Wikipedia talk:Blocking IP addresses/Archive 1

Archive 1

Failsafes against large ranges

Are there failsafes so that the IP range *.*.*.* isn't blocked? --ÆAUSSIEevilÆ 22:21, 28 November 2006 (UTC)

Yes. MediaWiki does not allow range blocks affecting more than 65,536 IP addresses; that is, it allows a maximum CIDR range of /16, or 0.0.0.0–0.0.255.255. Blocking every possible IP address would require 65,025 individual range blocks. For more information, see m:Range blocks. —{admin} Pathoschild 05:49, 29 November 2006 (UTC)

Blocking IP addresses in the UAE

Please can someone explain to me why UAE IPs keep getting blocked (usually by KhoiKhoi, Can't Sleep, Clown Will Eat Me, and one other name I've forgotten now).

When I read this page, it doesn't make sense. In the UAE, we are legally required (with some locational exceptions - eg Free Zones) to connect through the ISP's (Etisalat) proxy server, which assigns IP addresses dynamically. According to information on this page, any blocks are undesirable and if regarded as necessary, should only be temporary?

khoikhoi sent me this link as an explanation but I still don't get it. http://ecompany.ae/eco/isp/english/services/dialin/index.html

Thank you. signature 07:05, 27 March 2007 (UTC)

You could always just register an account, which is also what WP recommends for people that have to use a shared IP (students using the internet from their college's computers, for example). TJ Spyke 03:24, 15 April 2007 (UTC)

Tor proxies

Please have a look at this discussion on the blocking of Tor proxies: Wikipedia talk:Blocking policy#Softblock for Tor proxies. —Babelfisch 07:25, 3 April 2007 (UTC)

Merged from Wikipedia talk:Sensitive IP addresses

Other large governmental users

Should this list be perhaps expanded to included, for example, Saudi Arabian IPs, which all come through proxy servers? Physchim62 (talk) 15:12, 4 September 2006 (UTC)

This list is primarly used for editors FROM large government bodies, as a reminder to admins placing blocks on them to use extra clear block reasons. — xaosflux ^Talk 02:46, 5 September 2006 (UTC)

I concur - where's the CIA, the DOD/DNS, 'etc? Raul654 22:01, 26 October 2006 (UTC)

We may already block the CIA, and perhaps other intelligence agencies, somewhat. They use their own IP at least some of the time, but they also use anonymizing services, which we block when find out about them. A company that the CIA used went out of business, but they might be using a different one now. -- Kjkolb 09:43, 29 November 2006 (UTC)

Government of Canada

Do you think we could add the IP addresses allocated to the Government of Canada to this list? The range is 192.197.77.0 - 192.197.86.255, or in CIDR notation:

• 192.197.77.0/24
• 192.197.78.0/23
• 192.197.80.0/22
• 192.197.84.0/23
• 192.197.86.0/24

--  Denelson83  22:43, 6 September 2006 (UTC)

Yes, go right ahead. Then leaev a note to ComCom if they need to be added to the admin directions. — xaosflux ^Talk 02:31, 7 September 2006 (UTC)

Comment

If I understand the current blocking implementation correctly, putting an exlicit, indefinite anon-only block on this IP should prevent any autoblocks from having effect. This would mean that toolserver bots can't edit without logging in, but I think that should be considered a feature. Should we do this? —Ilmari Karonen (talk) 23:57, 24 November 2006 (UTC)

moved from Wikipedia:Sensitive IP addresses. --Daniel Olsen 07:49, 25 November 2006 (UTC)

Sensitivity

If blocks to these IP's would be so sensitive to the higher-ups at Wikipedia, why can't there just be a programming feature that allows only the highest levels of decision makers at Wikipedia to block these IP's?Tragic romance 12:03, 12 December 2006 (UTC)

The IPs may still need to be blocked, as vandalism can and has come from them. If one of these IPs is insanely vandalizing, it would not be efficient to call a higher-up and wait a few hours, while in the mean time it vandalizes hundreds of pages that all need to be reverted. Instead, the IP is temporarily blocked, and the political and public relations issues can be addressed. —Centrx→talk • 23:28, 3 January 2007 (UTC)

Qatar

Just wondering whether or not the Qatar ISP IP addresses (and any other IP address, which if blocked would effectively block another country or large region of users) shouldn't be listed on this page ? --Kind Regards - Heligoland | Talk | Contribs 00:50, 4 January 2007 (UTC)

Parliament of Sweden

I recently left a message on a IP user's talk page, and did a check to see if the IP address was shared. The address, 194.52.83.21, came back as the Parliament of Sweden. Seems they have the range 194.52.83.0-194.52.83.255, might be good to know. --Oden 00:47, 11 January 2007 (UTC)

Template?

Is there a template to identify these IPs on their talk pages? If not, should one be created? Either way would it be useful for SelketBot to tag them like it does with {{SharedIP}} and {{SharedIPEDU}}? --Selket ^Talk 23:19, 26 April 2007 (UTC)

How to determine if an IP address is shared and by whom

I would like to link a good description of How to determine if an IP address is shared and by whom from Template_talk:SharedIP#See_also. I feel such a section would do well in this article. I have used network-tools.in for trace and whois, but often end up with unclear results, ( 63.3.11.2 = uu.net = verizon ; but what to put with {{SharedIP}} ) Comments? ∴ here…♠ 21:55, 2 May 2007 (UTC)

I guess I wasn't very clear. Will someone experienced with using the {{SharedIP}} template and blocking IP addresses please detail: How to determine if an IP address is shared and by whom , for use in this article and elsewhere? ∴ here…♠ 19:25, 3 May 2007 (UTC)

Example, 207.237.78.111, RCN Corporation, comment: ADDRESSES WITHIN THIS BLOCK ARE NON-PORTABLE. Shared? What tools do you use? Where is the documentation for doing this? (I imagine it should go here...) ∴ here…♠ 19:44, 3 May 2007 (UTC)

IPv6 Addresses

What about IPv6 Addresses? Does MediaWiki support IPv6 users and blocking?
--FastLizard4 (Talk•Links•Sign) 22:02, 20 August 2007 (UTC)

Hughes Net

Hughes Net operates in 69.19.0.0 - 69.19.127.255 There is at least one complaint of inappropriate blocking (Wikipedia_talk:Arbitration_Committee#Hughes_Network). Is there a range block or some other problem? Fred Bauder 19:19, 10 September 2007 (UTC)

As a Hughes Net user, I have noticed there seem to be a lot of problematic users on Hughes Net, and not only at Wikipedia. The problem is that those users get their IPs banned, then are later assigned a new IP (dynamic IP) and the ban is passed off to another, often harmless and subsequently frustrated, user. -- 66.82.162.14 (talk) 08:01, 28 September 2009 (UTC)

Need some help with an IP

An anonyuser is semi-vandalizing pages with different similar IPs each time. User can be seen here in the revision history for Cliff Hangers, each time adding something to the caption of the image. First time was that he the contestant in the image is a "black contestant" (a few times), then a "male contestant", then "one contestant" (as opposed to simply "a contestant" as is normal). On its face, it could be good faith, but not en masse. One of the addresses also added a link to a person on the page Brad with the descripion "non-notable". It's not a mega-vandal, but it's pretty clearly someone just trying to be a smartass. I'm not sure exactly what to do about it, as the wp:vandal instructions aren't quite clear to me, and the template to put on the addresses was a bit confusing. Thanks TheHYPO (talk) 16:00, 15 December 2007 (UTC)

Sensitive IP addresses

The table that automatically transcludes on the right side of the "Block user" page does not match up with the addresses listed here. For example, Qatar is listed on the "Block user" page as 82.148.96.68/31.

Where is that table actually maintained and would someone please update it? Thanks. Rossami (talk) 22:08, 26 February 2008 (UTC)

It's at MediaWiki:Blockiptext. Algebraist 15:08, 20 March 2009 (UTC)

Block all IP addresses

All anonymous IP editing needs to be blocked. They're all vandals. This crap needs to stop now; it's seriously degrading the quality of our encyclopedia here. Dr. Cash (talk) 14:32, 20 March 2008 (UTC)

No, we're not. That is a flat-out lie based on WAY overgeneralization! — Preceding unsigned comment added by 69.9.48.198 (talk) 10:18, 15 August 2011 (UTC)

How do I look for all changes made by an IP range

user:Breathtaker operates under the 87.122 range to get around his block. I would like to be able to see all changes done by this range. I tried inputing "87.122.0.0/18" into the User Contribuitons page but that did not work. Any help would be greatly apreciated.--Dr who1975 (talk) 19:59, 17 April 2008 (UTC)

Blocking an entire browser?

The other day Slakr blocked the range 195.189.142.0/23, which I think blocks all users of the opera mini browser (the best browser for pocket PCs and smartphones, which uses a central proxy server to improve download speed and readability on smaller screens). Until the other day I was able to edit using it without problems, now as soon as I hit the edit button I get the "your blocked due to using an open proxy" message. I haven't changed any settings. Can someone please unblock it and contact the Opera Admins if there are any concerns about how their servers/proxies are set up, but in the meantime let us use a decent browser for our mobile editing, rather than having to use the substandard alternatives. I have left a message on Slakr's talk page without any response.The-Pope (talk) 13:02, 26 June 2008 (UTC)

Mobile Phone IP Template

I think Mobile phone carriers should have a specialized template such as:

---

Attention:

This IP address, Blocking IP addresses/Archive 1, is registered to {{{1}}}, a mobile phone company and is used by many of its customers.

For this reason, a message intended for one person may be received by another and a block may be shared by many. If you are editing from this address and are frustrated by irrelevant messages, you can avoid them by creating an account for yourself. In some cases, you may temporarily be unable to create an account due to efforts to fight vandalism; if so, please see here.

If this IP is blocked a range block should be used because the IP will frequently change.

---

Caution should be used when blocking this IP or reverting its contributions without checking - if a block is needed, administrators should consider using a soft block with the template {{anonblock|optional comment}} as the block reason.

Note: In the event of vandalism from this address, abuse reports may be sent to your network administrator for further investigation.

IT staff who want to monitor vandalism from this IP address can subscribe to a web feed of this page in either RSS or Atom format.

--HereFord 22:24, 1 August 2008 (UTC)

I don't see any objection to this myself, Hereford: you've already got a template that I can't find any fault in at all, so just go ahead and introduce the template into the mainstream. You might wish to list it at the matrix for user talk templates, Wikipedia:Template messages/User talk namespace (specifically, #Other, "Page headers"); post a notice on the appropriate village pump, so folks are aware that the template is now being used; write up some documentation (useful guidelines); and have a dig for any user talk pages that could be tagged with this template, in the way of testing it "out there" (although it does appear to work perfectly well).

With regards to whether it should be introduced or not, then I say yep, it seems like a good idea to me. ;)

Anthøny 13:01, 4 August 2008 (UTC)

Its now at Template:MobileIPHereFord 13:56, 4 August 2008 (UTC)

Tags documentation

Hi, for anyone who has experience with tagging IP pages, we are working on expanded docs to clarify how to determine which tag ({{ISP}}, {{SharedIP}}, {{whois}}, etc.) should be used on a given IP. Anyone who would like to help with this process, please feel free to join the discussion at Template talk:ISP. --Elonka 18:17, 10 December 2008 (UTC)

DoD IP Addresses

Do we have any DoD IP addresses on file? That should probably be added to the list. If we should add DoD addresses to the list, I (or an administrator) could try to contact DISA about it. Just my two-bits. Glacier Wolf 22:34, 24 March 2009 (UTC)

They used to have a class A allocated. Rich Farmbrough, 20:12, 12 December 2009 (UTC).

Updates required?

1. I notice that m:Communications committee is tagged as out-of-date. Is this body still operating?
2. After discussion at the pump (Toolserver IP editing logged-out again), the "Sensitive for technical reasons" section may need changing.

OrangeDog (τ • ε) 22:58, 6 December 2009 (UTC)

I don't know about ComCom. I would probably ask Cary or Jay what they best place to report these is (if its even still necessary). The Toolserver note is still mostly correct, as hardblocks will still break things, though its on the autoblock whitelist, so the comment about removing autoblocks isn't necessary. That ClueNet IP might be out of date as well. Mr.Z-man 20:20, 12 December 2009 (UTC)

Problems with blocking IPs

People can change their IPs easily and continue vandalising. The IPs that are passed on affect innocent people. If you block the IP range, there will be many innocent people affected. For regestered accounts, those people can log out and use the same way.--219.74.14.141 (talk) 10:55, 24 March 2010 (UTC)

This is why blocks are generally short in such cases, and need to be carefully examined. But yes, sometimes there is collateral damage, but there are also other methods of stopping such vandalism. --Dirk Beetstra ^{T C} 11:20, 24 March 2010 (UTC)

But how?--219.74.14.141 (talk) 11:46, 24 March 2010 (UTC)

How what? --Dirk Beetstra ^{T C} 16:53, 24 March 2010 (UTC)

How to deal with those types of vandalism. See, my IP has changed.--116.14.180.93 (talk) 13:09, 25 March 2010 (UTC)

IP block length

"consider blocks over a period of months or years instead" [1]

I have been operating under the impression that we typically only block for years when the IP is an open proxy. In my opinion, blocks of up to one year are more than sufficient; and ensures that extended blocks of IPs are reviewed on an annual basis. Should this be noted on the project page? –xeno^talk 16:48, 23 March 2010 (UTC)

I'd say so. An IP is generally only blocked for a year or more in the case of:

• 1 year: School IPs subject to frequent abuse and several previous long-term blocks, extremely static IPs subject to the same
• 2-5 years: Open proxies
• Indefinitely: By authenticated request of a technical officer from a school or company to functionaries.

Noting also that the latter occurs fairly infrequently. Hersfold ^(t/a/c) 20:51, 23 March 2010 (UTC)

Generally true, and should be practice. However, if an IP repeatedly comes back days after a long term block expires and starts vandalising or picking up old habits, I would suggest that indef(-like) blocks would .. clear up a bit. Same goes for some static IPs who run into severe problems in many places, where an indef(-like) block may finally get the message through: don't come back here. There is always the unblock-mailing list or ArbCom to get it unblocked again. --Dirk Beetstra ^{T C} 09:13, 24 March 2010 (UTC)

The IP will eventually move on. So how long would you propose "indef-like" blocks be left in place for? Is reviewing such blocks on an annual basis really such a timesink? –xeno^talk 13:20, 24 March 2010 (UTC)

Mwaagh, one or two edits and immediately being blocked again does not have to be a timesink, indeed. But some IPs do take more than that. Its not that I don't agree with you, Xeno, but some IPs are just not worth even one edit .. --Dirk Beetstra ^{T C} 16:52, 24 March 2010 (UTC)

@xeno - In the case of a few school ips that I could dig up, per Dirk the original vandal may have moved on (progressed through grades, dropped out from smoking crack, became obsessed with Twilight and lost interest in vandalizing Wikipedia) but the behavior post-previous long term blocks was essentially the same. Basically, nothing good is coming from the anon users at that IP. This doesn't cover any logged-in edits from the IP, and its been my assumption on the long term school blocks that either (1) nobody anon actually cares that they are blocked since we don't see unblocks or (2) all useful editors at that IP have either resigned themselves to editing at home or have created accounts and are being productive / vandalizing under those. I do watchlist the schools I've placed long term blocks on to see if we get unblock requests, but my feeling is that in some cases years is a valid and useful guideline. Its not that its a timesink to review on a yearly basis, Xeno. Its more the scientist in me railing at having to recheck that 2+2=4 every year, and finding that it does indeed still equal 4 making a note to come back in a year and recheck it again. Syrthiss (talk) 13:03, 25 March 2010 (UTC)

I suppose my (perhaps overly) optimistic hope is that in the short time the IP is unblocked each year, we might snag a constructive contributor or two. –xeno^talk 13:15, 25 March 2010 (UTC)

I know, which is laudable. It'd be interesting to know how many productive editors are editing from behind long term IP blocks, or even how many encountered a long term (or even the short term) IP anonblocks and then registered an account. Syrthiss (talk) 13:18, 25 March 2010 (UTC)

...and IIRC, open proxy ips are blocked indef. Not 'years' or 'year'. At least that was the guideline back in olden days. Syrthiss (talk) 13:10, 25 March 2010 (UTC)

Open proxies are (these days) typically blocked for 5 years, not indef. –xeno^talk 13:12, 25 March 2010 (UTC)

I think we need continuing education guidelines for admins then, tho at least to my credit I don't think I've done a proxy block myself since the guideline changed. Syrthiss (talk) 13:20, 25 March 2010 (UTC)

Nor I. We actually have a bot blocking proxies now, and it only blocks 2 months at a time. –xeno^talk 13:22, 25 March 2010 (UTC)

Open proxy IPs are as varied as any other IPs apart from two things. They usually have be open, advertised, and static for at least a couple of days to be found and used; and some types of open proxies are hosted on dedicated servers which are a) often static for long periods and b) less likely to be reassigned to innocent users when the proxy inevitably closes. ProcseeBot only detects HTTP proxies which are often computers on dynamic, usually residential IPs which will typically be closed and reassigned within a couple of months. Blocks of these IPs for any longer than a few months stand a high risk of only causing collateral. These are the IPs typically used by spambots, vandalbots, Bonaparte, and Jvolkblum. Web proxies, such as those usually used by Runtshit are typically on static dedicated webhosts and can be blocked for longer without much fear of collateral for a couple of years. I've written a proxy-checking guide, linked in the see also section overleaf, to try and help educate admins on the difference between the types of proxies, how to and how not to check them, and how to determine their block length. IPs with open proxies are not all the same. Most only need to be blocked for a few weeks. -- zzuuzz ^(talk) 14:15, 25 March 2010 (UTC)

We have ips that have been vandalizing whenever they're not blocked for years on end. Some have been hives of vandalism for 3 or 4 years, a few for 5, 6, or 7 + years. When an IP has gone from warnings to short blocks up through a series of longer blocks up to a year long block, and comes back with further vandalism every time the block expires, I really see no need to give them a free vandalism spree once a year, with the inherent extra cleanup work by required of good Wikipedians and extra reblock work by admins. Say an ip has a long record of some 10 previous blocks, the last two being for a full year, what's wrong with making the next block 2 years, and if they do it again, 5 years? If something changes about the ip in the meanwhile they can request an unblock on their talk page. -- Infrogmation (talk) 21:14, 1 April 2010 (UTC)

Impossible to edit?

Resolved

 – Waladil has edited his own user page since this request was posted, so he appears not to be personally blocked and is not affected by any IP hardblock. Unable to make progress without more info. EdJohnston (talk) 18:37, 23 March 2010 (UTC)

I attend a high school in my home town, and occasionally use wikipedia. Ok, fine. I often use wikipedia, gosh. In any case, I also sometimes want to edit wikipedia. For example, this morning I noticed the "gigittybyte" included in the list of byte sizes. After a google search to confirm my hypothesis (there were literally 0 results) I decided that it was a piece of vandalism and needed to be deleted. So I clicked "edit this page" and the "this IP has been blocked for the following reasons:" page showed up.

I should point out that there were no following reasons that my school was blocked. It was just blocked. It then went on to say that if I logged into my account, I could then edit. So I logged in, you know, username: waladil password: ************. It logged me in, no problemo, I went back to the offending page, hit edit, and: blocked. At this point, I'm thinkin damn this is annoying. I used the secure login button. Still blocked. Giving up on trying to fix it myself, I decide to send a message to the guy who blocked us (it was something like sterwalkerstr, I wasn't paying that much attention, so I'm no doubt off by several letters). I head over to his page, and look for some way to talk to him. His talk page is blocked and I can't see a way to send a pm. Now I'm at the "oh hell this sucks" stage. I decide to edit my _own_ talk page and put something intentionally inflammatory there to catch the attention of an admin so hopefully they'd notice that a change needed to be made. Take a wild guess: BLOCKED!

Well, aside from the simple cathartic effect of ranting about this, the point is that apparently there is no way to get around an IP block (short of changing your IP, perhaps through proxies which my school blocks) even if you log in which is supposed to let you edit. Am I insane(er) or is there some sort of problem?

Eagerly awaiting some sort of response,
--Waladil (talk) 05:33, 26 January 2010 (UTC)

Although I'm not an expert on this, it's possible that your school's IP was hardblocked. This prevents all editing from the IP, apart from sysops and users with specially granted permissions. It's impossible to tell without seeing the block log, though. —LedgendGamer 05:44, 26 January 2010 (UTC)

That's quite possibly it. For one thing, it seems to fluctuate. Some days I can do stuff, others I can't. Also, I have internet at home, which is how I can do things on my userpage and such from home even if the school-block is being unreasonable. The really hilarious bit is that last year, a friend of mine (who was/is a bloody admin) was held out by the hardblock. Maybe Wikipedia should be more considerate of situations like that. --Waladil (talk) 06:46, 17 April 2010 (UTC)

Toolserver IPs

As it is against both toolserver and enwiki policy for bots to edit while logged out, a proposal to permanently soft-block the toolserver IPs has begun at Wikipedia talk:Bot policy. Anomie⚔ 04:34, 18 May 2010 (UTC)

Bolditude

I've just added a section to the "policy" part of the policy. I don't think it's controversial in the slightest though; being good netizens and collaborating with the IRTs of organizations which are mature and responsible enough to be responsive to abuse reports is best for everyone involved.

Right now, I suppose that we can maintain the (short) list manually, and on this page — but I hope an increasing number of organizations will be willing to collaborate this way and we'll need to find a better way to make sure than admins who are considering blocking an IP or range can be made aware of an IRT that could handle the matter if there is one. — Coren ^(talk) 14:33, 27 August 2010 (UTC)

Addresses of organizations with a responsive IRT

It's been suggested here [2] regarding the section WP:Blocking_IP_addresses#Addresses_of_organizations_with_a_responsive_IRT that "BT doesn't belong on that list, probably" - "based on recent lack of responsiveness". Might be worth removing BT, or adding a note that their responsiveness is in doubt and that any future responsiveness is noteworthy. --Demiurge1000 (talk) 18:09, 4 November 2010 (UTC)

I'll respond over there. — Coren ^(talk) 18:37, 4 November 2010 (UTC)

(The short of it, in case someone stumbles upon this conversation later, is that the range mentioned here covers web access by employees of BT and the team than handles internal abuse. They almost certainly have nothing to do with customers of BT Internet or their NOC, and have been responsive and responsible in the past.) — Coren ^(talk) 18:46, 4 November 2010 (UTC)

You may disagree, but...(indef blocking of IP addresses)

I consider virtually every block listed at Wikipedia:Database reports/Indefinitely blocked IPs to be illegitimate. As far as I am concerned, there should never be an indefinitely blocked IP address for an extended period of time. I can think of the very rare exception of a WP:OFFICE or perhaps even a OTRS action, but other than that, I would strongly disagree with indefinitely blocking IPs for any of the following reasons:

-Death/legal threats - No more than a year for even the most severe

-Tor/other open proxy - Set period of years, like 5. Definitely shouldn't be an indef block. I've unblocked tons of IPs within a couple years of being blocked as tor, after running a check and finding they are no longer Tor.

-School administrator, bot owner, or someone else requests a block for their IP - IMO this should be treated like an open proxy and block 5 years. Maybe 8-10 at the most. But even schools change IP addresses sometimes, and an indef block isn't a good idea here.

-Spam/vandalism bot - No more than a year

-Other vandalism - No more than a year at once, except perhaps in extreme cases and multiple years of being blocked I could see going up to perhaps 2 years.

-Sockpuppet/block evasion - No more than a year

-BLP/copyvio/other long term abuse - No more than a year

-Any other reason other than an office/otrs action - No more than a year

Even OTRS and office indef IP blocks, IMO, should eventually be unblocked. I think this policy more or less agrees with my assessment by stating that "IP addresses should almost never be indefinitely blocked. Many IP addresses are dynamically assigned and change frequently from one person to the next, and even static IP addresses are periodically re-assigned or have different users." I would submit that the only time the almost in almost never should be implemented is during an office or OTRS action, or perhaps some other very, very rare case. Thus, 99.9% of the IPs in the list I link above need to either be unblocked or have their block lengths changed. I would say consulting with the blocking administrator first is a good idea for anything somewhat recent (in the last year?) if for no other reason than to alert them of our policy on indef blocking of IPs, but consulting probably isn't necessary for older blocks unless there is some special note or something you don't understand about the block. Any thoughts on this? VegaDark (talk) 07:55, 2 November 2010 (UTC)

I look through that list from time to time, and unblock a few each time as time has passed on. It generally shrinks a bit each time it's updated. There are some that I've chosen to leave blocked indefinitely, knowing that I'll review them again in the future. Some are still open proxies, some are very static with some potentially very nasty people on the other end. However I agree that indefinite should almost never be used. They always say that indefinite is not forever, but unless someone goes around reviewing those blocks then they are forever.

I have been helping for a long time to tone the policy down wherever it mentions indefinite blocks, in regards to IP addresses, and I have to say it's been working. You can fail an RfA these days for saying you'll indef an IP. There are still some old school admins who are stuck in their ways. I'd like to see the behaviour which caused the block to be completely detached from the block length in most cases. There is no reason at all why someone on a dynamic IP (perhaps switching IPs daily) issuing a legal or death threat, engaging in sockpuppetry, or egregiously violating BLP, should have their dynamic IP blocked for any longer than it normally would be for writing 'poop'. The same applies to open proxies which in many cases are dynamic and short lived. I definitely disagree with standard 5 year blocks for open proxies, unless due consideration is given to how they are run and hosted. I can't see any reason to block an IP for more than five years, even if it's been static for ten. All in all I think block lengths should be based directly on how static the IP address is, and the likelihood of vandalism continuing to stem from it with their next edit. -- zzuuzz ^(talk) 08:16, 2 November 2010 (UTC)

• Agree, and I've been going through the list and unblocked more than 20 so far. As for a 5 year block for open proxies, there are numerous open proxies indef-blocked on that list, so I was trying to be much more lenient by stating 5 years. I agree in general though that the block length should be based on how they are run and hosted, not to exceed 5 years- I could definitely live with, and perhaps support, a shorter period (3 to 4 years?). Also, I agree with the behavior thing. I've seen some indef blocks with reasons such as "defamed someone" without any prior blocks in the history, and it was run-of-the-mill vandalism that I'd have issued no more than a basic warning for. I think a good start is to clear out this list, and leave a note asking administrators to reverse their block for any new entries popping up on the list. VegaDark (talk) 01:49, 3 November 2010 (UTC)

Yes, the list was due for another review, and education is a good thing. As I say, I've checked a few proxies on that list, and found some that are still proxies, but I'll continue to review them. To me however a bigger problem exists in CAT:OP, comprising some 28,000 IPs, including over 7,000 dynamic IPs. There's also the non-templated ones, including a list of thousands arbitrarily blocked indefinitely by User:DerHexer some time ago (see this). The list mentioned above excludes all indefinitely blocked open proxies and checkuser blocks. -- zzuuzz ^(talk) 12:58, 3 November 2010 (UTC)

Ouch. I'd support a mass unblocking, but we would have to have a more centralized discussion about it. Now that we have a bot handling open proxy blocks (blocking for 2 months at a time), I don't particularly see an issue with flat out unblocking all open proxies and letting the bot deal with any that are still active. Would perhaps the village pump be the best place to bring this up? VegaDark (talk) 17:51, 16 November 2010 (UTC)

Wide, rolling blocks

My ISP, Smart Telecom, has been blocked entirely from anonymous editing by their domestic customers well for well over a year, in a series of three-month and currently six-month blocks over the whole /17 range, with short gaps between the blocks. This appears to be being done to deal with a single "problem" editor, who has been blocked from their original account, but keeps popping up on different dynamically allocated IPs over that range. Not vandalism, not large-scale editing, but a particular set of tendentious editing practices by one person.

There strikes me as being a huge disconnect between this as a practice, and the idea of anonymous editing as being a "sacred principle" of Wikipedia. Is anyone able to offer any insight or suggestion as to how to bring the practice closer to the documented intention, or to bring the documentation in closer alignment with the apparent practice? Smartiger (talk) 22:36, 25 January 2011 (UTC)

IP addresses of Qatar

Can anyone check whether the two IP addresses mentioned in the Sensitive due to public relations section (82.148.96.68, 82.148.97.69) are still relevant? I see that there are no more contribution from those addresses since July 2008. Razvan Socol (talk) 05:18, 18 September 2011 (UTC)

I'd agree. Either they have moved, in which case it's probably worth a mention, or Qatar have become more liberal by using XFF or more IPs/proxies. I'm sure I've seen other Qatar IPs but I couldn't tell you if they're as critical as these used to be. These IPs are also listed at MediaWiki:Blockiptext, and used by some bots/scripts. -- zzuuzz ^(talk) 08:32, 18 September 2011 (UTC)

IPv6

This is really really early, but I'd like feedback on how acceptable would be my guideline here for IPv6. Comments are welcome on the talk page.Jasper Deng (talk) 23:30, 12 January 2012 (UTC)

List of sensitive addresses

I was just asked to block a Homeland Security address. I didn't, but I might have. I clicked on block & checked to make sure that the address wasn't on the list. It wasn't. But it is on this page (well, not this page but the main page). And yes, I now see that there is a blue link to this page - but no clue that it contains even more IP addresses, I'd assumed it was just a discussion of sensitive IP addresses and that I could see all sensitive IP addresses without clicking on another page. So, why can't they all be on the page I use for blocks? That might prevent unfortunate errors. Dougweller (talk) 15:50, 4 March 2013 (UTC)

IP Blocks Useless?

I have reason to believe that IP blocks won't do any good. A blocked user can trick Wikipedia into allowing them to edit and create an account. To prevent a user from editing, Wikipedia has to know who they are. To do this, it looks at their IP address. By changing their IP address (either by unplugging their router and plugging it back in or via the network settings on their computer), Wikipedia will think that you are someone else and allow you to edit and create an account as if you were never blocked. — Preceding unsigned comment added by 69.27.128.9 (talk) 13:37, 8 April 2013 (UTC)

I've responded already to this exact same remark that you posted over at WT:Blocking policy. Beeblebrox (talk) 18:41, 8 April 2013 (UTC)

Discussion on indefinitely blocked IP addresses

Hi, this message is sent to inform you that there is currently a discussion on the Village Pump located at Wikipedia:Village pump (policy)/Archive 112#RFC: Indefinitely blocked IP addresses which may affect this policy page. TeleComNasSprVen (talk • contribs) 20:51, 2 February 2014 (UTC)

Toolserver

With toolserver now having migrated to labs, should the entry for 185.* at Sensitive for other reasons be replaced with, say, the WMF range that includes tools.wmflabs.org (208.80.152.0/22)? —[AlanM1(talk)]— 09:39, 26 August 2014 (UTC)

Template:Sensitive IP addresses

Pursuant to a discussion at the village pump, I was minded to replace the current text that contains the "sensitive IP address" list with the list at Template:Sensitive_IP_addresses, so that the list can be synchronized with the one at MediaWiki:Blockiptext. Any objections? The template may need a few edits but I wanted to gauge support or opposition for the general concept first.Jo-Jo Eumerus (talk, contributions) 22:30, 12 July 2016 (UTC)

Block Levels

Block levels show the ways IP addresses are blocked, defined by numbers.

1 - You must now enter CAPTCHAs after your writing to submit edits. This prevents bots without registered Wikipedia accounts using the IP address from submitting edits, and it is useful mainly if there are good contributions from that IP address but spambots are using the same IP address to vandalize Wikipedia.

2 - Submiting edits will only submit them as edit requests when editing anonymously (without an account). This should be used if there is both good contributions and vandalism from the IP address.

3 - Combine 1 and 2. This should be used just like level 2 but as an increased block in case spambots are submitting too many edit requests for vandalism.

4 - (default level) This IP address or IP address range has been blocked from editing without logging in, but you can still register from that IP address or IP address range while blocked. This should be used if only anonymous users using that IP address are vandalizing Wikipedia and those who register Wikipedia accounts use the accounts on the same IP address for good contributions.

5 - This IP address or range has been blocked from editing without logging in, and you can't register from that IP address or range while blocked. This should be used just like level 4, but as an increased block in case some people are registering accounts using the same IP address or range to continue vandalizing Wikiepedia.

6 - This IP address or range has been blocked from editing without logging in and, to submit every edit, entering a CAPTCHA, but you can still register from that IP address or range while blocked. This should be used if spambots using this IP address or range are cracking passwords to edit with accounts for vandalism.

7 - This IP address or range has been blocked from editing without logging in and even submitting edits when logged in will only submit them as edit requests, and you cannot register an account from this IP address or range. This should be used if there are good contributions from Wikipedia accounts, but humans who previously vandalized Wikipedia from that IP address or range are cracking passwords of accounts and using the accounts for vandalism.

8 - Just like level 7, but you must also enter a CAPTCHA to submit such edit requests. This should be used in case spambots are also cracking passwords of accounts and using the accounts for vandalism, to prevent too much vandalism edit requests from accounts used by spambots.

9 - This IP address or range has been blocked from editing whether editors on that IP address use an account or not, and you cannot register an account from this IP address or range. This should be used if the IP address or range is only used for vandalizing Wikipedia.

There is also semi-blocking, which only doesn't disable adding sections to talk pages or Wikipedia:Sandbox where the IP address or range is blocked, and editors using that IP address or range can still remove or edit sections added from that IP address or range.

--218.186.19.241 (talk) 08:57, 11 November 2011 (UTC)

We have no method of implementing any of the levels except #4 and #5. I don't think we need such complications. When we warn a vandal, stop means stop, no mr.-nice-guy after the vandal does not stop.Jasper Deng (talk) 02:43, 12 November 2011 (UTC)

Maybe we should [talk with MediaWiki's developers] about this if Wikipedia alone can't do it. At least have #9? --218.186.15.10 (talk) 11:54, 12 November 2011 (UTC)

I forgot to mention, #9 is also implementable too (by disabling the "anon only" option on a rangeblock or singel IP block).Jasper Deng (talk) 01:54, 13 November 2011 (UTC)

Maybe #4 and #9 are implementable (I have no idea), but how can anybody know whether or not any 'good' editors are also logging in from the suspect IP address? I thought one of the characteristics of logging in was that it hid the editor's IP address, so that nobody can know where they've connected to the internet? —DIV (120.17.118.20 (talk) 15:06, 8 October 2018 (UTC))

Blocking IP address ineffective for most users....

The Internet Service Provider, ATMC or Atlantic Telephone Membership corp, I'm using has multiple ranges, and I get sent to a different one each time I take the modem out the wall and plug it back in. This is obviously a problem for CheckUser, as it is easy to avoid detection. Fortunately, I do not not disrupt, but anyone using this IP will be sent to a different one each time they unplug the modem. No IP address should ever be blocked more then a year if it is not static, as mine is dynamic and I'm sure a lot of others have that. Of course, I'm just showing you the IP address do change frequently, and I am in IPV4 mode, but I do have IPV6 mode ready.... --209.188.60.186 (talk) 00:38, 11 September 2013 (UTC)

I agree. My internet IP address from home changes automatically every time I connect & disconnect, i.e. more-or-less daily, which has been the case for probably more than a decade. (At work it's been more stable, but still even the work desktop IP address changed every two or three years for various reasons.) A few times I've been subject to 'collateral damage'. Let's say 1% of the time (I must have made thousands of edits over the years, and have been collaterally subject to a handful of blocks). But what that indicates to me is that EITHER the anonymous people on (presumably) shared IP addresses who were (allegedly) causing mischief are also usually not caught by any of the crude range blocks OR the ranges were extremely poorly specified and included both shared and non-shared IP addresses. —DIV (1.144.108.197 (talk) 05:05, 1 July 2019 (UTC))

Wikify "collateral damage"

Under "Range blocks" I suggest linking the term "collateral damage" to WP:COLLATERAL. —DIV (1.144.111.80 (talk) 10:51, 1 July 2019 (UTC))

An analogue of Checkuser to assess IP Editors' collective contributions within given IP address ranges

The article currently states: "If you propose to block a significant range, or for a significant time, consider asking a user with checkuser access to check for collateral damage – that is, for the presence of other users who may be unintentionally affected by the range block. "
So far as I can understand, the 'checkuser' tool is only able to investigate registered accounts, and therefore will be silent on contributions from 'IP editors'.
I recommend that a companion tool be developed that can provide summary statistics pertinent to IP address ranges, which might include metrics such as:

• number of IP addresses with the range (and percentage) from which edits were made by IP editors in the given range within the past week / month / year;
• number of edits made by IP editors in the given range within the past week / month / year;
• number of reversions (and percentage) to edits made by IP editors in the given range within the past week / month / year;
• number of vandalism tags (and percentage) on edits made by IP editors in the given range within the past week / month / year;
• some sort of 'clustering analysis' to determine whether reversions and vandalism tags are more likely to be associated with just a few IP addresses within the given range (and hence recommend a narrower range/set be blocked), or whether they are dispersed more-or-less randomly throughout the given range.

Although the threshold statistics might not yet exist, if analysis as above is made available then before long benchmarks will be established to say that — for example, hypothetically — on a typical IP range around 5 to 10% of edits are reverted, so a threshold of 20% of edits being reverted could be enforced as a prerequisite to implementing an IP range block.
I am concerned that current policy & practice in this specific matter is based too much on ad hoc and subjective interpretation by individual 'administrating' editors (not necessarily their fault, if tools are currently lacking), and that while registered users within the given range are able to be identified and exempted with the checkuser tool, no such sophisticated tool exists to respect the rights & contributions of good-faith IP editors.
You will note that at least the output from the proposed tool is highly aggregated, and therefore effectively anonymised, so I don't foresee any problems there. (Of course, IP editors are already warned that all contributions will be logged against their IP address anyway.)
—DIV (120.17.34.20 (talk) 02:34, 15 March 2019 (UTC))

A few tools are listed at Wikipedia:Administrators'_guide/Blocking#Blocking_a_range_of_IPs and Wikipedia:Blocking_policy#Collateral_damage...

...but they all seem to be either dead or misdirected links.

So I cannot assess what they do.

But moreover it suggests (to me) that editors who apply rangeblocks are not taking seriously the recommendations to check thoroughly beforehand and treat rangeblocks as measures of absolute last resort.

—DIV (1.144.111.80 (talk) 11:03, 1 July 2019 (UTC))

Problems with the process of blocking a range of IP addresses

I have experienced collateral damage due to blocking a range of IP addresses. I document some of my concerns at User_talk:120.17.85.26#Block_appeal. Some of the matters to be improved:

• Provide more information in the block message, possibly including a link/links so that anybody reading the block message can easily discover what contributions were made from the range of IP addresses.
• Provide (flexible) guidelines on how much collateral damage is acceptable. E.g. if there are (apparently) 2 vandals and 98 good-faith users all sharing a range of IP addresses (2% bad users), is that sufficiently bad to warrant a block? Or if there are 100 annoying (but easily, automatically reverted) instances of 'page-blanking' and 300 helpful contributions from a shared range of IP addresses, does that warrant blocking?
• Clarify the WP position, as I understand it, which is that there is indeed a 'right' for all good-faith persons to edit WP, regardless of whether they use a user-account or not.
• Provide a clear sequence of escalation (and de-escalation) steps, which would involve contacting the administering ISP/organisation so that they can take action prior to any consideration of blocking an IP range.
• Provide a public policy declaring the consequences to Editors of: (i) good-faith but poorly judged blocks, and (ii) bad-faith blocks.
• Clearly distinguish between active/intentional "evasion" by a vandal versus happenstance changing of a vandal's IP address due merely to the vandal accessing the internet through a certain organisation/ISP.

—DIV (120.17.228.20 (talk) 03:09, 20 September 2018 (UTC))

More problems:
Apparently my good-faith efforts to improve WP are not being appreciated. Well, that is an understatement. Apparently my suggestions are being called "abuse" and (possibly, depending upon interpretation) the actions of a "troll". See responses at User_talk:120.17.85.26#Block_appeal. To me this is indicative of what I suspect is a bigotry in some WP Editors with user-accounts against WP Editors without user-accounts (also known as 'IP Editors').
For the benefit of all readers, see this text from WP:SPOTVAN: "IP editors should not be approached with the assumption that they are vandals. Although many vandals do vandalize without registering an account, there are many IP editors who are great contributors to Wikipedia. Always read the actual changes made and judge on that, rather than who made the changes or what was entered in the edit summary."
Yes, that's a clear policy, but the fact that it needs to be stated at all indicates to me that there are some Editors who have been unaware or 'forgetful' of this. Although perhaps on this particular Talk page there'd be fewer of those.
Sorry that this is going slightly off-topic, but I think it relates to the general attitude of Editors with user-accounts towards IP Editors.
—DIV (120.17.228.20 (talk) 03:52, 20 September 2018 (UTC))

• No you are trolling if you continue to claim you are currently blocked, as you did here; all the evidence I can find suggests that IP has never been blocked. (the range 120.17.192.0/19 has been blocked, for blatant and repetitive removal of content without explanation) Perhaps if you demonstrated edits that were intending to improve the content of the encyclopedia, rather than making (spurious) process complaints, people would be more likely to assume good faith. power~enwiki (π, ν) 03:57, 20 September 2018 (UTC)

Trolling? "Never attribute to malice that which is adequately explained by stupidity." — Robert J. Hanlon(DIV, 120.17.34.20 (talk) 02:51, 15 March 2019 (UTC))

• Nobody has the right to edit Wikipedia, and even administrators can be blocked when they cause too much trouble. I agree that block messages should have more information, though. I've thought about proposing some changes, but I don't have any concrete ideas yet. With regard to the rest of your complaints, creating an account will prevent you from getting caught in range blocks. If you choose not to create an account, you'll have to deal with being caught in range blocks. NinjaRobotPirate (talk) 07:43, 20 September 2018 (UTC)

• Hello, power~enwiki. Though I tried to be careful, after getting consistent feedback it seems that one of the IP addresses I thought was blocked, actually wasn't. I maybe connected initially on a blocked IP address, gotten the "blocked" message when I tried to make an edit, somehow become disconnected from the internet and then automatically reconnected without me noticing, navigated around to find that IP address's talk page. .... No, not sure if that makes sense either. Actually, I'm not sure how I made that one mistake. But the other 8192 IP addresses on 120.17.192.0/19 are definitely confirmed to be blocked for 3 months, and the philosophical and practical points I've made still stand.

If you want to spend some time searching, you will be able to convince yourself that all of my edits have been made in good faith, and predominantly improve the quality of WP. If you don't want to take my word for it, and don't want to spend time convincing yourself, then nothing I can say will convince you.

—DIV (120.17.118.20 (talk) 15:24, 8 October 2018 (UTC))

• Hello, NinjaRobotPirate.

I think you are overlooking one of the guiding Principles of Wikipedia alluded to by User:Smartiger in an earlier comment on this Talk page:

""You can edit this page right now" is a core guiding check on everything that we do."

The eight enumerated Principles describe "at some ultimate, fundamental level, [...] how Wikipedia will be run". Moreover, the particular Principle stated here is the only one described as a "sacred" principle.

I accept your point that editors who deliberately/repeatedly/seriously do the 'wrong thing' give up that right, but I was never (intentionally) arguing otherwise.

—DIV (120.17.118.20 (talk) 15:24, 8 October 2018 (UTC))

Things have changed. Try to edit the main page or {{citation needed}} – you can't. Only administrators can edit them. Like I said, the solution to your problem is easy. If you choose not to avail yourself of it, you can't really complain. NinjaRobotPirate (talk) 16:54, 8 October 2018 (UTC)

Hi, NinjaRobotPirate. You say that "the solution to your problem is easy" — but this is not an issue only for me, but for all 'collaterally' blocked users. You say that "Things have changed" — are you trying to say that those guiding Principles no longer apply? Do you have any documentation as to the consensus where that was established, or citations of any new Principles that now apply instead?

—DIV (120.17.65.80 (talk) 13:09, 10 November 2018 (UTC))

Further to the above comments, the current guidance from MediaWiki is illuminating:

"People often naively suggest lock-down [prevention of editing by anonymous users, forcing them to create an account with a username and sign in prior to editing] as best solution to wiki spam. It does reduce spam, but it is a poor solution and a lazy solution, because you are introducing something which massively inconveniences real users. Having to choose a username and password is a big turn off for many people. The wiki way is to be freely and openly editable. This "soft security" approach is one of the key strengths of the wiki concept. Are you going to let the spammers spoil that?" (Bold here is the same as in the source text.)

—DIV (1.129.111.11 (talk) 15:16, 27 April 2019 (UTC))

Guiding principles

I have already quoted above a few key principles applicable to Wikipedia, such as the "sacred" guiding Principle of Wikipedia that

" "You can edit this page right now" is a core guiding check on everything that we do. "

which describes "at some ultimate, fundamental level, [...] how Wikipedia will be run".

Further to this, I also note that the Principles currently enumerated for Wikipedia include:

"Wikipedia is free content that anyone can edit and distribute." (Link here is replicated from the source text.)

It links to Wikimedia's Founding principles page, which states:

"Wikimedia projects have certain founding principles in common. These principles may evolve or be refined over time, but they are considered ideals essential to the founding of the Wikimedia projects [...]. People who strongly disagree with them are nonetheless expected to either respect them while collaborating on the site or turn to another site. Those unable or unwilling sometimes end up leaving the project.

These principles include:

1. [...].
2. The ability of almost anyone to edit (most) articles without registration.
3. [...].
4. The creation of a welcoming and collegial editorial environment.
5. [...].
6. [...]."

Point 2 is vital.

—DIV (1.144.111.80 (talk) 11:45, 1 July 2019 (UTC))

IP Address block workaround

I am just curious, what happens if a person who has gotten their IP addresses blocked creates new ones to work around the IP address block? What consequences will they face if they are extremely persistent? MaRoFu (talk) 22:19, 17 February 2018 (UTC)

The first problem is to detect it. How will you know if vandalism from IP address w:x:y:z is caused by the same person as had been vandalising previously from IP address a:b:c:d? Only if they are repeatedly vandalising the same article(s) in a similar manner would I expect it to be picked up. See also Wikipedia's policy on sock puppetry. —DIV (1.144.111.80 (talk) 12:02, 1 July 2019 (UTC))

IP Editing: Privacy Enhancement and Abuse Mitigation project

Hello all,

I’m writing to let you know about a new project, IP Editing: Privacy Enhancement and Abuse Mitigation, that the Wikimedia Foundation is starting.

Because people in general are increasingly technically advanced and privacy conscious, our users are now more aware of the collection and use of their personal information, and how its misuse may lead to harassment or abuse. The Foundation is starting a project to re-evaluate and enhance protections for user privacy through technical improvement to the projects. As part of this work, we will also be looking at our existing anti-vandalism and anti-abuse tools and making sure our wikis have access to the same (or better) tools to protect themselves.

The project page is on Meta. This project is currently in very early phases of discussions and we don’t have a concrete plan for it yet. We’d like your input. And please share with other people who you think would be interested. SPoore (WMF), Strategist, Community health initiative (talk) 18:23, 1 August 2019 (UTC)

Dynamic vs. static IP

Is there a conventient and fairly quick way to single out dynamic IPs (and IP ranges) from static ones? Tommy Kronkvist (talk), 19:35, 28 January 2020 (UTC).

In my opinion, not really. To some extent most IPs can be reassigned, so dynamic is just a question of duration. Having said that, some blacklist tools on the web will offer some suggestion, including our go-to geolocation link (whatismyipaddress.com) on the IP's talk page. Just don't take it too literally. -- zzuuzz ^(talk) 20:10, 28 January 2020 (UTC)

Registered account linked

Can a IP address block affect the registered account too? SpinnerLaserz (talk) 18:44, 1 April 2020 (UTC)

Yes, in some circumstances. IP addresses can be hardblocked by admins, globally hardblocked by stewards, and/or autoblocked by vandals. Tor nodes are also hardblocked, by the software. Most IP blocks do not affect registered users. -- zzuuzz ^(talk) 19:36, 1 April 2020 (UTC)

Lift an excessive block?

I've just noticed 195.194.15.5 (talk · contribs), which is partway through a 3 year block. It was stomped in order to stop a sock, Qais13 (royalty articles), but also silenced another constructive editor who had been working on Television licensing in the United Kingdom. Note that this IP is part of Coventry University, so that's a lot of potential readers and editors within its scope.

Three years on a shared seems excessive to stop one sock. They're just not going to be static for that long. Can we get this lifted?

Checking through the SPI from 2017, they aren't listed as using this IP anyway. However they do switch to the O2 mobile network, with 82.132.213.118 (talk · contribs) and others. I look at that and there's a whole /18 range IP blocked! Wow! Do we even do blocks that big on a major non-static ISP like that? I've spent long enough chasing Europefan (talk · contribs) and was always refused any possibility of IP blocks on their far smaller IP pool (Vodafone Germany). What did O2 do to piss off Wikipedia so much?

Is a long /18 block reasonable? Andy Dingley (talk) 23:25, 25 March 2020 (UTC)

Not sure about the block on the 195.194.15.5, but Berean Hunter’s block on that /18 looks pretty justified just by checking the CU log. You gave the answer why in your question: it’s a major non-static ISP (in the UK, which is almost as bad as India if not worse when it comes to how dynamic stuff is). If there’s substantial abuse in the form of LTAs who create a host of sleepers, an account creation block on the range is usually the best way to solve it. TonyBallioni (talk) 23:33, 25 March 2020 (UTC)

" it’s a major non-static ISP (in the UK" And that's reason to block it? We do still have some intention to recruit editors, right? This is a massive obstacle to that!

Also, can I please have Vodafone DE for Lower Saxony blocked as well, because if LTA is justification for such draconian actions, then at least let's get shot of Europefan. And if we can't justify blocking Europefan with such measures, then why are we blocking O2? Andy Dingley (talk) 23:38, 25 March 2020 (UTC)

Yes. If the abuse is bad enough, being a non-static ISP in a country where the IPs are extremely dynamic, an account creation block may be the only way to deal with the problem. In cases like this, where it appears to be a mobile internet provider, good faith users can just create an account on non-mobile networks. Most LTAs that use mobile don’t want to do this because those are easier to block. The result is that there’s very little actual disruption to recruitment for in-person events (which I think you seem to be worried about) or people who want to create an account, while squelching a lot of undesired account creation.

I don’t know anything about Europefan, but every case is different, and usually CUs have a bigger picture of what’s going on underneath the publicly available data. This looks like a pretty justifiable block to me. The university one can likely be lifted, but I’d talk to the admin who placed it first. TonyBallioni (talk) 23:52, 25 March 2020 (UTC)

• " good faith users can just create an account on non-mobile networks." And how do they get access to such a network? We've a large rural population where you can't get access other than through the mobile networks. We've a generation of millenials who don't own a computer, but live on their phones. We've 5G rolling out which will make this even more mobile-first for connectivity.

Is WP really so high-handed as to take that sort of "Go away plebs, you don't have Real Internet" attitude? Or is this just because they're in the UK? Andy Dingley (talk) 00:06, 26 March 2020 (UTC)

The entirety of the T-Mobile IPv6 network has been blocked both locally and globally for the better part of 3 years for the same reasons. No, this is not a UK thing, my point was that the handling of IP assignment in the UK, especially on mobile, unfortunately makes these blocks more necessary. This is unfortunately what often happens with very dynamic ranges with multiple abusers.

In the UK, 98% of households with internet access had home broadband compared to 77% who had used internet “on-the-go” (Source). For the 2% of internet using UK household without home broadband, WP:ACC remains an option. We’re not going to change the considerations we make for 2% of the internet using population. It sucks that they can’t directly register, but sometimes it’s the only option to combat issues here. TonyBallioni (talk) 00:43, 26 March 2020 (UTC)

• re the ONS reports (that's an old one), they're a whole can of worms (UK rural development campaigning has long issues with their methodologies on these: far too much is based on counting people who have either apples or oranges, then politicians making claims that those people have both apples and oranges. For one thing, they're stacking the results by counting people who have internet but doesn't use it as "having" (there's a population of older people who have cable TV, but no router), and downplaying businesses in central Wales who have only the bandwidth of wet string.)

We do not have 98% broadband availability. (the report doesn't even say that. It says 98% of those with access have broadband access)

The ONS definition of "broadband" is flakey.

That same ONS report also highlights how some people might have potential access to broadband, but what they're actually using is their handheld device, connecting via the 4G network, not the wired networks or 21CN. These are the group who we're locking out right now. Andy Dingley (talk) 13:16, 26 March 2020 (UTC)

In 2018, of all households in Great Britain, 90% had access to the internet. This has started to level off over the last few years, rising by only 1 percentage point since 2016. However, household internet access has increased by 33 percentage points since 2006 (Figure 1).

Fixed broadband has continued to be the most popular type of household internet connection since 2015, with 98% of households with internet access having this type of connection in 2018.

No. It states that 90% of households have any type of internet access (which would include mobile.) Of those 98% have access to a wired connection. The most recent numbers (2019) don’t go into as much detail. I’m sorry, but I’m going with the U.K. government statistics over anecdotal evidence. The overwhelming majority of good faith users in the United Kingdom who would be prevented from making an account on mobile have access to at least one different unblocked network daily. There is absolutely no reason to remove this block. TonyBallioni (talk) 14:05, 26 March 2020 (UTC)

• 98% of 90% is 88%, not 98% How about you go with government statistics, over mis-read government statistics.

The overwhelming majority of good faith users ... have access to at least one different unblocked network daily or anecdotal evidence. Andy Dingley (talk) 15:31, 26 March 2020 (UTC)

I said 2% of households with internet access. The 10% don’t have any internet access, so your mobile point doesn’t apply to them since they don’t have mobile internet either. That 10% isn’t ever going to edit Wikipedia unless it’s at a public library or coffee shop. If you look at the raw data, and not just the summary, it’s even more clear that it considers mobile use as part of the determination of home internet access. There are no grounds to say that blocking a mobile range prevents a significant percentage of individuals in the U.K. from creating an account on another network. The data simply doesn’t support those claims. TonyBallioni (talk) 15:54, 26 March 2020 (UTC)

Another way to deal with the problem

" Yes. If the abuse is bad enough, being a non-static ISP in a country where the IPs are extremely dynamic, an account creation block may be the only way to deal with the problem. "

Actually there are plenty of ways to deal with the problem.

1. Don't let anyone edit (no matter whether IP, or VPN, or logged in, or admin). Just like the hospital that has closed down where no patient ever dies, this is a surefire way to completely eliminate vandalism!
2. Protect (or semi-protect) the articles that are being attacked by some vandals.
3. Prevent specific users (including those defined implicitly by their IP address) from editing an article directly, but let them do pretty much everything else.

The last point is one I want to raise as a suggestion. Most vandals want to see their handiwork implemented in an obvious public place.

Applying a 'semiprotect' style block to a user* would stop that user from editing an article, but would allow them to post in the article's Talk page, post on other users' Talk pages, and so on. The functionality may not exist yet, but it should be easy enough to create, because the functionality already exists for articles (and for some languages it seems to be the default for all articles).

Another possibility is to allow the user* to edit articles directly, but to withhold their changes until they've been moderated by some administrator, or perhaps just by anyone with a WP account.
This would perhaps mean more work for the community in moderating, but would still prevent vandalism from making it to the public, and thus would presumably also reduce the frequency of attempted vandalism. An advantage would be that the good-faith general public (often on dynamic/shared IP addresses) would still be able to edit articles as they see fit, rather than having to describe changes that they recommend (in the semiprotect rangeblock mode), which would then have to be double-handled by logged in WP editors in order to firstly moderate the suggestions and then actually implement the changes in the article.

* For "user", read "user or users, including those defined implicitly by their IP address".

I can foresee some 'helpful' administrators chipping in with, "but that won't work all of the time, so there's no point doing it." The short answer to that is, "NO!" The longer answer is that every block should be the minimum that can be justified (which is the WP policy/guideline, BTW); if that subsequently proves to be inadequate in some very specific cases, then the next level of action is taken, and so on.

—DIV (1.129.106.87 (talk) 10:42, 24 February 2021 (UTC))

Arbitrary IP blocks

I posted here with regards to the random/arbitrary blocks handed out to ISP users:

https://utrs-beta.wmflabs.org/public/appeal/view?hash=f7b4586f83f455ce04353ee36e8aed69

However my points (that there is no reason to place a blocking cookie just because I have used a VPN, and that if you are going to block IP ranges that are dynamic and potentially shared by several hundred million people, then the block is essentially pointless) were completely ignored.

Could someone please explain the rationale for blocking users for using a VPN, who are now no longer using a VPN? And could you also explain the purpose of handing out long blocks to IP users who will with absolute certainty not be using the same IP when the ban expires? Unless, as I observed at that page, you want to block the entire ISP.

125.162.47.233 (talk) 09:43, 13 February 2021 (UTC)

Have just restarted my router and got a new IP. All three IPs are me at the same physical location; however if I had been unfortunate enough to try and edit Wikipedia from the first IP then I'd be banned from account creation. 36.76.106.29 (talk) 09:47, 13 February 2021 (UTC)

Still me, same router same location. 180.251.14.234 (talk) 09:51, 13 February 2021 (UTC)

And me again. How much sense do these ip range blocks make to you, exactly? 180.241.120.210 (talk) 09:56, 13 February 2021 (UTC)

What's your point? That account creation should be mandatory? - wolf 17:36, 13 February 2021 (UTC)

How do you propose to create an account when account creation is prohibited by the blocks, which run for 3 or 6 months, not 'a few hours' as per the policy? https://en.wikipedia.org/wiki/Special:BlockList?wpTarget=36.71.138.122&blockType=&limit=50&wpFormIdentifier=blocklist

And let's say I'm using a VPN, because there's a good chance that anyone in Indonesia who is going to be fluent enough in English to be contributing to Indonesia will be, and I try to create an account, which is forbidden because VPN, so I logout of my VPN but I'm still poisoned with a 'you were using a VPN yesterday' cookie, so I still can't create an account. 180.248.160.7 (talk) 19:33, 13 February 2021 (UTC)

You are raising several issues. Your main point about cookie blocks is well founded. Users can request unblock, wait a bit, clear the cookie, or more likely for VPN users, just open another incognito window. VPNs are blocked for good reasons. Long blocks, and long range blocks, are handed out for long term disruption, which has usually been going on for years. Wide range blocks are also handed out for significant disruption. If I recall correctly, we have previously had to block just about every range you are currently using, at the same time, due to significant disruption. If we have to block half of Indonesia, even though we don't want to, then that's what we have to do. There is an element of luck and incompleteness with any large range blocks, and they're not always easy, but I wouldn't usually call them arbitrary. -- zzuuzz ^(talk) 19:49, 13 February 2021 (UTC)

I'm not suggesting that VPNs be unblocked, I was merely observing that the use of one shouldn't prevent you creating an account after you sign out from one.

Given that all the ranges I'm using are random IPs I get when I sign in to the same ISP in the same spot (home, not mobile), it would make sense to ban them ALL in one go, if you are going to do it, but it is clear that banning typically a few hundred IPs out of a range of maybe hundreds of thousands or millions of IPs that a given user on the same ISP is likely to get, for a period of three or six months is just going to frustrate random innocent users across a wide period of space and time, and not actually prevent any specific individual who is actively trying to be disruptive (since they can just re-connect and get a very different IP that is not blocked).

I took a look at the contributions log: https://en.wikipedia.org/w/index.php?title=Special:Contributions/36.71.138.0/23&offset=&limit=500&target=36.71.138.0%2F23

It appears to relate to https://en.wikipedia.org/wiki/Wikipedia:Sockpuppet_investigations/Manda_1993/Archive

It seems like this user was editing at 36.71.138.242, ‎36.71.139.150 , ‎36.71.138.183 on 9 November. But in the middle there is the innocent user ‎36.71.138.4 / ‎36.71.138.100 , who appears to make edits frequently at https://en.wikipedia.org/w/index.php?title=2020_in_Japan&action=history who is also hit by the block

Also if you look at the edits on https://en.wikipedia.org/w/index.php?title=User:Danu_Widjajanto/Vandal_Langsa_Log&offset=&limit=500&action=history you can see that the same vandal is editing on 180.251.46.98, 125.162.47.93, 180.241.14.217, 180.241.88.94, 125.165.36.43, 36.71.142.41, 36.79.7.225, 125.165.51.126, 36.79.3.243, none of which are currently blocked. So given that the block I posted about is directed at this user, it's not clear why all of the possible IPs are not blocked, since I'm quite sure he has the same state as me, i.e. a random IP from a wide range every time he connects. The blocks I mentioned are clearly not effective. Protecting the relevant pages and IP blocks to the SINGLE IP only for perhaps 48 hours might have some use, but the only other option is clearly to block entire vast swathes, otherwise this doesn't really do anything 36.76.108.179 (talk) 21:11, 13 February 2021 (UTC)

Two points I would pick up on

1. "blocks, which run for 3 or 6 months, not 'a few hours' as per the policy" — I too am dismayed by the blithe dismissal of official WP policy and guidelines by Editors who've implemented IP range blocks. Typical responses by administrators who've implemented draconian blocks include (each of which I have personally seen occur):
   • "That policy/guideline no longer applies at this time." (But no evidence provided.)
   • "This is an exceptional case, so the policy/guideline doesn't apply." (But no details provided to demonstrate the exceptional nature.)
   • "You are complaining about the process, but any matters concerning the process cannot be used to justify the lifting of a block." (Surely it's legitimate to ask for a block to be lifted or modified if the official policies/guidelines haven't been followed.)
   • "You are complaining a lot about this block, so you must be the vandal." (But no evidence provided.)
2. "The blocks I mentioned are clearly not effective." — I have raised this in the past too. Unless WP goes crazy and bans every IP offered by the vandal's ISP (which I am not proposing), then it's not going to stop the vandal. And they can then pick up their friends' mobile 'phone (on another network) or whatever. I myself have a couple of SIM cards with data plans on different ISP's.

Further to the first point, I would like the actions to match the policies/guidelines. Personally I would argue that it is better for the draconian blocks to be reined in to match the relatively reasonable policies/guidelines. But if we can't manage that, then let's change the policies/guidelines to match the current draconian reality, so that at least the nature of these blocking actions will be plainly described for all who read it to blush about.

—DIV (1.129.106.87 (talk) 11:05, 24 February 2021 (UTC))

Wikify "collateral damage"

I say again: under "Range blocks" I suggest linking the term "collateral damage" to WP:COLLATERAL. —DIV (1.129.106.87 (talk) 11:16, 24 February 2021 (UTC))

How much of a priority is this?

"There strikes me as being a huge disconnect between this as a practice, and the idea of anonymous editing as being a "sacred principle" of Wikipedia. Is anyone able to offer any insight or suggestion as to how to bring the practice closer to the documented intention, or to bring the documentation in closer alignment with the apparent practice?"

The above quote is from more than a decade ago. It got no replies, and — as far as I know — there was no constructive action taken since then to match principles and practices either.
—DIV (1.129.106.87 (talk) 12:18, 24 February 2021 (UTC))

Definition of 'long-term'

This page fails to define what 'long term' actually means when blocking an IP address, and I feel that it should. My own view is that blocking IP addresses for up to 3 or 6 months is, on occasions, quite appropriate for certain types of egregious editing behaviour, and is less than commensurate with the blocks we happily administer to registered users for far minor indiscretions. I would not regard such a time period as 'long-term', though might if it reached over 6 months to a year or more. Maybe others would take a different view? Either way, I feel the meaning of 'long-term' ought to be spelled out, and would suggest that >1 year would fit that meaning. Nick Moyes (talk) 22:16, 10 July 2021 (UTC)

How do I report an IP?

I found Wikborg Rein article was edited(and manipulated) by an IP belonging to that company.[3] 212.71.78.130 How can I report it? --Greatder (talk) 11:15, 2 September 2021 (UTC)

Queries like this should be asked at the Help Desk. Conflict of interest related editing may be reported to WP:COIN. Be aware of WP:OUTING. 331dot (talk) 13:52, 2 September 2021 (UTC)