fixing questionable sources should be a reasonable high priority but it should be done in a manner to be disruptive. Arguably, we have many many many more pages with completely unsourced material that should be fixed first if we're talking sourcing issues (no source being worse than questionable sources per WP:V) but obviously that's not the focus here, nor am I saying that should be the focus. These should clearly fixed in time, and I commend if there's a concerted effort to do in a careful manner that seeks to try to find replacement sources and/or make sure removal of sources and material that could only be sourced to that material be done in a matter that is not disruptive and respects the fact that deprecation is a phasing approach, not a "do or die" pact. Again I do stress there are unique cases like the DM that need to be treated differently where because the RFC determined that the source fabricated material and thus should not be used, that gives a strong reason to remove more quickly and without the care of what mess may be left behind. Hence why I feel by conflating the DM situation with other sources that we have deprecated in other fashions, its messing up how to discuss the right ways to go about resolving deprecated sources. --Masem (t) 01:01, 30 November 2020 (UTC)
- The 2017 RFC says
Volunteers are encouraged to review them, and remove/replace them as appropriate.
You don't think removal is appropriate, and repeat your arguments over and over; consensus is consistently against you on this - David Gerard (talk) 12:59, 30 November 2020 (UTC)- The nature of this dispute appears to include substantiated doubts as to whether those volunteers are actually "reviewing them" and handling the situations "as appropriate". WhatamIdoing (talk) 05:35, 16 December 2020 (UTC)
- If you have an accusation to make, bring diffs. If you can't bring diffs, don't make the accusation - David Gerard (talk) 08:07, 16 December 2020 (UTC)
- I'm not accusing any individual of bad editing. I'm making an observation about what this dispute, right here on this page, is about.
- The pattern I'm seeing on this page, is that multiple editors say things like "I've seen some patterns of editing that concern me", and then you say "You're accusing people! Bring diffs or it never happened!"
- The main problems with the "bring diffs" argument (aside from making everyone else feel like you think they're untrustworthy liars, or perhaps more cynically, that you are protesting too much to not have seen this yourself) are that:
- It leads to arguing about the individual specific edit (=not the overall pattern).
- It unfairly singles out individual editors (=not the overall effect caused by our potentially inadequate advice to those editors).
- Since you keep demanding diffs, I'm going to give you three diffs with an explanation of why these three concern me. I challenge you to not focus on the individual edits or the individual editor, and instead think about how the community could change its advice, so that the concerns I identify here would be addressed.
- Three diffs: [8][9][10]
- Concerns:
- Note that the timestamps prove that these three edits, to three different articles, were made in the space of less than 60 seconds.
- This was part of a larger editing run that did nothing except replace the cited sources with Template:Citation needed. Logically, there are only three possibilities:
- the editor negligently blanked the sources without reviewing the statements,
- the editor skipped any articles where the source should have been kept, or
- the source was never appropriate in any article (not even appropriate enough to keep around for a brief time with a tag on it).
- As evidence against option 2.3, in the first diff, the editor removed a source that supports a statement about what a person said. The source, although thoroughly biased, is actually reliable in WP:RSCONTEXT for that particular statement, and now we have an indirect quotation from a named BLP with zero source behind it. Quotations are WP:MINREF material; they are singled out both the nutshell and lead of WP:V as well as the lead of BLP as absolutely requiring an inline citation, with no exceptions. So in that edit, either the whole thing should have been removed, a different source should have been substituted immediately, or the deprecated source should have been kept as a temporary measure. There is no policy-compliant option that results in only removing the source.
- As evidence against option 2.2, the steadiness of the editing rate and the uniformity of the edits during this run makes me doubt that any articles were skipped.
- Please remember my challenge to you: Don't focus on what this individual did on this one day. Think about how we could support all editors in getting these sources out of all (or at least almost all) articles in a thoughtful, context-sensitive way, and especially so that our zeal to remove the bad sources doesn't accidentally result in violations of WP:V and WP:BLP. WhatamIdoing (talk) 16:57, 16 December 2020 (UTC)
- If you have an accusation to make, bring diffs. If you can't bring diffs, don't make the accusation - David Gerard (talk) 08:07, 16 December 2020 (UTC)
- The nature of this dispute appears to include substantiated doubts as to whether those volunteers are actually "reviewing them" and handling the situations "as appropriate". WhatamIdoing (talk) 05:35, 16 December 2020 (UTC)
- The 2017 RFC says
- All the RFCs did was to label the sources as deprecated and implement any appropriate edit filters for that; it did not give any editor the okay to start removing sources beyond the principles outlined on this page (which call for a careful nuanced approach, not the rush job that has been demonstrated in discussion above and before). Deprecation is meant to be a phased transition, and while we take immediate steps to prevent new uses of deprecated sources for good reasons, we should be treating old uses as if they were originally added in good faith and with valid intent but should be implicitly questionable sources that should be replaced -- excluding the case of sources like the DM which I outlined are downright harmful. Because WP is a work in progress and there is no deadline to get it right, fixing questionable sources should be a reasonable high priority but it should be done in a manner to be disruptive. Arguably, we have many many many more pages with completely unsourced material that should be fixed first if we're talking sourcing issues (no source being worse than questionable sources per WP:V) but obviously that's not the focus here, nor am I saying that should be the focus. These should clearly fixed in time, and I commend if there's a concerted effort to do in a careful manner that seeks to try to find replacement sources and/or make sure removal of sources and material that could only be sourced to that material be done in a matter that is not disruptive and respects the fact that deprecation is a phasing approach, not a "do or die" pact. Again I do stress there are unique cases like the DM that need to be treated differently where because the RFC determined that the source fabricated material and thus should not be used, that gives a strong reason to remove more quickly and without the care of what mess may be left behind. Hence why I feel by conflating the DM situation with other sources that we have deprecated in other fashions, its messing up how to discuss the right ways to go about resolving deprecated sources. --Masem (t) 01:01, 30 November 2020 (UTC)
- As an aside, another possibility is that an editor might prepare for the edits ahead of time (in this case, reviewing the individual uses of the source) and then perform the edits afterwards. I just feel like I need to mention that since it's a workflow that I've used myself in the past. Sunrise (talk) 13:32, 17 December 2020 (UTC)
- And really - the usages of deprecated sources get extremely generic. They tend to be used in the same bad ways repeatedly. It can take a few seconds glance to decide if a given usage should be removed, {{cn}}'ed or go looking for a substitute. This can go quite quickly.
- I submit that you are attempting to have your cake and eat it: you are casting aspersions on other editors, but noticeably not sticking your neck out to state an actionable complaint, and I suspect that is because you know very well that you don't have the substance for one.
- WhatamIdoing's behaviour here is as the arbcom ruled 10-0, as quoted in WP:ASPERSIONS:
It is unacceptable for an editor to routinely accuse others of misbehavior without reasonable cause. Legitimate concerns of fellow editors' conduct should be raised either directly with the editor in question, in a civil fashion, or if necessary on an appropriate noticeboard or dispute-resolution page.
- Please stop casting aspersions on other editors. Make your case or don't. - David Gerard (talk) 15:11, 17 December 2020 (UTC)
- Sunrise, you're right that this happens on occasion, but I don't think that's what usually happens. You are also right that WP:MEATBOT becomes a problem after a while. Also, I hope that it's not what happened, because IMO "didn't pay enough attention, sorry" is a much more easily forgiven sin on wiki than "carefully reviewed the content and decided to that violating BLP was the best choice".
- Hello, @David Gerard. You have repeatedly, on this page, accused others of violating ASPERSIONS. ASPERSIONS, as you quote, focuses on editors who "routinely" (i.e., not one time) "accuse others" (i.e., individual editors, usually named) of "misbehavior" (e.g., not saying that an advice page needs to be improved; not saying that some editors may be the innocent victims of the incomplete advice provided on this page). Would you please stop casting aspersions yourself, against everyone who disagrees with you about whether this page is already providing the best possible advice? The only editor on this page who is "routinely" accusing anyone of actual bad behavior appears to be you. Please stop.
- For your convenience, I will re-state my "actionable complaint": This page (NB: not any individual editor, so ASPERSIONS cannot apply) IMO does not provide adequate information to editors. This page IMO should be improved. For example, this page provides basically no advice about what to do when an editor encounters a deprecated source other than to remove the source. Here are three specific examples of how this page IMO fails to provide adequate advice in this particular area:
- The page links to Template:Deprecated inline in the ==See also== section, but it never explains what the template is, how to use it, when to use it (or to not use it), or what to do if you find a source tagged with that in an article.
- This page says "Citations to deprecated sources should not be removed indiscriminately", but provides no information or examples of how editors should decide whether the source should be removed or kept (temporarily or otherwise).
- This page says nothing about what to do if you think that the source should be kept. Leave it so that someone else can remove it? Tag it? Add a hidden HTML comment so that a future editor will know that it's been reviewed? Start a talk-page discussion? Something else?
- Note that there are more areas that I think could be improved, and that these are not the only three problems within this particular area. I am giving these examples so that it will be clear to everyone that I have an actionable complaint that has absolutely nothing to do with the behavior of any individual editor. WhatamIdoing (talk) 19:04, 17 December 2020 (UTC)
- Sunrise's point is a fair one, but I will caution that this has also been an "excuse" used by one well-known editor that has been banned for such bot-like actions (BetaCommand). Yes, you can load up several pages, spend several minutes to review each, and then in a space of a minute, hit "submit" to rapid-fire those changes. That's an acceptable process, if the actions are clearly showing human review. But looking at the diffs and the editor actions around the same time, this does not look like their process; it looks like outright blind removal with no careful attempt to find replacement sources or evaluate the material that the references were supporting. --Masem (t) 15:30, 17 December 2020 (UTC)
- As I said in what you're responding to, and have said to you repeatedly (including in this discussion): make your case or stop casting aspersions - David Gerard (talk) 15:32, 17 December 2020 (UTC)
- This discussion has not been about individual behavior until you made it such. It has been about the community's approach to how deprecated sources should be removed from articles appropriately, only point to how some individuals have done it as examples of extremely fast removals that this guidance seems to caution against. The apparent conclusion of this discussion would be that if a slower, more evaluative removal approach is what the community wants, those individuals currently removing such links as quickly will be first cautioned about that change or affirmation to handling deprecated sources and then if behavior is not changed, taking that to AN. Taking the attitude that this is attacking other editors was not part of this until you brought that up. --Masem (t) 15:41, 17 December 2020 (UTC)
- I agree with @Masem that this discussion is about the advice we're giving (or failing to give) to editors, and not about editors' conduct.
- I'm less certain whether the community wants a slower, more evaluative approach. It's possible that, if we had an RFC, that the result would be that the respondents would prefer getting it done over getting it right. I think this because a surprising percentage of votes to deprecate a source, especially among newer editors, vote for the "Publishes false or fabricated information" option and give reasons that support only a claim that it's a WP:BIASED source (which could be potentially reliable). IMO editors who aren't differentiating between having a different political POV and regularly publishing hoaxes aren't very likely to care about getting things right. WhatamIdoing (talk) 19:17, 17 December 2020 (UTC)
- Which is why I suggested way up at the top that we may need a different term for sources that we are calling "bad" for outright fabrication like the Daily Mail. Something stronger than "deprecation" so that removal does not have to follow the careful curation that this guideline suggests and can be a more "remove first ask questions later" approach that we'd want for a bad source. Things with political bias (including state-sponsored media sources) should be handled with more care, but sources that we simply cannot trust have every reason to go quickly, and if we had a different term to present that in the various RS/N discussions on those sources, it would help clarify, implicitly, how the source removal will be handled to !voters. --Masem (t) 19:36, 17 December 2020 (UTC)
- This discussion has not been about individual behavior until you made it such. It has been about the community's approach to how deprecated sources should be removed from articles appropriately, only point to how some individuals have done it as examples of extremely fast removals that this guidance seems to caution against. The apparent conclusion of this discussion would be that if a slower, more evaluative removal approach is what the community wants, those individuals currently removing such links as quickly will be first cautioned about that change or affirmation to handling deprecated sources and then if behavior is not changed, taking that to AN. Taking the attitude that this is attacking other editors was not part of this until you brought that up. --Masem (t) 15:41, 17 December 2020 (UTC)
- As I said in what you're responding to, and have said to you repeatedly (including in this discussion): make your case or stop casting aspersions - David Gerard (talk) 15:32, 17 December 2020 (UTC)
- As an aside, another possibility is that an editor might prepare for the edits ahead of time (in this case, reviewing the individual uses of the source) and then perform the edits afterwards. I just feel like I need to mention that since it's a workflow that I've used myself in the past. Sunrise (talk) 13:32, 17 December 2020 (UTC)
- IMO, we are required by AGF to start from the assumption that editors are reviewing the sources ahead of time (or are simply able to evaluate the sources faster than we personally find plausible, etc). Of course we can conclude that AGF is broken in any particular circumstance, but it’s not that the actions have to clearly show human review; instead, others have to show that the actions clearly don’t show human review.
- More generally, in order to claim that a problem exists to be addressed, we need that to have happened at least a few times. The evidence for that would have to be that people have been brought to ANI (or wherever) and been warned or sanctioned. Otherwise we’d have to essentially claim that unspecified editors in good standing are editing inappropriately. It’s true that ASPERSIONS primarily seems to contemplate accusations against specific editors, but the distinction runs into problems if there are only a few people the statement is likely to be referring to. Additionally, at least one of the Arb cases cited there (ARBGMO, the one I’m familiar with) did deal with general statements of that nature. I think that David Gerard’s interpretation is at least reasonable, unless e.g. they were to be explicitly cited as an example of an editor whose behavior is not problematic.
- With regards to offering guidance on removal processes, my primary concern would be that we shouldn’t say anything that doesn’t apply to unreliable sources in general, as it could produce the perverse result that deprecation makes unreliable sources harder to deal with instead of easier. Sunrise (talk) 02:26, 18 December 2020 (UTC)
- Sunrise, given that this complaint about overzealous removal of sources has been made on multiple different pages, by different editors, in different subject areas, for several years now, I don't think that we should be concerned about these complaints referring to "only a few people". We seem to have a systemic problem.
- Masem might be correct that we need to split the "deprecated" section into something like "basically hoaxes" (e.g., Daily Mail) and "hopelessly biased" (e.g., academic papers published in China that say Traditional Chinese medicine is very effective at treating objectively measured medical conditions [such as high cholesterol]). That would address the problem on the front end (i.e., what labels are used on the page).
- I think that is a separate project from providing better advice about what to do when you encounter one of these sources. WhatamIdoing (talk) 23:54, 20 December 2020 (UTC)
- If you are in fact in support of removing deprecated sources from Wikipedia, could you please list some diffs from your own work to do so? - David Gerard (talk) 13:00, 30 November 2020 (UTC)
- It appears that historical references to DM are also removed en masse. E.g to articles from the 1920s to the 1980s. I suppose it is true that DM currently is trash (I don't know much about the newspaper) but was it as bad 40 or 50 years ago? The details they support are very hard to verify because the articles are not public - you need a subscription to a newspaper archive to do it. ImTheIP (talk) 08:44, 30 November 2020 (UTC)
References
- ^ Sputnik. "Russian Prime Minister Medvedev Says TPP, TTIP 'Encroaching on WTO'". sputniknews.com. Retrieved 3 August 2016.
- ^ Dwight D. Eisenhower (1963), Mandate for Change: The White House Years 1953-1956: A Personal Account, Doubleday, Wikidata Q61945939, p. 372.
- ^ Peter Binkley (2006). "Wikipedia Grows Up". Feliciter (2): 59–61. Wikidata Q66411582..
- ^ Feng Shi; Misha Teplitskiy; Eamon Duede; James A. Evans (4 March 2019). "The wisdom of polarized crowds" (PDF). Nature Human Behaviour. 3 (4): 329–336. arXiv:1712.06414. doi:10.1038/S41562-019-0541-6. ISSN 2397-3374. PMID 30971793. Wikidata Q47248083..
- So at what point was the Daily Mail not trash? This has never been established at any point in these discussions - David Gerard (talk) 12:59, 30 November 2020 (UTC)
- 1978 was the last year that The Daily Mail was not trash. "Paul Dacre joined dmg media limited as US Bureau Chief in 1979. Appointed Editor of the Evening Standard in 1990 he was appointed Editor of the Daily Mail in 1992"[11]
- Also see Sex, children and Mail Online and Boris Johnson 'asks former Daily Mail editor Paul Dacre to chair Ofcom' --Guy Macon (talk) 16:38, 30 November 2020 (UTC)
- So at what point was the Daily Mail not trash? This has never been established at any point in these discussions - David Gerard (talk) 12:59, 30 November 2020 (UTC)
Iranian content
editHi, I nearly only edit articles on automobiles. When it comes to the Iranian automobile industry, there are very few sources available. Deprecating PressTV is causing an enormous amount of trouble, removing tons of sources that are not replaceable. I am not suggesting we use PressTV for anything relating to the Holocaust or anything political whatsoever - but when it comes to press regarding SAIPA, this cannot be thought of as anything but a factual source. Same thing with the Daily Mail, whose coverage of the British automobile market in the 1970s was hardly concerning. The deprecated source list is quickly becoming very limiting to all sorts of work which is entirely uncontroversial. Please consider those of us who aren't making political or contentious edits. Mr.choppers | ✎ 20:36, 12 July 2021 (UTC)
Is YouTube a blacklisted source?
editHello Wikipedia. I always thought that YouTube was a blacklisted source, but I've recently seen a few articles that get away with citing it, and I have just done it myself whilst editing and no objection has been raised by Wikipedia as it usually does. Has it recently been taken off the blacklist? Thank you in advance for your help.
- Not at all - it's just usually a bad source ;-) It's just a host. If it's definitely officially from a source, it has the same usability as that source. In most cases it shouldn't be used - but in some, it's fine. See WP:RSPYT - David Gerard (talk) 16:24, 24 August 2021 (UTC)
- (ec) Youtube is just a host - some videos uploaded on there are WP:RSs and are copyright compliant, so can be used as sources, others do not meet the requirements of RS or are not copyright compliant. It depands on the iploader and the actual videos uploaded.Nigel Ish (talk) 16:31, 24 August 2021 (UTC)
- A good way of looking at it is that a video gains no reliability simply from being posted to YouTube, since anyone can post any video there. If, say, the BBC uploads something to its official YouTube channel, we can cite that the same way we could cite the BBC normally, but random YouTubers cannot be cited outside of the strict limitations of WP:SELFPUB / WP:ABOUTSELF. --Aquillion (talk) 16:58, 24 August 2021 (UTC)
Thank you - I am extremely grateful for all your kind and swift responses. RogerBuchanan (talk) 19:58, 27 August 2021 (UTC)
Blocking and deleting references to deprecated sources are attacks on Wikipedia:Prime objective
edit@Luisina Ferrante (WMAR): @PDiazR (WMCL): @Asaf (WMF): @Deryck Chan: @AKeton (WMF):
I'm concerned that Blocking and deleting references to deprecated sources are attacks on Wikipedia:Prime objective:
"Imagine a world in which every single person on the planet is given free access to the sum of all human knowledge. That's what we're doing."[1][2]
This concern with blocking and deleting deprecated sources doubtless sounds bazaar to some. However, communication requires both a sender and a receiver. People who think Wikipedia is biased against them are not likely to access it, especially if sources they find credible are never cited and are discussed only in the most superficial and derogatory ways. People making decisions to block and delete references to deprecated sources are saying in actions that speak louder than words that we know the truth, and if you disagree, you are ignorant. That's arrogant, especially with the limited transparency involved in deprecation decisions today; see my comments on "Increase transparency of deprecation decisions" below. I think current procedures for blocking and deleting sources deemed deprecated also conflict with Wikipedia:Assume good faith.
Current deprecation practices also deprive readers of access to what the deprecated sources say, an issue discussed in the section on "Reducing the risks from major conflicts" below.
I think we will better serve our mission by increasing the transparency of deprecation decisions and by encouraging editors who want to cite a deprecated source to join the discussion about why that source has been deprecated. I think Wikipedia should reference deprecated sources while clearly describing the credibility of the different sources as it related to the particular subject of the article.
We should not allow interminable discussions. User:Asaf (WMF) at Wikimania:2021:Program said we should set a limit on discussions. I don't know the optimal limit, but I would set something like two weeks; much longer than 15 seconds and shorter than a month.[3]
Doing this will, I think, have the following positive effects:
- 1. Increase the credibility of Wikipedia.
- 2. Reduce the number of people who think Wikipedia is against them.
- 3. Increase the number of people editing Wikipedia and promoting it to others.
To accomplish this, we will also need to improve procedures to limit the damage that ideologues and paid editors do. This would probably require moving beyond Wikipedia:Assume good faith to more aggressively enforcing safe space policies like Wikipedia:Meetup/ArtandFeminism/Safespacepolicy and Foundation:Friendly space policy.
I also think we should change procedures to increase transparency of deprecation decisions.
Increase transparency of deprecation decisions
editIn reading Wikipedia:Reliable sources/Noticeboard/Archive 355#CounterPunch, I saw multiple references that were NOT mentioned in the Wikipedia article on CounterPunch.
I think the following will help increase transparency and the positive impact of these deprecation discussions:
- Examples of poor quality publications used to justify deprecation should be discussed with appropriate citations in the article on the source, like CounterPunch, if such exists, presumably in a section with a title like "Controversies". People contributing to the deprecation discussion are already posting these comments to the Request for Comments (RfC) regarding deprecation, so it should not be much more work for the contributors to post essentially the same information to the article on the source being considered for deprecation.
- A flag should be posted in that article, to invite people who "watch" the article on a source to participate in the deprecation discussion. If a decision is made NOT to deprecate, then this flag should be deleted. If a decision is made to deprecate, the flag should be changed to include a discussion about how to reopen the discussion for users who think the decision was wrong.
- The Wikimedia Foundation (WMF) legal department should be notified, in case they want to say something. Some deprecated sources may try to sue the WMF. Our attorneys should be invited to monitor deprecation discussions in case they feel a need to contribute either to the general procedures or to the discussion on a specific source being evaluated. Relevant law is changing, as discussed in a presentation by Anna Mazgal at Wikimania:2021:Program on Wikimania:2021:Submissions/Do Something Doctrine - looking back on the Terrorist Content Regulation in EU. She said that the recently enacted Terrorist Content Regulation in the European Union (EU) is bad law, potentially criminalizing discussions of terrorism or even eco-terrorism that may make it difficult to talk rationally about anything that someone in power might want to label as "terrorism".[4]
If we can actually make Wikipedia more inviting in these ways, I think we can make major contributions to reducing the risks from major conflicts, as I discuss in the next section.
Reducing the risks from major conflicts
editWikipedia has been described as remarkable in getting different parties in conflict to engage each other in negotiating a narrative that all can more or less live with. We need more of this, not less. I believe we should actively encourage organizations concerned about conflict to hire Wikimedians in residence to recruit and train people to try to improve the respectful coverage of the concerns of major parties in conflict. This could make it easier for people to understand their designated enemies, so they are less likely to do things that manufacture recruits for their opposition and more likely to do other things that increase the chances for reasonable resolution. Blocking and deleting references to deprecated sources (a) strengthen the ability of censors to suppress the dissemination of information they don't like and (b) are obstacles to these kinds of conflict resolution efforts.
On May 23, 2013, then-US President Obama noted that terrorism caused fewer American deaths than car accidents or falls in the bathtub. He occasionally had to be badgered by advisors into choices commensurate with popular fear. He also worried that counterterrorist priorities swamped his other foreign policy aspirations.[5] This suggests to me that the President of the US understood that some of what he was doing was counterproductive, but the political environment effectively prevented him from pursuing policies that would likely have substantively reduced terrorism and increased US national security. US Presidents Dwight Eisenhower and Lyndon Johnson made crudely similar comments about their policies toward Vietnam, as documented in the section on "US foreign interventions in opposition to democracy" in Wikiversity:Winning the War on Terror.
I suggested encouraging such use of Wikimedians in residence in two presentations earlier this year:
- Discussing "Wikiversity:International Conflict Observatory" at the 2021 Military Operations Research Society Symposium.
- Presenting "Wikiversity:Managing conflict on Wikipedia and internationally" at WikiConference North America 2021. The latter presentation was in joint with User:Luisina Ferrante (WMAR), Education and Human Rights manager for Wikimedia Argentina, and Patricia Díaz-Rubio, Executive Director of Wikipedia Chile. Since 2016 Luisina has helped organize dozens of edit-a-thons throughout Latin America and the Caribbean that have gotten people to contribute over a thousand new images used in over a thousand articles that have received tens of thousands of views and offered the public a broader perspective of concerns, esp. re. violence by police. Patricia said that this work in Chile was instrumental in obtaining voter approval for the current process of rewriting the Chilean constitution, scheduled to be completed by July 2022. Without it Chile would likely still be burdened with the constitution written by the Pinochet dictatorship.
Editors and administrators paid to distort Wikipedia
editWhat percent of Wikipedia editors today are paid to burnish the images of their clients and drive away honest volunteer editors? Almost certainly, that number is greater than 0 but less than 50 percent. If Wikipedia's credibility continues to grow, more people with power will likely hire more people to damage Wikipedia in these ways.
In 2013 the French interior intelligence agency Direction centrale du renseignement intérieur (DCRI) invited Rémi Mathis, a volunteer administrator of the French Wikipedia and resident of France, to their offices and ordered him to delete a Wikipedia article they did not like. Mathis complied, understanding he would not be allowed to leave otherwise.[6]
An even more serious example was mentioned by User:Deryck Chan in a presentation at Wikimania:2021:Program on "Wikimania:2021:Submissions/Cross-wiki ideological conflict and Wikimedia's vision of knowledge equity". Some editors in Hong Kong of the Cantonese language Wikipedia were told they would be reported to the national security police if they continued trying to make an article reflect what could be documented from reliable sources when that conflicted with official government policy.[7]
Even if we ignore the Chinese, there are tens of thousands of people with power around the world who may believe that information published on Wikipedia threatens their social status enough for them to pay people to infiltrate the Wikimedia system. Some of their employees may be attacking honest volunteers posting content their employers don't like. Others on their payroll may make mostly reasonable edits and become an administrator, from where they could do substantial damage to the Wikipedia:Prime objective. The French DCRI and many of their counterparts elsewhere may have learned since 2013 that they might be more effective in managing their image on Wikipedia by paying disruptive editors like this.
In particular, what percent of the people contributing to these deprecation discussions and then blocking and deleting references to sourced designated as deprecated are actually being paid to burnish the images of their employers and tarnish the images of their designated enemies? I don't think we have any way of knowing. We can hope that it's zero today. However, we can be confident that at some time in the future, some -- perhaps the majority -- of people engaging in deprecation discussions and deleting deprecated references will be paid to do that, to the general detriment of humanity.
The supporters of the Pinochet dictatorship in Chile, in the US, and around the world evidently have not yet penetrated the Wikimedia system in this way. It seems likely that they will figure it out eventually. We need procedures that will limit their ability to damage Wikipedia's integrity, including using deprecation. DavidMCEddy (talk) 15:05, 20 October 2021 (UTC)
References
- ^ "Wikipedia Founder Jimmy Wales Responds". Slashdot.org (Interview, Q&A). SlashdotMedia. July 28, 2004. Retrieved October 3, 2017.
Wikipedia is an excellent project, and Slashdot readers' questions for Wikipedia founder Jimmy Wales were just as excellent -- as are Jimmy Wales' answers to 12 of the highest-moderated questions you submitted.
- ^ wikimediafoundation.org Vision. Consulted on 30 September 2019
- ^ Wikimania:2021:Submissions/Conflict Resolution: Collective Responsibility
- ^ "The European Union Moves to Fight Terrorist Content Online". The Soufan Center. 15 June 2021. Wikidata Q108679526. Regulation (EU) 2021/784 of the European Parliament and of the Council of 29 April 2021 on addressing the dissemination of terrorist content online, Wikidata Q103793218 For more see Wikimedia.brussels' blog.
- ^ Samuel Moyn (2021). Humane: How the United States Abandoned Peace and Reinvented War. Farrar, Straus and Giroux. ISBN 978-0-374-71992-0. OL 34118912M. Wikidata Q108896140., pp. 268, 299-300.
- ^ After that deletion, the Wikimedia Foundation sent several letters to the French government requesting an explanation. When none arrived, a French-speaking Wikipedia administrator in Switzerland restored the article.
- ^ Selina Cheng (14 July 2021). "Hong Kong Wikipedia editors take precautions amid fears mainland peers may report users to national security police". Hong Kong Free Press. Wikidata Q108667501.
- This has been posted to the wrong place. The deprecation procedure was questioned in an RFC at WP:RSN, and kept: Wikipedia:Reliable_sources/Noticeboard/Archive_275#RFC:_Moratorium_on_"general_reliability"_RFCs - if you want to change this community consensus, WP:RSN is probably the best place to start. You could also try WP:VPP, though I suspect they'd refer you back to RSN in the first instance. I'd make your proposal shorter than this, though - David Gerard (talk) 17:54, 20 October 2021 (UTC)
- There hasn’t been a Pinochet dictatorship in Chile to support since the 1990s. Horse Eye's Back (talk) 18:29, 20 October 2021 (UTC)
- The current Chilean constitution was written by the Pinochet government. Patricia Díaz-Rubio, Executive Director of Wikipedia Chile, implied that their Wiki Human Rights campaign gave voice to people who wanted a constitution that gave more power to common folk and less to the ultra-wealthy, who had supported Pinochet. DavidMCEddy (talk) 21:12, 20 October 2021 (UTC)
Deprecated sources as reliable reference to their own texts?
editWP:Deprecated sources fairly states, Deprecated sources ... fail the reliable sources guideline in nearly all circumstances."
Can the other circumstances include use for direct, interpretation-free reference to their own materials? I'm thinking that a source might be cited perhaps solely with a link to archive copy from a close to publication time. GregKaye 05:27, 28 July 2022 (UTC)
- I can't think of an exception right now. Literally ALL content here must come from, or be filtered through, RS.
- Deprecated sources have zero due weight, so actual RS that discuss or quote them would have to be used. That's the only way we can document many fringe views that are unnotable in mainstream RS, so we are then forced to resort to the WP:Parity rule, because sometimes only skeptical or quackbusting sources mention such fringe views.
- A deprecated source does not have any due weight, not even enough for WP:ABOUTSELF, so if parity reliable sources don't mention it, it gets no mention at all here. -- Valjean (talk) (PING me) 06:06, 28 July 2022 (UTC)
- I agree, Valjean, that
"A deprecated source does not have any due weight, not even enough for WP:ABOUTSELF"
but I'm thinking about using it as a WP:PRIMARY source for its own text. As an example, in the case of media articles that have been the centre of libel lawsuits such as the article[12] in The Sun at the centre of the Depp v News Group Newspapers Ltd dispute or, in the case of Depp v. Heard, that article along with an article[13] in The Washington Post which gey both referenced, just with one without a citation. In the second case (where I've been involved in editing) I'd have thought it could present readers with possibilities if they were provided with ready reference to both articles to facilitate comparison between arguably very different article contents. Web archive present copies[14] of the Sun article from May 6 2018[15] for a"27th April 2018, 10:00 pm
version following a 7.58am title change[16]. My thought was that, in a situation like this, web archive might act as a reliable and direct primary reference of what texts said/still say. GregKaye 13:23, 30 July 2022 (UTC)- Well, there are exceptions to every rule, so why not show us the version right here, with the sources and quotes, that you're talking about. Then we all can look at it and maybe come up with a version that doesn't break too many rules. It sounds a bit too complicated for me to understand all the nuances without actually seeing it in practice. There are also others here who can chime in with their advice.
- Regarding The Sun, WP:RSP says:
- "The Sun was deprecated in the 2019 RfC. There is consensus that The Sun is generally unreliable. References from The Sun are actively discouraged from being used in any article and they should not be used for determining the notability of any subject. The RfC does not override WP:ABOUTSELF, which allows the use of The Sun for uncontroversial self-descriptions. Some editors consider The Sun usable for uncontroversial sports reporting, although more reliable sources are recommended."
- So ABOUTSELF use has an exception "for uncontroversial self-descriptions". That is only about "self-descriptions" of the site itself, not about commentary on others, especially per BLP restrictions against using questionable sources for BLP matters like this case. An archive copy doesn't ameliorate any of this. If the issue is significant (has due weight), reliable sources will have commented on it. Otherwise, it would likely be OR to use different archived versions to make a point not made by RS. If RS make the point, then quote them and their quoting of The Sun, IOW the content is "filtered through a RS". I hope that all makes sense to you. -- Valjean (talk) (PING me) 20:28, 30 July 2022 (UTC)
- File:Freddiehamster.jpgI think the main potential reason here not to cite the The Sun article could be in potential applications of BLPPRIMARY. But for this, I think that the lead for Depp v. Heard worked as a better base for reader comparisons here were readers could directly compare the (incendiary) Sun article [6] (not judged as defamatory) to the (relatively subtle) Washington Post article [10] (judged as defamatory). Readers could compare both article contents with parallel means being provided for reference. I don't see how the archived writing of The Sun can be depreciated as reference to the writing of The Sun. "If a bear shits in the woods" and the shit was rapidly archived then that archived shit would surely represent a reliable source regarding the activity of the bear. Personally, I do not think that RS are necessarily reliable in regard to their potential bias when covering topics relating to defamation and free speech. (Also personally, I've been significant in arguments against the lead inclusion of a "wife beater" reference, which was very temporarily used by The Sun online. However I still think that inclusion of direct article reference can, in some circumstances, can provide reliable direct reference to the to the content of that article). GregKaye 08:19, 19 August 2022 (UTC)
- If it does go in - which I am entirely unconvinced it should, the judgement suffices IMO - it should be an archive as close to publication as possible. Deprecated sources can't even be trusted as to their own websites, we keep finding. Or an image of a print version.
- (also, all of this is really one for RSN.) - David Gerard (talk) 11:51, 19 August 2022 (UTC)
- File:Freddiehamster.jpgI think the main potential reason here not to cite the The Sun article could be in potential applications of BLPPRIMARY. But for this, I think that the lead for Depp v. Heard worked as a better base for reader comparisons here were readers could directly compare the (incendiary) Sun article [6] (not judged as defamatory) to the (relatively subtle) Washington Post article [10] (judged as defamatory). Readers could compare both article contents with parallel means being provided for reference. I don't see how the archived writing of The Sun can be depreciated as reference to the writing of The Sun. "If a bear shits in the woods" and the shit was rapidly archived then that archived shit would surely represent a reliable source regarding the activity of the bear. Personally, I do not think that RS are necessarily reliable in regard to their potential bias when covering topics relating to defamation and free speech. (Also personally, I've been significant in arguments against the lead inclusion of a "wife beater" reference, which was very temporarily used by The Sun online. However I still think that inclusion of direct article reference can, in some circumstances, can provide reliable direct reference to the to the content of that article). GregKaye 08:19, 19 August 2022 (UTC)
- I agree, Valjean, that
An archived copy is a mirror version, and a mirror of an unreliable source is just as unreliable as the original. -- Valjean (talk) (PING me) 01:53, 20 August 2022 (UTC)
- It's a snapshot of the unreliable source at a particular time, though, which is relevant if it's being used as a primary source link - David Gerard (talk) 16:28, 20 August 2022 (UTC)
- Yes, as long as it's only used for neutral self-descriptions, as described above, and only if the RS which discusses the issue has mentioned it. -- Valjean (talk) (PING me) 16:40, 20 August 2022 (UTC)
- yep, absolutely. e.g. In the few cases we include a Daily Mail primary source, I've made them early archives - David Gerard (talk) 16:45, 20 August 2022 (UTC)
- Yes, as long as it's only used for neutral self-descriptions, as described above, and only if the RS which discusses the issue has mentioned it. -- Valjean (talk) (PING me) 16:40, 20 August 2022 (UTC)
Identifying the deprecated reference
editI made an edit here (User:Semsûrî/sandbox/3) where 78 references were added and when I saved it I received a notification telling me that I was about to add a deprecated source. I still went ahead and saved the edit. Now, my question is, how do you figure out what reference is the problematic one? I did check Wikipedia:Reliable sources/Perennial sources but couldn't immediately figure out which one it is. Semsûrî (talk) 15:43, 11 April 2023 (UTC)
- @Semsûrî: I recommend using the WP:UPSD script, which highlights sources in the reference list based on their status at RSP. Currently the deprecated source in your sandbox is ref 118 (Telesur). Sunrise (talk) 13:53, 29 April 2023 (UTC)
Interviews an acceptable use?
editDo interviews conducted between an article subject and a journalist for a deprecated source fall under acceptable use criteria? While not optimal, there are instances when an interviewee may share a potentially important piece of insight relevant to the article. Krisgabwoosh (talk) 12:42, 19 November 2023 (UTC)
- As per #Effects of deprecation, "quotations should also be avoided, since they can be manipulated or fabricated. If the source contains material that cannot be found in more reliable sources, it may be valid to assume that the material in question is incorrect". Nikkimaria (talk) 14:15, 19 November 2023 (UTC)
5pillarsUK
editWikipedia's requirement for writing articles is "verifiability, not truth.", in this article, https://5pillarsuk.com/2023/12/29/east-london-jewish-charity-is-raising-funds-for-idf/, the site says The IDF has been widely accused of committing war crimes, and even genocide, by human rights organisations. Without providing any proof for the spurious claims. It should therefore be deprecated if it hasn't already. Steveonsi (talk) 07:43, 4 January 2024 (UTC)
- That specific article does not appear to be cited in any Wikipedia articles at the moment. If anyone would like to inquire about the reliability of 5Pillars, which is currently cited in 43 articles , a new discussion would need to be started on the reliable sources noticeboard.
- However, please be aware that the Arab–Israeli conflict is a contentious topic with an extended confirmed restriction, which means that your Wikipedia account does not yet have enough activity to participate in this topic area, with the exception of making edit requests. Please see the messages I have posted on your user talk page for more information. — Newslinger talk 04:12, 5 January 2024 (UTC)
Medium (website)
editI recently ran into a problem when I tried to save some notes in a userspace page of mine. I was notified that a deprecated link was on the page, and only when I deactivated the link to Medium (website)'s website was I allowed to save the page. I didn't even have any intention of ever using the link in an article. What's going on? Why get a warning for Medium? It isn't deprecated at all. It should just be used very cautiously. -- Valjean (talk) (PING me) 21:54, 17 January 2024 (UTC)
- (Discussion at RSP talk page) Sunrise (talk) 23:39, 21 January 2024 (UTC)
Why are right wing sources more predominately depricated/restricted?
editTitle says it all. Just for fun I compiled a list of 55 sources of American media which have bias ratings from AllSides. 36 of the sources are left wing. Of these, only one is rated as unreliable (Alternet). 19 of the sources are rated as right wing. Only one of these is considered reliable (ReasonTV)! Seven are rated as unreliable, and six more are completely depricated. Just curious how you justify this? 209.171.85.237 (talk) 15:45, 25 February 2024 (UTC)
- The bias is not a reason for deprecation (see WP:BIASED). A source may be biased but if it's reliable we can use it balancing it with other sources having different biases. If there are specific "left wing" sources that have published falsehoods, I suggest you check the archives of WP:RSN and raise this issue there. Alaexis¿question? 21:06, 25 February 2024 (UTC)
- You missed Occupy Democrats (RSP entry), which is a deprecated American left-wing source. AllSides has yet to rate other American far-left sources that have been deprecated, including The Grayzone (RSP entry) and MintPress News (RSP entry).Also, AllSides only rates bias on the left–right political spectrum and does not rate reliability. If you check a media rater that also classifies media outlets for reliability, such as Ad Fontes Media, you'll find that Wikipedia editors and media raters tend to identify the same sources as unreliable. See Wikipedia:Wikipedia Signpost/2020-11-29/Op-Ed for details.Finally, keep in mind that AllSides's media bias ratings are from the perspective of American politics. The Wikipedia community is international and has a much broader range of political perspectives that frequently differs from American political discourse. — Newslinger talk 02:20, 27 April 2024 (UTC)
Closed RfC deprecating Tasnim
editThis closed RfC on Tasnim News Agency should probably be added here by an uninvolved editor. Thanks. - Amigao (talk) 17:32, 28 May 2024 (UTC)
Deprecated Sources - Actual Wikipedia vs Current Description
editActual wikipedia seems so very different to what is written in this article. In reality, a deprecated source has a de-facto ban, completely unlike what is currently described.
"Acceptable uses of deprecated sources Deprecation is not a blanket retroactive "ban" on using the source in absolutely every situation, contrary to what has been reported in media headlines. In particular, reliability always depends on the specific content being cited, and all sources are reliable in at least some circumstances and unreliable in at least some others. Citations to deprecated sources should not be removed indiscriminately, and each case should be reviewed separately. While some deprecated sources have been completely eliminated as references, others have not."
In reality, using a deprecated source means facing a challenge when posting, then being flagged by a bot, then having editors reject the source. There is no corresponding check whether the use is as a primary source, an opinion, or whether the criteria of factual sourcing apply. There seems no way to have an 'acceptable use' be accepted.
My experience is wiki/User_talk:Amigao#Wall_Street_Journal,_reference_from_Global_Times 14.201.39.78 (talk) 03:51, 11 August 2024 (UTC)
- I agree. This seems to be a rather straightforward case in which there can be no reliability issues. I cannot comment whether the content is due in that article. Alaexis¿question? 19:41, 11 August 2024 (UTC)
- Having read other topics on this page, I just wanted to highlight that my point is that it is confusing for new users to have the documentation so very different from the practice. Also, if the aim is for the documentation to be followed then there needs to be some balance against the zealous deletions based on source and not reliability. I appreciate comments on my own edit, but prefer they be added to the discussion here. 14.201.39.78 (talk) 10:08, 13 August 2024 (UTC)
There is no such thing as a "depreciated" source
editThe word is "deprecated". Please get it right and fix any misspellings. -- Valjean (talk) (PING me) 16:02, 13 August 2024 (UTC)