Wikipedia:Sockpuppet investigations/Gantlet/Archive

31 October 2015

 * Suspected sockpuppets


 * User compare report Auto-generated every hour.
 * Editor interaction utility

In March 2010, blocked Rajeshbiee as a sock of Gantlet (YM was a CU back then). Gantlet only merited a short-term block. Now, years later, the Rajeshbieee is active, and (per the name and behaviour, as explained by here) is pretty much a WP:DUCK of Rajeshbiee. The problem is that both Rajeshbieee & Gantlet are currently active... if Rajeshbiee was a CU-confirmed sock of Gantlet and Rajeshbieee is a DUCK of Rajeshbiee, then the logical outcome is that Rajeshbieee is a sock of Gantlet... however, the entire basis for that conclusion is a five-and-a-half years old CU result that was poorly documented at best, and since both Gantlet and Rajeshbieee have edited very recently, a new CU check is a possibility.

I'm still unsure of what action should be taken if this turns out to be confirmed (perhaps indef the mostly retired Gantlet account and warn Rajeshbieee against using any account other than Rajeshbieee?). ☺ ·  Salvidrim!   ·  &#9993;  17:55, 31 October 2015 (UTC)

Comments by other users
''Accused parties may also comment/discuss in this section below. See Defending yourself against claims.''

I recently got a notification telling that me & some user named are sock puppets. I don't even understand how you came to the conclusion. This is my only account & I don't even know who this other person is & I've been editing & contributing sincerely for Wikipedia for a long time. I can see that he's given his photo & everything on that page & I'm sure that I'm not him. Gantlet (talk) 17:29, 1 November 2015 (UTC)

Clerk, CheckUser, and/or patrolling admin comments

 * Self-endorsed - ☺ ·   Salvidrim!   ·  &#9993;  17:55, 31 October 2015 (UTC)
 * Before I ran the check, like, I speculated what should be done if the two accounts were confirmed. My view was quite different from Salvidrim!'s possible action (he doesn't state it categorically). I would have indefinitely blocked both accounts. However, the two accounts are not confirmed. Therefore, I'll leave it up to Salvidrim! or another clerk to decide what's best.
 * and are.
 * Gantlet and are ✅ to each other. However, .--Bbb23 (talk) 04:03, 1 November 2015 (UTC)
 * : Adj48n is already blocked as a CU-confirmed sock of Smileverse; Gantlet is a lot older than Smileverse. Thoughts on this? If you're confident in the CU results, I propose to merge the SPIs under Gantlet, indef Gantlet as a CU-confirmed master, retag Smileverse socks as under Gantlet; although I'm having a hard time conclusively linking behaviour of Rajeshbieee with Gantlet or Smileverse socks., you're familiar with all these cases, can you point out behavioural similarities in the edits of Rajeshbieee or Rajeshbiee and Gantlet, Adj48n, or other Smileverse socks? ☺ ·  Salvidrim!   ·  &#9993;  04:15, 1 November 2015 (UTC)
 * I'm confident in the CU results, but I would not block based on the technical evidence. Unless there's a clear behavioral connection, I'd leave it alone. Ironically, I was the one who confirmed Adj48n. I looked at the contributions page and saw SmartSE's block but never looked at the userpage. No harm in waiting for the inestimable SpacemanSpiff's views, but I think your views are important. If we err, we should err on the side of not connecting the two masters.--Bbb23 (talk) 04:30, 1 November 2015 (UTC)
 * Interesting... you've basically CU-confirmed Gantlet to blocked sockmaster Smileverse, but despite Gantlet remaining active to this day, you would rather see the account continue unblocked? I'm not sure I quite follow, I'm sorry. :/ ☺ ·  Salvidrim!   ·  &#9993;  04:35, 1 November 2015 (UTC)
 * You misunderstand. I would not block Gantlet based solely on the technical connection with Adj48n. That doesn't mean I wouldn't block Gantlet for other reasons.--Bbb23 (talk) 04:51, 1 November 2015 (UTC)
 * There is the possibility that all are linked because of the PR agency -- they all have film related clients, BLPs etc. It could also be that they are all part of the same "agency collective" -- multiple editors who are either directly or indirectly working with at least three PR agencies / SEO marketeers. The difference in point-in-time snapshots of CU findings also suggest the same -- there's a strong linkage at one point, and a weaker linkage at another, hinting at some sort of common activity in phases. If you look at the specific topics of the various editors, there's a macro-level link, but then there's segregation at the second level with individual distribution of articles based on sub-topic. A look at Rajeshbiee/e's contributions suggest paid promotional editing (just like all the Smileverse accounts) but there's also a hell of a lot of fanboy stuff editing (this is not uncommon in the Indian film paid editing area). The same with Gantlet, there's the "my hometown is the best" sort of editing mixed with sort of promotional editing around a couple of film topic areas. My opinion is that there are multiple editors in some sort of collective involved which gives different IPs/geolocation at times, and then suddenly converges for either role-account type editing or meeting-to-discuss-client-activity type editing. It's very similar in behavioral pattern to this SPI where I'm sure there are at least two geolocates with 3-4 editors involved. I'd expect that most of this group's activity is based out of Bangalore and Kochi and possibly a third common point. Given my lack of technical knowledge of the group, this analysis is just my expectation based on a look at the original two after Salvidrim! posted to my talk page on it, and now a re-check on Smileverse -- and for clarity, I did not expect Gantlet to be linked to Smileverse, but I now see a possibility sort of a behavioral possi-likely. As for promo editing by Rajeshbieee, some of it has already been caught by, so alerting him. &mdash; Spaceman  Spiff  04:44, 1 November 2015 (UTC)
 * Most of Rajeshbieee's work needs assessment by an expert--they're Tamil language films, which will have only Tamil language sources. I have no way of telling from the information readily available which are actually notable. They're the sort of articles which could as well have been made by a hobbyist or a press agent. I would expect an Indian paid editor to be interested in his country's films, just as are the rest of us--and most of WP's film articles from all countries are every bit as much subject to fan coi. Good sourcing for films from India is often a problem even worse than the language barrier, as it is my understanding that all newspapers from that area are unreliably promotional  for this industry.  (I'm going to try to find something reliable in English at NYPL Performing Arts, but that will take a week or two. (And if I do find it I'll really need to check & properly source every single article in the entire area (and see what we're missing,) or help someone else do it.)  DGG ( talk ) 06:53, 1 November 2015 (UTC)
 * There are two parts to the editing here: (a) Old movies (mostly Tamil) and for these online sourcing is not going to be easy to come by, but the editing in this space is more of fan editing, (b) Marginal actors and new films -- it will be incredibly easy to find online sources for anything like this if they are notable, so the lack of such sourcing implies some sort of COI and given the G11 nature of some of the stuff it's easy to say that there's paid editing involved. Language should not be a problem in this case as the English media covers all commercial and artsy Tamil and Malayalam films, while there may be additional information in vernacular sources, it's just that -- I've provided my behavioral analysis above, will let or  make the final call. cheers. &mdash; Spaceman  Spiff  12:30, 1 November 2015 (UTC)
 * I have a few random comments. I went back and looked at the history of Adj48n, and my finding wasn't confirmed, but likely. I changed the tag on the userpage to confirmed, but that's something I sometimes do with a likely finding. My limited experience with paid editing is it's often very hard to pinpoint whether there is one master or multiple masters, or even worse from a socking perspective, whether it's coordinated by one or more people and almost no one matches technically because they're all in fact different people. Some of the IPs used by the Smileverse socks are dodgy, meaning they obfuscate one way or another the true location of the account. None of the IPs I saw in the Gantlet check were like that. They were consistently from India (as were many of the Smileverse socks). Location is often unreliable, not just in India, but in other countries. It's inherently dicey, and it's compounded by different geolocators coming up with different results. FWIW, my instinct is that Gantlet and Smileverse are different masters, but instinct is not evidence. One comment on the Gantlet check. They key finding was that Gantlet and the other account were possilikely. Other CheckUsers might have come to a different conclusion based on the data. Could have been less conclusive, e.g., only possible, or more conclusive, e.g., likely. There may even be CheckUsers that would have confirmed the connection. The data is in a gray area, which is one of the reasons I waited so long to post my findings. I wanted to think about the data longer. It wouldn't surprise me if YellowMonkey, who of course is no longer around to ask, might have had a lower threshold for a confirmed finding. If anyone wants my non-evidentiary instinct on that relationship, I would say that the Gantlet and Rajeshbieee should be blocked as socks of each other, irrespective of whatever we do with the Smileverse question.--Bbb23 (talk) 13:01, 1 November 2015 (UTC)


 * Alright, I'll heed Bbb23's sage advice -- Rajeshbieee as a sock of Gantlet, and Gantlet  as master of Rajeshbiee and Rajeshbieee. However, now that was can tag suspected alternate masters, I will also tag Adj48n as a suspected sock of Gantlet and Gantlet as a suspected sock of Smileverse . ☺ ·   Salvidrim!   ·  &#9993;  17:41, 21 December 2015 (UTC)