Archives

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 s/Archive1 s/Archive2 s/Archive3 s/Archive4 s/Archive5 s/Archive6 s/Archive7 s/Archive8 s/Archive9 s/Archive10 s/Archive11 s/Archive12 s/Archive13 s/Archive14 s/Archive15 s/Archive16 s/Archive17 s/Archive18 s/Archive19 s/Archive20 s/Archive21 s/Archive22 s/Archive23 s/Archive24 25

I normally have the patience of Job with SPI, but I AfD-nommed several articles before I realized there were numerous socks involved. Since the socks concentrated on only a handful of articles, those 5 articles are the bulk of the evidence, and if they get deleted it will be hard for non-admin clerks to review and may also make it harder for admins to review. The SPI in question is WP:Sockpuppet investigations/Keevaymusic. (I also requested CU but I'm beginning to think CU may not be possible, in which case the evidentiary articles become even more important.) Softlavender (talk) 09:21, 28 March 2018 (UTC)[reply]

@Softlavender: This seems straight forward so I will take this. I have applied standard block on the recent two accounts, and I will check around again when I return home. Alex Shih (talk) 09:48, 28 March 2018 (UTC)[reply]

I've already closed it, but you can do what you wish with it, Alex.--Bbb23 (talk) 14:22, 28 March 2018 (UTC)[reply]

Brilliant, I think this is good enough for now! Thanks Bbb23. Alex Shih (talk) 15:20, 28 March 2018 (UTC)[reply]

• Thanks. I'd personally feel calmer if the other socks were blocked too, since if the user remembers those passwords he can simply go back to one of them. Also, there's also the WP:COI element to be dealt with -- all of the still-existing articles the farm created seem to be about the user's own relatives (as noted in the SPI). There's also block evasion to consider -- all of the socks were evading the permablock of the master -- should their articles be CSDed on that score? Softlavender (talk) 02:31, 30 March 2018 (UTC)[reply]

In response to a particular long-term sockpuppet I'm considering setting a hardblock on an IPv4 /19 range (~8,190 addresses) to block edits from registered accounts. I know about collateral damage and know to ask a Checkuser before doing such a thing, but my question is more generally: what are our norms on this, and/or what's everyone's opinion as to weighing the potential for collateral damage? As far as I can tell from anon edits the range hasn't been used at all except for this one sockmaster since at least 2011. This is a user who creates multiple sleepers at a time to go around restoring their old (sometimes very old) versions of pages, disregarding any progress or discussion. They quickly create a new account or cycle to a new IP if they are blocked and don't really make much of an effort to evade detection at all, strongly suggesting they don't care if they're blocked and 7 years of data indicate they intend to continue. The topic they frequent is popular among other unregistered editors, meaning semiprotection also leads to collateral damage, and the list of articles they frequent is quite long. Two of their IP ranges I've tracked for a while are currently anonblocked for terms measured in years, and it has stemmed their IP disruption but doesn't stop them from creating new accounts. My thought is that the possible collateral damage of hardblocking the range (unknown) should be weighed against the damage from possible mass-semiprotection of their frequently disrupted pages (somewhat high to extreme), versus myself and several other users playing a long and tedious game of whack-a-mole over many years. Before I do anything further I'd like to know what other clerks and CUs think. Ivanvector (^Talk/_Edits) 17:49, 2 April 2018 (UTC)[reply]

I haven't mentioned which case it is because I'm interested in general opinions, but I can get into specifics if it benefits the discussion. Ivanvector (^Talk/_Edits) 17:50, 2 April 2018 (UTC)[reply]

I'd strongly encourage you get a CU to take a peek, as collateral isn't always obvious, any IPBE can be dealt with, and the accounts might not all be coming from where you think they're coming from. This is the norm. Quite simply, you can't properly assess the collateral if you don't know who's editing on the range. You can take a guess, but you'll probably be missing something. This is especially true when a range is already anonblocked and you won't have any idea about the activity, because only registered users are using it. -- zzuuzz ^(talk) 20:01, 2 April 2018 (UTC)[reply]

Sure, I'd ask a checkuser first, I thought I said that originally. I'm really asking for opinions on whether this sort of thing is justified, like, if there are x1 registered (unrelated) accounts editing on the anonblocked range, versus x2 articles that are frequently disrupted and would need protection to prevent disruption (multiplied by y2 unrelated IPs interested in those articles), is preventing disruption worth the inconvenience to those x1 users of seeking IP block exemption? I know there's not going to be a bright line rule here, just a judgement call. Ivanvector (^Talk/_Edits) 13:43, 6 April 2018 (UTC)[reply]

I've never seen any policy or guideline give advice on range blocks. It seems to be one of those things that's left entirely to admin discretion. NinjaRobotPirate (talk) 02:18, 7 April 2018 (UTC)[reply]

Ivanvector, over the period since 2011, how many named sock accounts do you think have been created from the range that you are proposing to hardblock? (I'm not a clerk or CU). EdJohnston (talk) 02:43, 7 April 2018 (UTC)[reply]

Looking at about 50 checkuser-confirmed accounts in the case archive, plus another 25 in my block log that link to the SPI although there could be overlap in those two sets, and only since about this time last year. The user only edited from IPs up to that point, but after I investigated an ANI report and figured out an effective range to block, they switched to mass account creation. Of course there is no CU confirmation that the user and the IPs match, but the behavioural evidence is very strong, one of the most obvious cases I've seen. Ivanvector (^Talk/_Edits) 14:30, 9 April 2018 (UTC)[reply]

This is in big red letters on the main page, can someone fix it? Thanks. Doug Weller talk 12:59, 9 April 2018 (UTC)[reply]

I don't see what you're referring to, Doug Weller. ​—DoRD (talk)​ 14:45, 9 April 2018 (UTC)[reply]

According to WP:VPT there was a site-wide typo thing earlier, which has apparently now been fixed. -- zzuuzz ^(talk) 14:48, 9 April 2018 (UTC)[reply]

Thanks. When I saw it had gone, I presumed that it must have been something like that, particularly as I hadn't been able to find where it had been introduced. Doug Weller talk 14:50, 9 April 2018 (UTC)[reply]

Anyone know if this is related to anything? Ivanvector (^Talk/_Edits) 21:04, 11 April 2018 (UTC)[reply]

FYI Ivanvector, see the current WP:AN and the last two archives of it (probably more, but that's as far back as I went) for headings with "white space" in them. Some of the rangeblocks being asked for are out of the question, iirc. ​—DoRD (talk)​ 21:31, 11 April 2018 (UTC)[reply]

Oh yeah, I did see that. Yeah, let's block all of China over someone inserting and removing single newlines, that seems like a reasonable response. :) I was more just wondering if anyone's bothered to create a container case for these reports. Ivanvector (^Talk/_Edits) 21:45, 11 April 2018 (UTC)[reply]

There's one in Brazil that I've range blocked a few times. That's the one that GoodDay is usually on about at WP:AN. I don't think any other white space vandals are related, but some people apparently do. I think it's just a "thing" recently. NinjaRobotPirate (talk) 01:57, 12 April 2018 (UTC)[reply]

Could someone replace the double tags on wikinger/calvero socks with the global locked sock one? (i would do it myself, but I'm not autocomfirmed. Also, could someone fix the tagging on atlanticdeep socks? Sincerely, Cæsey ^Shine! 05:20, 12 April 2018 (UTC)[reply]