Wikipedia

Wikipedia talk:Harassment/Archive 3

< Wikipedia talk:Harassment
Latest comment: 9 years ago by Spartaz in topic WP: OUTEX
Archive 1Archive 2Archive 3Archive 4Archive 5→Archive 10

WP: OUTEX

Latest comment: 9 years ago245 comments55 people in discussion
Consensus is against these proposals and policy/acceptable practice remains unchanged. Spartaz Humbug! 13:16, 4 May 2015 (UTC)

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

Seeing as the RFC concerning other accounts at other websites has closed with sometimes it is okay to post such information, is there any agreement as to what information, and where it may be posted. I'm looking to clarify and improve the wording regarding what would be considered an exception to outing. --Kyohyi (talk) 15:10, 26 February 2015 (UTC)

Yes agree this is an exceedingly important next step. I would propose that one situation in which it should be okay to post other accounts is when the other accounts are involved in the business of paid advocacy editing (such as account on Elance, Fiverr, Odesk, etc). That this should be allowed at both WP:SPI, WP:ANI and WP:COI noticeboards. In this situation making statements that the account belongs to a Wikipedia user is typically not needed. Most it is "1) this Elance account took a job to write X 2) Wikipedia user Y has just written X. Can we get a CU?" Doc James (talk · contribs · email) 19:53, 26 February 2015 (UTC)
A good exception would be when someone someone (using the same username as their wikiaccount) makes a canvassing post in a public and visible forum, or a post smearing or attacking another editor. It seems legimate to point out that someone sharing a username with an editor is doing this offsite. This seems to be reflective of comments in the RfC.Bosstopher (talk) 15:47, 27 February 2015 (UTC)
Yes agree that is another reasonable use. I guess the question is should we have a RfC for each situation in which linking to another account on another website is okay? Doc James (talk · contribs · email) 00:43, 28 February 2015 (UTC)
Would it be considered disruptive to have multiple RfC's on the same page? Because I've got a few ideas for parts of OUTEX I want to propose.Bosstopher (talk) 16:57, 1 March 2015 (UTC)
When I made this section I was thinking that we'd have a general discussion which would list a number of cases, then have RFC's on specific wording around each of those cases. --Kyohyi (talk) 19:32, 4 March 2015 (UTC)
I am thinking one RfC for each instance. So yes multiple RfCs is fine. Doc James (talk · contribs · email) 00:37, 2 March 2015 (UTC)
The potential downside here is that you won't be able to think of all possible scenarios in advance. If you try to make a list of all scenarios where it is okay to link to an off-site account, the assumed implication will be that it's forbidden for any case that isn't listed. That prevents the organic development of consensus. If such a list is eventually made, please add a sentence to say that it makes no claim to being exhaustive. Andreas JN466 13:50, 4 March 2015 (UTC)
Yes the first question will be "can we get consensus for any case" Doc James (talk · contribs · email) 18:36, 4 March 2015 (UTC)

Dated comment to prevent this being archived before being closed. Thryduulf (talk) 14:16, 28 April 2015 (UTC)

I propose the addition of "Link to accounts on other websites through which paid editing appears to be taking place may be used on Wikipedia during discussion at WP:COIN. This includes accounts from websites such as Elance or from websites that specialize in Wikipedia editing such as WikiLinkPro among others."

Amended proposal:


WP:OUTING already says: "Posting links to other accounts on other websites is allowable on a case by case basis."

Add, following that: For example, links to freelancing sites or other sites involved in soliciting paid editing, where editors disclose their Wikipedia usernames, take paid editing jobs, or provide examples of Wikipedia articles they have written for pay, or the like, may be used to address issues related to paid editing. (note - amended proposal per discussion below Jytdog (talk) 21:48, 4 March 2015 (UTC))


Support

  • Support as proposer. The WMF has made it clear that they do not have the personnel to enforce our terms of use with respect to undisclosed paid editing. They feel that the community should be involved. For the community to be able to take this on we need to feel comfortable discussing issues openly. Doc James (talk · contribs · email) 03:36, 28 February 2015 (UTC)
Support amendment. Doc James (talk · contribs · email) 22:55, 4 March 2015 (UTC)
  • Support This makes sense. Many freelancers on these sites actually link to their WP user pages and to articles they have written for pay. Used in conjunction with actual edits to articles, this would provide ~pretty~ clear proof of paid editing, and help us deal with paid editors using multiple socks - many of these editors have a stable username here and create one-off, throwaway accounts here per job. But they need to fix an identity somewhere and those brokers are the place where they want to establish their reputations. Jytdog (talk) 01:39, 2 March 2015 (UTC) (add a bit and strike a bit - have learned more since my !vote Jytdog (talk) 14:10, 4 March 2015 (UTC)) (striking, too broad as stated; needs limits per discussion below Jytdog (talk) 11:43, 25 March 2015 (UTC))
  • Support Absolutely. §FreeRangeFrogcroak 03:59, 2 March 2015 (UTC)
  • Neutral I've been thinking about this for a while and I decided to withdraw my support. While I do believe that we need to be more aggressive towards the demographic of more blatant COI editors, the core of the COI guidelines is broken to begin with (because it relies on good faith and has no preemptive "teeth") so this will be just a band-aid on top of a gaping wound. There is no way that I can see to reliably make an exception to WP:OUTING that we can be sure won't affect other editors, and privacy of the overall editor community is more important than fighting the Wiki-PRs of the world. This will come back and bite us if we expand what OUTING considers to be allowable on a "case by case basis". What happens when we start playing sleuth on the internetz because we believe someone is operating under a COI, out them and it turns out we were wrong? It's just not worth it. I commend Doc James (and Jytdog for his efforts as well) for bringing this idea forward for discussion but we need to attack this problem from the bottom, not the top. §FreeRangeFrogcroak 18:28, 13 March 2015 (UTC)
  • Support. Good idea, commonsensical. Would tweak to allow utilizing of such links at ANI and other official noticeboards. Coretheapple (talk) 14:13, 4 March 2015 (UTC) To clarify, per Guy's comment below, it is important to distinguish between "outing" as a form of harassment and punishment and self-disclosure via advertisements, which this passage is specifically designed to address. One caveat: see my comment re "joe jobs" in the discussion section.Coretheapple (talk) 21:43, 4 March 2015 (UTC)
Re the amended version: good stuff, helpful improvement. We don't want people banned or blocked for quoting from sites where people advertise. My only concern is the possibility of "joe jobs." If I don't like User:PainInTheButt I post an ad saying "My name is "User:PainInTheButt on Wikipedia and I am for sale." But in cases in which there is clear advertising, and we are sure it's not a joe job, we need this kind of clarification. I am pretty well satisfied by Doc James' comments in the discussion section, and do support this proposal. Coretheapple (talk) 22:04, 4 March 2015 (UTC)
  • Cautious support subject to careful analysis of any cases. The present system is broken. A user can be banned for noting a link made elsewhere exposing a material conflict of interest. This is not a good thing. We don't want to become a venue for witch-hunts, and we certainly don't want to enable or support the kind of obsessive stalking of Wikipedia opponents that has led to bans in several cases, but there is a definite need to protect the integrity of the project against conflicted editors. Guy (Help!) 17:09, 4 March 2015 (UTC)
Re the amended version: this is a clarification of what should be current practice. In reality people simply get banned for outing, and I am told this happens for a first offence in some cases, even to long-standing editors who are acting in good faith. That is unacceptable. I think it's important to include this clarification. Guy (Help!) 21:57, 4 March 2015 (UTC)
  • Support amended version. John Carter (talk) 22:33, 4 March 2015 (UTC)
  • Support the outing claims below are clearly not in keeping with practice - with or without this section being proposed being explicitly stated, it will continue to be practice as it always has (and true consensus) to say 'this subject of this article is being advertised at [link]'. Alanscottwalker (talk) 17:26, 6 March 2015 (UTC)
  • support as an individual. The relative importance of different policies is a matter for discussion, and when they come into conflict, the necessary course is to adjust one or the other either explicitly or by interpretation. I think our practices have gone much to far in the direction of protecting anonymity, to the extent that it is harming the encyclopedia. A limited degree of investigation of identities without our actually revealing them in public is already permitted in sockpuppettry, and this has proven very useful in addressing disruption. The current epidemic of undisclosed paid editing is as much a threat to the encyclopedia as the usual forms of puppetry, and we need as way of dealing with it. The TOU is explicit that the enforcement of the terms of use is jointly by the foundation and the community, and we therefore have the right to do so in any part that it is directly related to enWP, and not specifically prohibited by the Foundation. I think in particular we have a responsibility to enforce it in areas directly related to enWP where the WMF does not currently do so, even if we wish they would take the burden from us. As a minimum we have the obligation to do so in those cases where the WMF does not have the capability and we do. The existing manner of investigation of sockpuppettry is the appropriate manner here also, and our success there shows that we do have the capability to do it. There are obvious limitations to what we can effectively do; some of them have been treated in the discussion below,and others will be devised by guidelines and customs just as with other policy questions. Personally, I am not sure we do not already have the right to do as this proposes, but I think it necessary to be explicit. It is probable that some members of arbcom may oppose this proposal, but this is an instance where they need to be guided by the community. I regret very much that they never adopted this rule explicitly on their own, but I would not like to think that arbcom would actually refuse to implement a community policy. DGG ( talk ) 00:00, 7 March 2015 (UTC)
    Wait, DGG, I'm confused. You're supporting a proposal that is intended to allow us to link, onwiki, to off-wiki identities/accounts, but your rationale is that we should be investigating paid editing in the same manner as we do with sockpuppetry - which is to say, private checkuser investigation, public administrative action, but no onwiki publishing of personal/offwiki information about the suspected account. A sockpuppetry investigation will, as I understand it, pretty much never publicly connect an account name to an IP, exactly because an IP can be used to track down undisclosed personal information about the account holder in much the same way an off-wiki account (especially one related to real-life employment) can. So your two points seem mutually-exclusive to me - if you're supporting publicly posting the information, then you can't be doing it on the basis of the right system being one in which we're careful to not post private information...can you? Or am I misunderstanding something from your rationale?

    For the record, I agree with you that something along the lines of SPI would be a good starting point for handling paid advocacy editing - a small group of trusted users empowered to privately look into the "man behind the curtain" details of questionable accounts, and then act onwiki based on the results of their investigation. I just can't figure out how that reconciles with a proposal explicitly approving public posting of that non-public info. A fluffernutter is a sandwich! (talk) 01:32, 7 March 2015 (UTC)

Details will need to be developed, and there will be more several methoda needed. One is to permit the posting of useful links to published sources; another is to permit the private investigation of the relationships; a third, as mentioned in the discussion is the involvement of OTRSA, which is where people with COI conflicts often ask for proper assistance.
I wish this and the proposal below had not been done in this manner. The first steps that should have been taken is1/ a statement that to avoid all doubt, the WMF TOU apply at the enWP,2/ that accruing to the TOU, the community at the WMF has have a joint responsibility with the WMF to enforce them, 3/ that the Arbitration Committee has the power and the responsibility to investigate and enforce them within the limits of its abilities, with appropriate deletion to other group and individuals, all in accordance with guidelines to be developed. And then 4/ this particular proposal of permitting linking in certain cases being one of the methods. It would have been more helpful not to have skipped the intermediate steps, although , to be personally, the first 3 of them seem pretty obvious. DGG ( talk ) 04:44, 7 March 2015 (UTC)
Clarifying the other three can still be done. The paid editors likely are not going anywhere anytime soon :-) Doc James (talk · contribs · email) 05:09, 7 March 2015 (UTC)
  • Support This should be common sense, it clearly is a basic thing to allow people to point out someone is breaking the terms of conditions by essentially selling themselves as paid advocates:
"but this would publicly link a WP account to (potentially) a real name, employment history, photo and area of residence." Well, they've already published it themselves, this is merely notifying other wikipedia editors that someone is breaking the terms of use.
"it would violate every WP:OUTING criteria" So what? We are talking about someone doing undisclosed paid advocacy. They should be exposed.
"Likewise, this is already covered in "case-by-case." That's what people say until someone gets sitebanned for proposing it. It's impossible to bring up a suggestion of a COI on wikipedia because even indicating that there is a clear identifiable COI is itself seen as a violation of OUTING (yes, wikipedia editors are nuts).
" It's against the policy and ethos of Wikipedia, and is very unpleasant. " Paid advocacy is breaking policy and ethos as well as the terms and conditions of wikipedia. What's worse, undisclosed paid advocacy or those who expose paid advocacy by doing the very basic task of looking at websites where they sell their services and making a pretty uncontroversial link between advertised paid advocacy and an account.
" It's part of the bedrock of the project's morality that the pursuit and publication of real-world identities is not allowed" Morality? Wow ... "' As written, the proposed wording would then allow me to publicize that Elance account here on Wikipedia and so do the outing" You are inventing some bizarre set of circumstance. It's much more likely they'd just set up a fake account in the persons own name or some such.
"If the evidence is so compelling, emailing a single admin should be sufficient. If it's not that compelling it definitely should NOT be posted on-wiki. " Except that bans on private evidence don't lead to blocks, and emailing OUTING information to admins can lead to blocks as far as I know.
". Some of our most damaging biased editing is politically or ethnically motivated rather than financially." Most of the damaging editing is from the fact that most editors are as dumb as two sticks. But, on the COI, how does anyone really know the scale of financial issues when it's absolutely impossible to discuss it on wikipedia. The admins and arbcom actively collude in encouraging conflicts of interest to persist wherever possible. I know people with blatantly clear conflicts of interest, ArbCom knows, everyone who has seen the evidence knows. But ArbCom is quite happy to allow editors with COI do their damage willy nilly. Someone with a COI has unintended biases even if they do try and not exploit their COI. It's fantasy talk to think a paid advocate or someone with a COI is going to play nice in the topic where they have a COI.
People who suggest problematic edits be pointed out instead of a COI are basically putting the burden on people to do lots of work. An editor has to do an ungodly amount of research on a topic to see what the actual state of research is and then see how the advocates edits don't match that. They have to be clear issues because most wikipedia editors don't do subtly. You also have to convince people to care. The fact is, ad hominem arguments are not always fallacious, and sometimes a persons motive is a rational thing to take into account as in the clear case where they have a financial conflict of interest.
"The problem is with the article, not the person doing the editing." Someone with a COI can happily slant articles all day, because if it reaches ArbCom the people who judge a slant at the end of the day are just random members of the public with no special knowledge of the topic to judge what is slanted or not. I absolutely guarantee most arbs looking for a slant look for the kind of obvious tone issues, rather than the actual clever POV slanting. Not to mention the days and weeks wasted by others arguing with them. A good civil POV pusher can waste years.
"Anonymity is also important so that outside individuals may not unduly influence editors who do not engage in paid editing in how they edit." This is a rather bizarre defence of anonymity. Do you think real encyclopaedias worry about showing the name of their editors and writers in case they are influenced? No, clearly not, because that would be insane. People are anonymous on wikipedia because everyone else is, and there's a good chance that some of the editors a person has to interact with are either nuts, will file frivolous lawsuits against you or will harass you. Let's not pretend that people are anonymous for any lofty goals here. Second Quantization (talk) 22:22, 9 March 2015 (UTC)
  • Support Absolutely. We need to formalize the COI process to either be more like SPI or roll it into SPI. We need to send a message, if you want to do paid editing here, you have to disclose it and play by the rules. If you don't, we will find out, and undo all your work. Right now our hands are far too tied to put any pressure on paid advocates to disclose, and very few have disclosed. Gigs (talk) 18:21, 13 March 2015 (UTC)
  • Support per Doc James. Gamaliel (talk) 21:23, 17 March 2015 (UTC)
  • Limited support. The opposers give plenty of reasons not to make this change, and the COI situation I believe generally to be overblown. What matters is the final result: good stuff by a COI editor is no different from good stuff by an uninvolved editor, and the same is true of bad stuff. However, I see no reason to object to one thing mentioned above. There's no way we should object to something like "Someone at http://www.example.com/JohnDoe says that he provides 'Wikipedia Editing Solutions' and gives a link to User:John Doe". As Coretheapple notes, we still have to be careful with this kind of thing, lest people think they can start framing their opponents. Nyttend (talk) 13:28, 19 March 2015 (UTC)
  • Support. It is obviously important to protect the identities of wikipedia editors, but we have now reached the point where our chosen wording makes it harder to tackle serious editing problems on-wiki because it's forbidden to point out off-wiki evidence. Paid spammers are a big problem, but there are canvassing and pov-pushing problems too. We must try to fix those problems. bobrayner (talk) 04:34, 20 March 2015 (UTC)

Oppose

  • Oppose We've always argued that WP:OUTING trumps WP:COI, and I'd rather a different solution than sacrificing this. The problem is the amount of information typically provided on an Elance or other account. If it was simply a matter of linking a username to a username and jobs list I'd be ok with it, but this would publicly link a WP account to (potentially) a real name, employment history, photo and area of residence. - Bilby (talk) 06:49, 2 March 2015 (UTC)
  • Oppose: I was about to !vote support, but reading Bilby's comment changed my mind. If an external profile lists things like employment history, photograph and other real world details, it would violate every WP:OUTING criteria. We'll need to find another way. After all we're handling sock puppets (a greater evil) in other ways as well and even that does not take precedence over outing. --lTopGunl (talk) 18:53, 2 March 2015 (UTC)
Many of these users are using armies of sock puppets. And we are not handling them thus this proposal Doc James (talk · contribs · email) 19:32, 2 March 2015 (UTC)
Check user solves that problem as far as you are able to find basis for reporting at SPI (if not, linking an identity to one of the suspected accounts won't change that). The issue here is outing a person which was not even considered for things like plain socking, and should not take precedence over paid editing either... paid editors eventually get detected anyway due to their obvious promotional editing. It shouldn't take much time building a consensus among experienced editors with that. I guess, WP:COIN is what needs a reform. --lTopGunl (talk) 19:39, 2 March 2015 (UTC)
Often on Elance one can figure out that account X has taken jobs to write Y and Z. One can than discover that both Y and Z are written by brand new but different accounts. This is good support that these two new accounts are likely sock puppets. Doc James (talk · contribs · email) 20:08, 2 March 2015 (UTC)
You are right about that one... I think emailing a private list such as the ArbCom would avoid outing and still get it handled. It's something already being done about socks when outing is involved. --lTopGunl (talk) 20:44, 2 March 2015 (UTC)
Yes so arbcom does not feel they have the mandate to deal with undisclosed paid advocacy editing. The WMF does not feel they have the personal to deal with undisclosed paid advocacy editing and wish the community to take this on. Doc James (talk · contribs · email) 22:18, 2 March 2015 (UTC)
The WMF didn't have a problem with unilaterally changing the TOU. Now they want us as a community to enforce their unilateral policy? Yeah, right. They've got more than $50 million in the bank, they're free to unilaterally hire a little bureaucratic detective force and to issue their unilaterally created SanFranBans™ (from which there is no appeals process) against alleged perpetrators. This entire RFC process is a waste of time, going nowhere. Carrite (talk) 19:33, 6 March 2015 (UTC)
What arb com thinks on this is uncertain, because different members of arb com at this point have expressed in public different views on this. Speaking boldly for myself, I cannot imagine that the committee would not welcome an explicit decision of the community. DGG ( talk ) 01:41, 5 March 2015 (UTC)
Exactly. Arbcom are not arbiters of their own remit; they are given a mandate by the community. There is no reason the community can't add to that remit; all arbcom can do about it is whinge that they are overworked (and they might have a point). GoldenRing (talk) 01:10, 9 March 2015 (UTC)
  • Oppose but for a different reason - this is too much specificity. We do not write in WP:NPA "you may not call other editors 'halfwit space pirates from the planet zorp.'" Likewise, this is already covered in "case-by-case." Hipocrite (talk) 14:13, 4 March 2015 (UTC)
  • This is a request to change policy into allowing real-life sleuthing of other editors. It's against the policy and ethos of Wikipedia, and is very unpleasant. Editors have been banned from Wikipedia for doing such things. It is fair to advise users that their contributions may not be complying with our policies and guidelines, but it is not appropriate to seek out the real life person behind the account, and to publish revealing details about them, including where they work - regardless of what that place of work is. SilkTork ✔Tea time 15:35, 4 March 2015 (UTC)
Exactly, as much as we love to work here as volunteers on this project, after all, Wikipedia is just a website (something like we tell the ones we block and tell them to take it easy and go elsewhere). Outing a real life person on the other hand is something more serious and can have unknown repercussions. --lTopGunl (talk) 15:42, 4 March 2015 (UTC)
Holy cow you are both missing the point, and so much! The goal here is not to find out anything about anybody's RL identity. It is only the identity they create on Elance and similar sites, and only to to the extent that they link that identity to WP. How strongly they link that identity to their real life identity is none of our concern. None at all. do you see what i mean? Jytdog (talk) 02:04, 5 March 2015 (UTC)
Wikipedia is not about making links to the real people doing the editing. If there's a problem with an article, we have policies and guidelines to help us resolve that. If there's a problem with an editor, we have policies and guidelines to help us resolve that. Pointing a finger at Editor X and saying, "look there's a user with the same name on a website that pays people to edit on Wikipedia" is not dealing with the issue on Wikipedia, it's about chasing the person behind the account. If Editor X is making inappropriate edits, they need to be told. If they continue to make inappropriate edit, they need to be blocked. Being paid to edit is not in itself a problem. It is inappropriate editing that is the problem - and that can be done by paid or unpaid editors (and most of it is by unpaid editors, who are often enthusiastically unaware that what they are doing is wrong). Devising techniques and tactics to deal with inappropriate edits (such as this method to counter what WikiLinkPro is claiming they do), is more cost effective and less risky than sleuthing the user behind the account anyway, because while we can block an account, we can't prevent the user from carrying on editing Wikipedia. SilkTork ✔Tea time 03:02, 5 March 2015 (UTC)
  • Strong Oppose. It's part of the bedrock of the project's morality that the pursuit and publication of real-world identities is not allowed, and deliberate outing is anathema to that principle. It's hard to find undeclared paid editors, sure, but it's more important to protect people's real lives and safety, and we should not sacrifice the latter for our convenience. Identifying information for paid-editing investigations could be sent privately to ArbCom, for example, and that should be good enough. Squinge (talk) 18:28, 4 March 2015 (UTC) (ArbCom doesn't want to know, apparently. Squinge (talk) 18:02, 6 March 2015 (UTC))
    I still oppose the amended version. Suppose I know the real life details of a Wikipedia editor and I want to cause trouble for them and out them on Wikipedia. All I'd need to do is create a fake profile at Elance (or wherever) and name their Wikipedia account along with listing those personal details. As written, the proposed wording would then allow me to publicize that Elance account here on Wikipedia and so do the outing. Links to external accounts should only be allowed if the link has been made here on Wikipedia and not only on the external site, so we know the external account really does belong to the named Wikipedia editor and is not a fake. Squinge (talk) 05:01, 5 March 2015 (UTC)
  • Oppose - We don't need editors opening their Junior Detective bags and start publicizing real world information about other editors. If they are wrong, it damages the RW person and if they are right it provides free publicity. Suspected paid editors should be reported to SPI or other appropriate and discrete places so it can be hashed out. A quiet ban as a paid editor and subsequent quiet blocks for socking is much preferred over an advert. As an example, publicly outing User:MyWikiBiz as a real world someone that will write articles for pay doesn't help the encyclopedia but it does point customers to his real world name and website. Like most things, the less drama the better and we should avoid advertising activity we don't want. --DHeyward (talk) 22:15, 4 March 2015 (UTC)
@DHeyward: I'm all for denying paid editors recognition, and agree with your "free advertising" point. However, as a practical matter how does one raise the issue of an editor advertising his services somewhere without going to a public noticeboard? One can write privately to somebody-or-other on the arbitration committee perhaps, but that would not give the accused paid editor an opportunity to defend himself, which is important in the case of a "joe job." Coretheapple (talk) 22:58, 4 March 2015 (UTC)
You can't unring the bell. An on-wiki outing of a person that isn't violating policy is not acceptable. Their defense should be private as well as the accusation. I understand the need/desire to remove paid editors but when I fill in the Prisoner's dilemma chart, the best choice is always private. --DHeyward (talk) 10:32, 5 March 2015 (UTC)
But@DHeyward:, that's what I'm asking, how does one raise it privately? Who do you write? The subject has come up in the past and I would really like to know; this is not a rhetorical question. Coretheapple (talk) 15:24, 5 March 2015 (UTC)
ArbCom has mailing list, admins have email enabled, etc etc. If it's compelling, that would be enough. If it's not compelling then it shouldn't be posted anyway. --DHeyward (talk) 20:33, 5 March 2015 (UTC)
If they are being paid to edit then they are, de facto, violating policy. It's a breach of the terms of use. Guy (Help!) 15:20, 5 March 2015 (UTC)
If the evidence is so compelling, emailing a single admin should be sufficient. If it's not that compelling it definitely should NOT be posted on-wiki. --DHeyward (talk) 20:33, 5 March 2015 (UTC)
  • Oppose I believe the ban on outing should be a blanket one. Anything else invites wikilawyering around the edges. If the paid editing is damaging the encyclopaedia, then that should be clear from the edits without having to link the accounts on-wiki. There's nothing stopping you from looking at elance, finding jobs that have been taken, finding the relevant edits and reporting problems with them; allowing linking the accounts on-wiki is unnecessary for this purpose and invites problematic actions in other areas. GoldenRing (talk) 23:16, 4 March 2015 (UTC)
GoldenRing - 2 questions. First, in the world you are painting, on what basis do we report problems if we cannot cite the source? 2nd, what do you make of the fact that this policy already allows linking off-wiki on a case-by-case basis? Thanks! Jytdog (talk) 02:02, 5 March 2015 (UTC)
Jytdog, I think you are looking at this from the wrong end of the telescope. The problem is with the article, not the person doing the editing. Your comment suggests you are seeing paid editors as the problem rather than biased or otherwise problematic editing. Some of our most damaging biased editing is politically or ethnically motivated rather than financially. And while we should be alert to what organisations or groups plan to do on Wikipedia, we should not ever be in the business of asking users "Are you now or have you ever been a member of the Communist Party Elance?" SilkTork ✔Tea time 03:15, 5 March 2015 (UTC)
Yes, I think SilkTork puts it well. I'm not worried if an account is on elance; I'm only worried if their edits are bad. If their edits are bad, then they're bad and linking to elance doesn't change that. If they're good edits, which improve the encyclopaedia, then yay, someone is paying to improve the encyclopaedia. GoldenRing (talk) 04:53, 5 March 2015 (UTC)
And I don't think the policy should allow linking off-wiki on a case-by-case basis (I !voted this way in the recent RfC on this, IIRC). It makes it too easy for someone to phrase outing in a way that looks at least plausibly related to one of the exceptions. Then that editor's supporters pile on the ANI report and nothing happens. GoldenRing (talk) 04:56, 5 March 2015 (UTC)
SilkTork and GoldenRing there is no fucking McCarthy commission here - i have been subject to that - and i am being subjected to that even right now on Jimbo's Talk page by POV-pushing editors accusing me of COI. You are preaching to the fucking choir. So please get off your high horses about that and get down here in the dirt and look at the problem; about the "telescope", I would say that your "telescopes" are not directed to the actual issue here at all. Paid editing is a well documented problem on WP, and we have already shown, with concrete diffs in this discussion, how single editors working through elance take jobs and use throwaway sock accounts to create articles, which are generally shit.
We are talking about how to strategically turn off the shit spigot, in a way that is consistent with an exception that is already part of policy: by
  • taking away the protection of OUTING for those who advertise their undisclosed paid editing services in violation of the ToU; via email some of those editors have mockingly cited their protection via OUTING to me.
  • for those who go ahead anyway, stopping their activity by connecting concrete edits in WP with concrete claims in elance
This is not about witch-hunting. I have never seen either of you dealing with COI issues and to be frank I don't think you even are trying to understand the problem we are trying to address. Which is not what consensus is about. We have tried to show that we are very concerned with OUTING and have zero desire to harm anybody, but rather, to connect concrete actions here and there, which are done in blatant violation of the Terms of Use. All this amendment seeks is an example of the existing case-by-case exemption, so we can proceed with this important work safely.
I want to emphasize that paid editing has a place in WP, as do paid editors I am not opposed to paid editing at all and I work with paid editors who properly disclose and who ask for content to be added to articles through edit requests. Which is something else that I have never seen either of you do. Many of their suggestions (definitely not all) are great improvements. Jytdog (talk) 13:47, 6 March 2015 (UTC)
Well, yes, that stream of abusive ad-hom has certainly convinced me... Sorry, I just don't see making exceptions to OUTING as the right way to deal with this. I'm not saying there's not a problem; I'm saying that your 'solution' will cause other, bigger problems. To give a semi-hypothetical example: Suppose that Ryulong hadn't made the link between his wiki account and various off-wiki accounts public. About six months ago, someone with an off-wiki account that looked a lot like Ryulong accepted some money from people on one side of the GamerGate controversy; Ryulong was also very heavily involved in editing the article about that controversy. Can you please explain how your proposed change would not have enabled a large, co-ordinated harassment campaign, already long-established off-wiki, to also spread on-wiki in this situation? I'm no fan of Ryulong, don't get me wrong, but that's not to say that I want to enable that sort of harassment. GoldenRing (talk) 02:46, 9 March 2015 (UTC)
  • Strong Oppose Per Squinge and DHeyward. This proposal and its tenets are extremely problematic and asking for all kinds trouble. -- WV ● ✉ ✓ 05:31, 5 March 2015 (UTC)
  • Oppose, while there is an issue with paid editing, which falls under WP:COI. Anonymity is also important so that outside individuals may not unduly influence editors who do not engage in paid editing in how they edit. IF there is a false COIN investigation, that proves false, than that person's RL information has become a matter of record here on Wikipedia. While I understand that the proposer means well with this suggestion, IMHO WP:OUTING is more important not to violate. If there is the belief that someone is violating the alternate accounts policies that is what WP:SPI is for. If we need to empower certain bureaucrats to be able to do higher level investigations (with necessary checks on their abilities, so that they don't abuse their powers), than propose that maybe.--RightCowLeftCoast (talk) 05:32, 6 March 2015 (UTC)
RightCowLeftCoast the point of this amendment, adding an example to the existing "case by case" exemption, is exactly to facilitate the work at SPI and ANI with these undisclosed paid editors. That is exactly the point. Jytdog (talk) 14:39, 6 March 2015 (UTC)
  • Oppose per WP:CREEP and WP:BEANS. Andrew D. (talk) 00:00, 7 March 2015 (UTC)
  • Oppose per Wikimedia:Privacy_policy. NE Ent 02:51, 7 March 2015 (UTC)
  • Oppose It is absolutely never acceptable to disclose another editor's address, phone number, name, SSN, or the like when they have not actively put out such information on Wikipedia. WP:OUTING should absolutely trump WP:COI, and that's the way it's should be. This also seems to stem from a misunderstanding of what qualifies as outing. Saying that there is X username on Y site, which X username corresponds with a user on Wikipedia (in all exactness), and asking the user whether that is them is not outing. Outing implies personal information about the user. This is also just putting more bureaucratic nonsense into Wikipedia. If you can't discuss a COI with a user without outing them, you probably shouldn't even be discussing said COI with them. Name, address, personal information, etc. There is no excuse for it and there should never be an excuse for it. Tutelary (talk) 19:11, 7 March 2015 (UTC)
User:Tutelary We are trying to clarify that "Saying that there is X username on Y site, which X username corresponds with a user on Wikipedia (in all exactness), and asking the user whether that is them is not outing" is the case. Thus this RfC. This is not about disclosing the person address, phone number, SSN or the like. It is hard to tell if people are using real names on Elance but many are not. Doc James (talk · contribs · email) 20:02, 7 March 2015 (UTC)
  • Strong oppose. I am expressing only my own opinion here, not the opinion of the whole Arbitration Committee, however I am speaking as someone who would be dealing with the consequences of anything posted. In my view it is never acceptable to out anyone on Wikipedia, regardless of the motivation for doing so - and that explicitly includes the possibility of a person being a paid editor. While it is possible that no harm will come from suggesting that User:X here is User:X elsewhere, this is not always going to be the case. Even if the linking is correct, and especially if it is incorrect, there always exists the possibility of real world harm. Investigations should also not be undertaken - Wikipedia editors do not have the skills, resources or powers to conduct such investigations. I personally will not be doing any, even if asked as a member of the Arbitration Committee, without the prior and explicit written approval of the WMF to investigate the specific user and the prior and explicit written statement from the WMF legal department that they will cover 100% of the legal costs that I might incur from undertaking that investigation (for example if I am sued for defamation or libel). I do not have the resources to defend myself against any legal action, especially considering that I am subject to the jurisdiction of the courts of England and Wales, which are the world capital for libel tourism (and not without reason). As others have said more eloquently above, the problem is not with paid editors (disclosed or undisclosed) but with biased edits and biased articles, which may arise for any number of non-financial reasons. Thryduulf (talk) 17:25, 8 March 2015 (UTC)
    • Thanks for weighing in, Thryduulf. Are there example of WP users - as Wikipedia users with anonymous accounts - bringing suit in the real world related to connecting their WP identity to some other online identity? Thanks. I'm trying to understand how much of your concern is theoretical and how much is based on the experience of this institution. Thanks. Jytdog (talk) 18:57, 8 March 2015 (UTC)
      • I am not personally aware of any, although I have not researched the issue in depth. However, I am equally not aware that any such linking of an anonymous/pseudonymous account on Wikipedia to a real-world identity in a way that alleges them to be an undisclosed paid editor has happened many times. I do know though that real world legal action has been launched based on the content of pages on Wikipedia, and so it is not an irrational fear - particularly as the consequences of such a legal action happening would be very severe. If you have proof beyond reasonable doubt (i.e. sufficient for a United States court) that a user is violating the terms of use, and you have not violated the terms of use yourself (specifically including the privacy policy) to obtain this information, then contact the WMF as they have the resources and unambiguous mandate to deal with this issue. Thryduulf (talk) 23:10, 8 March 2015 (UTC)
Thryduulf (sorry for pinging, if you are watching) - thanks for your reply. I appreciate risk management very much; I do it as part of my job. I am generally pretty skeptical when folks bring up Important Legal Risks dressed up in Very Serious Fancy Language in the course of community discussions; it usually turns out to be well meaning but overdressed speculation. Let's ground this in advice from WMF Legal. I am pinging Stephen LaPorte who was involved in launching the amendment to the ToU and was helpful in past discussions of paid editing and COI. Hey Slaporte (WMF) - Can you take a moment to review the proposed amendment to the OUTING policy above, review Thryduulf's comments above, and comment on the legal risks Thryduulf mentions? Thanks! Jytdog (talk) 23:43, 8 March 2015 (UTC)
Jytdog I'm aware of two cases involving ArbCom (the WMF were involved with both); the risk of defamation/invasion of privacy suits are going exponentially increase with this proposed amendment. While we're on the subject, the WMF's Legal Fees Assistance Program explicitly excludes payment of "fines or damages (including any award of attorney's fees)". So, I also have a question for Slaporte (WMF). Will the WMF be paying fines, damages, and awards of costs in respect of purported enforcement of the TOU?  Roger Davies talk 07:10, 9 March 2015 (UTC)
Hi Jytdog and Roger Davies, I'm happy to weigh in with a few thoughts from a legal perspective, but first I have a disclaimer: I can't provide specific advice about your legal risk. Each user is responsible for ensuring that his or her activities on Wikipedia comply with the relevant law. Generally, this proposal is not illegal. But there could be legal risk whenever you post information about someone else online, depending on what you say (for example, if you say something false and harmful, depending on your local laws). The Legal Fees Assistance Program may cover defense support, for example if an admin acts lawfully and needs help hiring a lawyer, but it does not cover damages that result from a user's violation of the law. When thinking about the legal consequence of this proposal, I recommend assessing: 1) what is the risk of false positives or other impact on innocent users? 2) how can this policy minimize the amount of personal information that it allows? Best, Stephen LaPorte (WMF) (talk) 01:09, 10 March 2015 (UTC)
For more details, here is a link to the Legal Fees Assistance Program. Stephen LaPorte (WMF) (talk) 06:30, 10 March 2015 (UTC)
Hi Slaporte (WMF) thanks so much for taking the time to respond. Your disclaimer is giving me pause, and helping me to understand some of the objections to this being part of community-wide policy (I noted your use of "admin" as someone whom the WMF may be legal fees for; i also noted that it doesn't cover damages for anybody). If we are going to build some process by which links to Elance, it needs to actually be a project process, embedded in a trained bureaucracy that uses it with care, and not something that any random user has the right to do. And for the same reason, done in private. I am getting it, that a lot of the objections appear to be stemming from that. I am unclear on the arbitrator's objections, as they are the definition of both. I'll need to think about this more. Also, about the joe jobber objection that has been raised, which I don't have an answer for now, and I have not seen that anyone else does either. Jytdog (talk) 02:45, 10 March 2015 (UTC)
The joe job issue is what's mostly bothering me about this--it would be too easy to construct a joe job on someone and use that to OUT them with impunity. I don't like proposing an off-wiki process, but I think that may be exactly what is needed:
  1. Reports of undisclosed paid editors boasting about their sockpuppets and laughing at us because WP:OUTING protects them would be sent (privately) to the checkusers list, where they are investigated.
  2. A checkuser functionary then watches the off-site profile and corresponding sockpuppet activity on-wiki for a while, to let the undisclosed COI editors dig themselves nice deep holes.
  3. The sockpuppets (and sockmasters) are banned and a report passed to WMF for possible legal action should the abuse continue.
I'm not sure at what point in the above process the investigation should appear on-wiki. I would support mentioning nothing publicly until the sockmaster and sockfarm are publicly blocked and tagged. I'm still unsure about how to entirely protect against joe jobs--in the most extreme case, where the malefactor has planted malware on the victim's computer and is doing the paid edits through the victim's own computer (some bank-fraud malware uses this approach to steal money) there is no way that Wikipedia can distinguish between an actual paid editor and an extreme case of joe jobbing. In such a case, I think we would have to rely on "you are responsible for the security of your account" and WP:CIR. More generally, I think that, over time, checkuser evidence should be able to distinguish any lesser case of joe jobbing from an actual paid editor. A further protection is that the evidence itself would never be posted on-wiki. Only "this sockmaster and these socks are banned based on checkuser evidence after an investigation begun from private evidence" would appear publicly, so the undisclosed paid editor never gets OUTed on-wiki. Pathore (talk) 03:26, 10 March 2015 (UTC)
  • Here's what confuses me about this line of argument ("As an arbitrator I am not willing to do this because I might get sued"): there is nothing particularly unique about sanctioning someone for paid editing that doesn't also happen regularly while sanctioning people for any number of other things. "Defamation" can be claimed by any real-name account (or probably any pseudonymous one, too) if Arbcom sanctions them: "I am an upstanding citizen and this Arbitration Committee is putting it about that I am a [sockmaster|troll|jerk|in need of being told to be civil]! When people google me, those accusations come up now!" Real-world harm? Well, contacting an ISP, employer, or parent about a long-term abuser could do that. So could banning someone struggling with mental issues, and Arbcom doesn't shy away from that. Invasion of privacy? Well, any checkuser that's run can be argued to be that; a checkuser involves Arbcom getting their hands on roughly the same amount of private information as "X works for company Y" constitutes. And yet Arbcom - and the community generally - handle all these things. Why is paid editing, specifically, the one thing that won't be touched, especially when it's one of the few hard-and-fast rules that all users are considered to have actively agreed to by using Wikipedia? And especially when the people claiming to be utterly terrified of enforcing it are people who willingly volunteered for a job they knew would involve private information, hard decisions, and people disliking what they do? A fluffernutter is a sandwich! (talk) 13:45, 9 March 2015 (UTC)
    Points of information: (i) The committee has received legal threats over sockpuppet allegations. (ii) Sockpuppetry allegations aren't obviously defamatory. (iii) Accusing people of taking bribes or engaging in fraud is a serious allegation wit real life implications. (iv) Digging into someone's real life affairs carries a real risk of invading their privacy.  Roger Davies talk 11:29, 11 March 2015 (UTC)
    Query: Is a post on the order of "that would be like asking you to prove you aren't paid by the Koch brothers" beyond the pale - or is it "normal discourse" when posted to an editor? When asked to retract the apparent charge, the editor stated "In this case, the default position is that you are not paid by the Koch brothers, nor could one reasonably assume that you are based on an absence of evidence" which I found quite interesting, indeed. Cheers. Collect (talk) 11:36, 11 March 2015 (UTC)
    I note the legal issue is some arbitrators principal objection - just noting it's not mine. My concerns relate to the risks of false positives, the risks of real-life and on-wiki harassment via legitimization of amateur outing attempts, and the reality that this change is entirely unnecessary as we already have functional rules on COI, NPOV, UNDUE and RS that address the content of people's actual edits. If someone's edits are entirely accurate, neutral, well-sourced and appropriately weighted, why then would we support a poorly-equipped private investigation to uncover our best guess of their real-world identities and employers? If the edits are not any of these things, and consistently breach WP policies over time, then we can (and do) deal with this under current rules. -- Euryalus (talk) 07:22, 10 March 2015 (UTC)
  • Oppose for the reasons outlined by SilkTork above. -- Euryalus (talk) 06:16, 9 March 2015 (UTC)
  • Oppose as pretty much irrelevant. ArbCom recently closed a case where there were plenty of accusations of paid editing. The case was ultimately decided not on proving the truth of that claim, but by the editors' edits themselves. There's a line from the Bible, "Wherefore by their fruits ye shall know them". We don't need the amateur sleuthing, the inquiries, etc. If the edits can't follow NPOV, are advertising/puffery, etc, warn them, see if it continues, and if it does, block them. If they come back, send it to SPI. Absolutely no need to dig into who they are elsewhere on the internet. On the other hand, if the edits either are NPOV, useful, and compliant with policy (or are made that way after some combination of warnings and discussion), why do we need to (and why would we even start to) dig into their motivations? Judging people on their edits avoids all sort of issues, including when it is wrong. Look at GorillaWarfare's user page. Joe Jobbing is going to happen, because appearing to have a solid body of work to point to make potential clients of the imposter think the work is going to be good and put in by a contributor in good standing. I don't think we need this, not because paid editing is something I consider good, but because the solution that doesn't involve looking off-wiki is much simpler. Courcelles (talk) 08:39, 9 March 2015 (UTC)
  • I am not willing to be responsible for enforcing this. And I am not willing to give people the go-ahead to perform their own paid editing stings—as Courcelles just pointed out, I've had someone (or several people) impersonating me on Elance, Facebook, and who knows where else, and they have successfully had companies pay them to edit pages (see my Twitter post and my userpage for examples, and I've had emails from others). Without a checkuser (or even with, depending on the imposter's ability to use common UAs and fake a Boston IP address), this would easily appear to be me. As such, I feel like I am somewhat uniquely sensitive to the ease with which someone could joe-job a perfectly above-board editor, and am extremely uncomfortable with proposals such as this. GorillaWarfare (talk) 08:48, 9 March 2015 (UTC)
  • Oppose As GorillaWarfare and other of my colleagues have said, I won't do this. I agree that this will encourage joe-jobs (I've had fake Facebook and G+ pages created), and I don't see why we need this in any case. We have the tools to deal with editors who are biased or dishonest. A paid editor who follows our policies faithfully doesn't seem to me to be a problem. A paid editor who doesn't can be dealt with as we deal with any editor. Accusing people of being paid editors who are not is likely to chase away editors.
  • And finally, I know for a fact that the WMF will not reimburse me for any damages incurred in a law suit. Does the community really expect us to put our savings, our houses, our families at risk? Dougweller (talk) 10:45, 9 March 2015 (UTC)
  • Oppose per Bilby and Hipocrite. Kaldari (talk) 19:48, 9 March 2015 (UTC)
  • Opposeper Silk Tork, Dougweller, Goriila Warfare, ad others. The Wikimedia foundation won't handle this and the arbs can't handle this so let the community police this. Frightening given our penchant for pile ons. The issues must always come back to the articles. Look at the edits; if biased deal with the edits. Looking off Wikipedia for information on editors is a slippery, slippery slope that will, I am very sure, end up in a mud slide and a big mess at the bottom of the slope.(Littleolive oil (talk) 00:09, 10 March 2015 (UTC))
  • Oppose What the nom I think meant to say is: "WMF doesn't mind outing if it serves their purposes." That's a horrible message and a stupid policy. By the way, I'm a former and hopeful future paid editor. Chris Troutman (talk) 14:33, 17 March 2015 (UTC)
  • Oppose While my query supra was not answered, it is clear that my position is in opposition to the "amateur detecting" which would ensue if this passes. Collect (talk) 21:34, 17 March 2015 (UTC)
  • Oppose. At first reading I was going to vote support because Doc James presented it well, but then I read SilkTork and GorillaWarfare comments. GorillaWarefare makes an excellent point about being misidentified. And SilkTork is very good at reducing a problem to its essential truth. He's right. People have been banned from WP for this. Roger Davies makes a good point about the legal aspects. I'm not a lawyer but any policy change that could potentially bring litigation to the project should be avoided. SW3 5DL (talk) 01:51, 20 March 2015 (UTC)
  • Oppose. When the terms of use were changed, amendments like this were the main reason for my opposition to the change. As SilkTork says, we ban people for this kind of activity, and well we should. Amateur sleuthing is harmful to the reputation of individual editors and the project as a whole. As can be seen with Gorilla Warfare's experience, it is incredibly easy to fake; I myself have had to have pages taken down on Facebook and LinkedIn that purport to be mine. The nature and quality of the edits reveals conflict of interest far, far better than any sleuthing, and it also will reveal other biases that have nothing to do with money. As a checkuser, I would not conduct an investigation on any account absent on-wiki evidence that it is violating editing policies. Risker (talk) 16:28, 22 March 2015 (UTC)
  • Oppose. Silk Tork echos my thoughts on this. — Ched :  ?  16:37, 22 March 2015 (UTC)
  • Oppose. Per Risker, Gorilla, etc. The various harms outweigh the good, and the potential for abuse and deleterious effects on peoples lives and livelihoods is significantly greater, in my opinion, were we to allow amateur sleuthing and condone this kind of outing. -- Avi (talk) 02:42, 26 March 2015 (UTC)
  • Oppose as unnecessary - I don't want to create an invitation to out someone who is a good faith, well-respected user and fixes a typo on an article about their employer. If there is someone who is obviously an employee of one of the article-for-hire businesses, then it's already fine right now / today to deal with that situation within reason. We should not be encouraging users to do any kind of detailed sleuthing or any such thing. --B (talk) 17:26, 10 April 2015 (UTC)
  • Oppose no one should be outed under any circumstances, regardless of any alleged COI. Just because someone is being paid to edit doesn't mean that people are not required to observe core behavioral policies when dealing with them. --Jakob (talk) aka Jakec 02:01, 29 April 2015 (UTC)

Discussion

  • Comment from a paid editor Adding a couple qualifiers like "compelling off-wiki evidence" and "to investigate breaches in Wikipedia's Terms of Use" would help make sure the policy only supports legitimate cases where evidence is needed to investigate covert activity and not for harassment. CorporateM (Talk) 15:21, 1 March 2015 (UTC)
  • Amended comment: Regarding the amended version, it seems too specific. Policies aren't intended to cover every circumstance individually in that way. The second RfC below seems like a better direction to me. CorporateM (Talk) 22:57, 4 March 2015 (UTC)
  • Commnet - Could OTRS be used for this disclosure along with a link to the OTRS ticket in the conflict of interest discussions? That would seem to resolve the concern regarding WP:OUTING while still allowing an open discussion at WP:COIN based on trusted OTRS users vouching for the conflict of the Wikipedia editor being discussed. Vertrag (talk) 15:31, 2 March 2015 (UTC)
    • interesting ... will ask them! Jytdog (talk) 20:07, 2 March 2015 (UTC)
        • OTRS has a many month backlog. They cannot effectively take on more work as they are not able to do their current work in a timely manner. Doc James (talk · contribs · email) 20:10, 2 March 2015 (UTC)
            • i just looked over there and they have a really serious obligation to keep things private. i don't reckon they would look kindly at bringing brought into an "enforcement" scenario which would muddy their "helping" mission.... and the backlog is a problem i didn't know about. i like the thinking creatively! we need to find a way to better deal with these broker companies. Jytdog (talk) 20:13, 2 March 2015 (UTC)
  • question to TopGun - would you please say more about what kind of reform you think is needed at COIN? Thanks. Jytdog (talk) 20:07, 2 March 2015 (UTC)
Nothing specific in mind, but the point was to hone in on the issue at hand... having more experienced editors watch it would certainly help. An RFC-like format for analyzing specific cases (perhaps calling in an RFC at COIN) would help too. --lTopGunl (talk) 20:53, 2 March 2015 (UTC)
Hi TopGun, thanks for your reply. I would love it if more admins hung out at COIN, but it seems that few are interested now. Atama was spending time there for a while and was a great presence, but has moved on to other things. I understand this... this remains a volunteer project and folks work on what interests them, when it interests them. But the point of this RfC is that our hands are really tied by OUTING. I don't know if you are aware of this, but freelancers on Elance and the like are so brazen as to link to their WP user pages and give examples there of articles they have created for pay. (This is one of those rare cases when "brazen" is actually apt.) They do this, knowing they are protected by OUTING. It would be good for the encyclopedia to take that protection away from them, for this narrow case. Do you see what I mean? Jytdog (talk) 13:58, 4 March 2015 (UTC)
Totally see what you mean... I dropped in because I watch this venue rather than COIN but I guess it's a good idea to watch that as well... from my occasional peaks into COIN I do see many COI / paid editors being dealt with as well. But if a wikipedia editor goes so far as to link to their wikipedia profile from somewhere such as an elance profile and offers paid services, I think those are grounds to a private ArbCom report already. If they have their hands full, maybe another private committee can be formed which can sort out the cases before finalizing them for ArbCom. My objection is due to the fact that outing a person's private details is something that was not even done for those who break every rule of wikipedia and sock on articles - they are also dealt with in private and I've had my hands tied a few times in that case as well. Maybe there can be another possibility if forming a committee is something too bureaucratic, find volunteer admins who would deal with such situations via email, list them under a category or something and get them to take a look at it so that there's a case of SPI or something. Just to add to it... say if some one searches up an editor, John Doe, registered with their actual name on wikipedia, makes a fake profile of them and puts in the same details that of their own or those collected from around the internet and claims that one of their article was "paid" for... they can game the system and out the person just by running them through the process (whether or not they can get them sanctioned). I just thought of this one, but I think there might be other flaws and possibilities of gaming the system with this too. Whereas, as of now, any outing is oversighted right away. --lTopGunl (talk) 15:31, 4 March 2015 (UTC)
  • Bilby - i hear your objection very clearly generally. and I think that Doc James sought this carve-out on purpose. Paid editing remains a problem in WP. It is great we have the Terms of Use but we need to enforce it. I would say that if WP editors are getting jobs on Elance etc and are coming here and making undisclosed paid edits, they put themselves at risk. No editor at elance who complies with the ToU and discloses paid edits, would get outed via the action of this exception. We need this exception to be able to make the ToU stick. Does that make sense? Also, I witnessed your block of johnmoore, and as far as i could tell, you more or less did what this exemption would make clear is OK. If you are going to stick to your guns on this, i would be very interested in hearing your thoughts on how the community can manage the other johnmoores among us. thx. Jytdog (talk) 22:28, 2 March 2015 (UTC)
The whole problem is ugly, and I'd like to see a solution too. But in the past we've taken the idea that we don't out editors as a kind-of golden rule, so if we had to publicly out an editor in order to prove that they had a COI, we couldn't prove the COI. We can communicate privately, though, and my approach in the past has been to get a case together to send to a checkuser or oversighter for independent verification, although normally evidence found on-wiki is easier to work with. (I haven't been overly pleased with working with ArbCom, and probably won't worry about contacting them in the future). In regard to Johnmoor, that was an unusual case as we had good on-wiki evidence, and if needed he'd previously made the connection through publishing his real name on WP with a link that connected the paid editing account to his WP one. - Bilby (talk) 14:16, 3 March 2015 (UTC)
Thanks, that was a thoughtful answer. I hear everything you said. The thing that I struggle with this is that the approach you discuss leaves us unable to work as a community to deal with paid editors... what we end up with instead, is editors who want to collaborate to work on the problem of undisclosed paid editing are forced to do that work off-wiki, in a non-transparent fashion... and ugly shit grows in the dark. I was the target of such an effort myself and when the idiots emailed me their "work" I was disgusted almost as much by sloppiness of their thinking and their conspiratorial glee in trying to tear someone else down, as by the effort itself. Much better to have the work be done in public and transparently. But we need the exception to OUTING to do that. The exception here is narrowly targeted to people who post on freelancer sites. Narrow. And as I mentioned, freelancers do their work in blatant violation of the ToU and they do it knowing they are protected by OUTING. I think we need to take away that protection, narrowly. You couldn't live with that? Jytdog (talk) 14:28, 3 March 2015 (UTC)

I have asked the foundation to create a group of functionaries whose position is to help with enforcement of the TOU. Doc James (talk · contribs · email) 22:43, 3 March 2015 (UTC)

I think that's a good step. There needs to be a way to balance "allowing clear-as-daylight evidence of paid editing" with "not posting people's personal info on Wikipedia", and "allowing Elance etc. as evidence but passed privately through functionaries" seems like a potential way to harmonize those goals. Like checkuser, sort of. –Roscelese (talk ⋅ contribs) 14:57, 4 March 2015 (UTC)
Doc James any progress on the front of having WMF create a group of functionaries? Thx Jytdog (talk) 23:18, 8 March 2015 (UTC)
Nothing yet. Doc James (talk · contribs · email) 02:29, 9 March 2015 (UTC)
  • Hipocrite about your oppose, and specifically the "covered by the 'case by case' exception" piece of it. In my view, OUTING is really serious policy in WP and we have, rightfully, a zero tolerance for outing. Without a clear exception I would never take the risk of getting banned for OUTING by bringing evidence from Elance. Please consider that. Jytdog (talk) 14:19, 4 March 2015 (UTC)
If you can write language that does not remove the case-by-case but rather adds on to it, then my objection is obviously ignorable. Hipocrite (talk) 14:20, 4 March 2015 (UTC)
Hm! Thank you! So something like "Posting links to other accounts on other websites is allowable on a case by case basis. For example, references to freelancing sites or other sites involved in soliciting paid editing, where editors disclose their Wikipedia usernames, take paid editing jobs, or provide examples of Wikipedia articles they have written for pay, or the like, may be used." Something like that? Jytdog (talk) 14:27, 4 March 2015 (UTC)
Fine w/ me. Hipocrite (talk) 14:32, 4 March 2015 (UTC)
Thanks! SilkTork, TopGun, don't know if you looked down here. Doc James, does this work for you? Hipocrite pointed out that there is already a "case by case" exemption to OUTING and we can include this the new language there, as an example. Please really do take a moment and consider this - I know you are busy and that this is a contentious issue, but the WMF made this part of our Terms of Use and we need means to enforce that, responsibly and respectfully. Jytdog (talk) 15:49, 4 March 2015 (UTC)
Yes thanks USer:Hipocrite this was not to replace the "case by case" but to add to it. Happy with Jytdogs wording. Doc James (talk · contribs · email) 18:32, 4 March 2015 (UTC)
worth amending the RfC and pinging those who have already !voted? I think the anchor in existing policy is very helpful and will help people be more comfortable with it Jytdog (talk) 18:55, 4 March 2015 (UTC)
Yes not sure if we need a new RfC for the new wording. Maybe we should ask if some of those opposing would agree to that. Doc James (talk · contribs · email) 20:10, 4 March 2015 (UTC)
ok amended, notices to AN, COIN, and talk:COI, and notice put on Talk page of each editor who has !voted or commented. Jytdog (talk) 21:57, 4 March 2015 (UTC)
  • Comment I support this proposal, but I do see one problem: the prospect of somebody trying to "get" an editor by posting a phony ad naming an innocent editor. Such "joe jobs" are possible even today, and might be hard to counter. Coretheapple (talk) 21:50, 4 March 2015 (UTC)
I totally hear that, Core. There are evil motherfuckers out there who could try to do that. Would take deeper investigation and probably cooperation of elance who ever to work something like that out.... hopefully in the real world this will happen infrequently if ever.. Jytdog (talk) 21:58, 4 March 2015 (UTC)
      • One needs a few things. 1) the content added needs to be promotional 2) the job posted at Elance needs to mention what Wikipedia article they are paying for 3) the person who took the job thus may be the person editing the Wikipedia article
      • Usually armies of sockpuppets are used for these edits. We want to be able to use this evidence at CU. Currently it is not allowed because it cannot be posted on Wikipedia. Doc James (talk · contribs · email) 22:02, 4 March 2015 (UTC)
  • Comment re some opposes. I'm seeing a bit of sentiment to the effect that we should have mercy on editors who advertise their Wikipedia work, and in so doing disclose personal information about themselves. Give-me-a-break. These people are voluntarily outing themselves in a public forum with the intent of violating our terms of use and denigrating Wikipedia for their own selfish commercial ends. Providing them with a cone of silence and protection because of their advertising efforts is downright perverse. Coretheapple (talk) 22:16, 4 March 2015 (UTC)
  • Comment The wording in the alternative doesn't seem like a significant change from the original, in that it reads as if there are three situations where linking to accounts is permitted - if they self-disclose their WP account by naming it; if they self-disclose their Wikipedia account by listing articles they have written; or if they take paid editing jobs. Am I reading this correctly? - Bilby (talk) 23:08, 4 March 2015 (UTC)
Those are the three specific examples offered within this larger example. The key thing is that Hipocrit commented above, that OUTING already allows use of external links in some cases. The RfC is now correctly framed to be an example of the kinds of "cases" where linking is permitted - as opposed to it being some radical new exception to OUTING. Context is important! Jytdog (talk) 23:19, 4 March 2015 (UTC)
In that case, the new wording doesn't fix the problems from the old, as it still allows linking to personal information where the editor hasn't made that connection on or off-wiki. The main change seems to be a broadening of the situations where outing is possible. - Bilby (talk) 00:32, 5 March 2015 (UTC)
Hi Bilby. Not sure I am following you, in two ways. First, the "case by case" exception was already there... right? Additionally, what we are looking to do, is to be able to link to somebody's profile or posting in elance etc, who says "i am user X in Wikipedia" or "i created article X in WP" or "i accepted the job to create article X in WP"... and there would only be action if there is undisclosed paid editing going on. Otherwise nobody cares. While I acknowledge that personal information might get revealed, in my view that is not our problem. It is like saying that prison is a dangerous place so we shouldn't send people there. People who do undisclosed paid editing would be putting themselves at risk of OUTING, and that is, as it should be, in my book. Jytdog (talk) 01:47, 5 March 2015 (UTC)
The "case by case" addition wasn't intended as a means of circumventing the rule not to post links to personal information. It was in regard to situations where, for example, someone has a Twitter account being used to canvas, with the same username as their WP account, referencing their on-WP actions. That's not the same thing as linking to someone's resume with a list of information that the policy specifically forbids revealing.
The situations where someone on Elance reveals who they are on WP are very rare - almost all Elance/Odesk/etc editors simply describe their background (x number of edits, editing for n years) without providing additional information. But be that as it may, the proposed wording isn't limited to that - it allows linking to an account based on a suspicion that they are engaged in paid editing, whether or not they reveal their WP account on the Elance/Odesk one. - Bilby (talk) 03:12, 5 March 2015 (UTC)
So as long as you include something that could be a photograph of you are a name that could be yours than you can use any webpage you want to run a Wikipedia editing business or engage in meat puppetry in your opinion? You can also use other accounts to create attack pages about Wikipedians and laugh about them on Wikipedia but as long as you do not link to them yourself you are safe. Doc James (talk · contribs · email) 03:17, 5 March 2015 (UTC)
I'm not sure if you were looking for an answer, but no. If someone posts personally identifiable information off wiki and doesn't link to it, then no, we can't link to it, but that isn't the same as saying that we can't address the problem. We just can't do it through outing. - Bilby (talk) 23:50, 6 March 2015 (UTC)
  • Comment Even with the amendments and new ideas, I think this proposal is misguided. Focus on edits, not editors. It's one of the core principles of how we work at wikipedia, but this sort of proposal encourages people to do exactly the opposite, focusing on who is making the edit, not the merits of the edit itself. GoldenRing (talk) 05:51, 5 March 2015 (UTC)
The major freelancer sites are willing to take down accounts when we ask them to. Doc James (talk · contribs · email) 05:55, 5 March 2015 (UTC)
  • Note - prompted by a thoughtful note on my Talk page, i have publicized this at WP:Village pump (policy) and WP:CENT.Jytdog (talk) 17:03, 6 March 2015 (UTC)
  • General comment regarding all of this: The fact that Arbcom does not believe the ToU are policy or that they are bound by such a policy is extremely frustrating for those of us trying to deal with paid editing. However, I don't believe the solution to their refusal is to throw away a protection that has been very important to both our users and the WMF for pretty much as long as Wikipedia has existed. It is not merely a hypothetical that publishing a user's details can lead to them being harassed irl. It has happened, repeatedly, and that's a very significant reason why we have reacted by not allowing personal details to be thrown around in public. If Arbcom won't privately investigate paid editing cases, I suppose we can't force them to do it (short of passing an arbcom recall procedure and then recalling the whole committee at once, which is about as likely to happen as me being made the next Queen of England), but we can create another committee to investigate it, or hand it over to the oversight team, or start talks with the WMF about them globally enforcing it. We could even empower individual admins to do one-person, private investigations-and-blockings, if we want to err on the side of false positives rather than catching nothing at all. All of these are better options than shrugging and saying, "Well, I guess it's open season on the personal details of anyone we think might be getting paid, even if we turn out to be wrong!" The answer to "we have no current mechanism to do this in private" shouldn't be "...so let's just throw away the idea of privacy!" A fluffernutter is a sandwich! (talk) 16:39, 6 March 2015 (UTC) Partially copied from now-closed RfC below A fluffernutter is a sandwich! (talk) 17:45, 6 March 2015 (UTC)
  • Comment - Of course is is a good idea and commonsensical, but it is in direct contradiction of site "outing" policies... What to do, what to do? Hey, wait, didn't opponents of the anti-paid editing crusade note time and time and time and time again that there is absolutely no way to "ban" paid editing without having restricted registration (to end socking) and an end to so-called "outing" rules so that bad actors can be identified and weeded out... Oh yeah, you anti-paid editing activists forgot that, didn't you? Well, the contradiction remains and there is no way to make the math of (apple + orange = banana) work out. Sorry about that. We can bluster and hurrah about the vanquishing of undeclared paid editing just so long as we are cognizant that it hasn't been vanquished at all, only driven deeper underground where the work is more difficult to scrutinize. Oh yeah, we warned you about that, too. Carrite (talk) 19:10, 6 March 2015 (UTC)
  • Comment - The unsolvable nature of this problem is why Wikipediocracy remains a valuable institution, by the way. Carrite (talk) 19:27, 6 March 2015 (UTC)
Carrite fwiw, I am not an anti-paid-editing-activist. Paid editing has a place in WP; paid editors just need to follow the ToU and the COI guideline, that's all. I work with paid editors who do so and many (not all) of their suggestions are valuable. HINT - we could use more editors who do work with disclosed paid editors; you can find edit requests awaiting help at WP:COIN. The point here is exactly to bring transparency. Jytdog (talk) 22:56, 6 March 2015 (UTC)
How can something remain what it's never been? NE Ent 02:53, 7 March 2015 (UTC)
  • Comment I have mixed feelings about this. I've spent a lot of time on conflict-of-interest problems. I was heavily involved in the attempt to stop the massive paid editing push at Banc de Binary. There it was discovered off-wiki that Banc de Binary was offering $10,000 for someone who could get their article rewritten to omit their problems with the CFTC, SEC, and the Federal courts. See WP:Administrators'_noticeboard/IncidentArchive844#Banc de Binary, Round 2 for how that went. That received press coverage in the Wall Street Journal, and in the ToS change on paid editing. We need to be able to deal with episodes like that without being accused of "outing".
Most conflicts of interest are a nuisance problem. I deal with a lot of those at WP:COIN. Usually it's enough to tell people that, no, Wikipedia is not a place for promoting your band/DJ/company/product. That deals with at least 3/4 of conflict of interest problems. In practice, excessive promotion from a single source isn't a big problem. They get reverted by multiple editors, and they either stop, or they hit 3RR and get blocked. Identity doesn't matter. Trouble that might lead to "outing" generally comes from multiple editors seemingly acting in concert. That's when "sockpuppet" or "meatpuppet" accusations appear, and have to be dealt with somehow. Still, outing in the commercial area tends not to be a big problem. Editors often admit they work for a PR firm when they're close to being blocked.
It's on popular controversial topics where Wikipedia runs into problems. WP:OUTING is intended to protect editors writing on controversial or dangerous topics from retribution. Abortion and Israel-related topics, where there are organized advocacy groups on both sides, are especially difficult. I don't know what to do there. We had some problems at Landmark Worldwide (which is either an educational operation or a cult, depending on point of view) recently come up at WP:COIN. I passed the buck to AN/I. They passed the buck to ArbCom enforcement.
So, as a practical rule, I'd suggest that we protect the anonymity of editors strongly on anything which could broadly be considered public or political discourse. In the private sector, where paid promotion is more likely, we can reasonably ask whether editors have some connection with the article subject. John Nagle (talk) 09:44, 7 March 2015 (UTC)
  • Question The canonical example given for the need of this policy is an elance account which lists multiple wiki accounts it has used or is using to do paid, POV-pushing editing. In my view, the correct way to handle this is to open an SPI which includes the text, 'See private evidence on checkusers mailing list,' and then send the evidence to the checkusers mailing list. In what way is that process not sufficient to handle the situation? GoldenRing (talk) 03:07, 9 March 2015 (UTC)
  • Comment My only concern with this is the possibility of joe jobs being used to OUT someone with impunity. We have no good public way to prove that User:X here is actually the person who claims to be User:X and does undisclosed paid editing. Suppose someone wanted to "get User:X" and could dig up enough information to (accurately) out User:X. Currently, that someone will have a very hard time doing this on-wiki without getting (rightly) banned, but with this proposal, all they would need to do is make a profile with User:X's details on Elance, offer paid editing services, edit a few articles for pay, then make a COI complaint. Presto: User:X is now outed and it's all above-board.
I don't like creating such an obvious loophole, but having COI editors gleefully boasting that WP:OUTING protects them is just ... grating. I like GoldenRing's idea of requiring such statements go through SPI with links to the offending profiles sent to the checkuser list instead of being posted on-wiki. Joe jobs are still a possible concern, but checkuser evidence would make them far more difficult. Checkuser functionaries are already trusted with private information, so that aspect shouldn't be a problem. Pathore (talk) 21:32, 9 March 2015 (UTC)
I'd not also that WP:CHECKUSER already says this:

CheckUsers are seasoned, experienced users, trusted to handle sensitive and privacy related matters and other user issues. Routine sock-puppet and editing issues requiring CheckUser review are handled at Wikipedia:Sockpuppet investigations. In keeping with WP:NOTBUREAUCRACY, CheckUsers are authorised to receive contact by other means, like on their talk pages, by e-mail, on IRC or a mailing list, and so on. If an incident is of a private or sensitive nature, you should never use public means of contact.

Doesn't this already cover this situation without all the problems of outing someone publicly? GoldenRing (talk) 23:04, 9 March 2015 (UTC)
  • Regardless of anything else, the investigations proposed here, and the outing that would result from them are completely unnecessary:
    • If an editor is biased, they can (and should) be dealt with on-wiki regardless of who they are and why they are biased.
    • If an editor is being paid, but is not making biased or otherwise problematic edits then the encyclopaedia is harmed by the removal of a good contributor while gaining nothing.
    • If an editor is wrongly accused of being an undisclosed paid editor then the encyclopaedia is harmed by the loss of a good editor for no reason, and the person making the accusation potentially ends up in legal trouble.
    When something has many serious downsides and no benefits, why is it even being considered? Thryduulf (talk) 23:40, 9 March 2015 (UTC)
  • Comment As someone who's taken many accounts to COIN...Necessary but not sufficient. We can and should make an exception like this, which would be useful but problematic however worded/restricted/enforced without embracing wider change. We should do it soon but not now as it currently throws the baby out with the bathwater... We need to grasp the fundamental issue first - a clash of cultures - reconciling:
    • legal culture of the WMF TOU (no undisclosed paid editing)
    • right to anonymity of editors (exemplified by WP:OUTING)
Posing the question for the overlap: Do we want anonymous paid editing?
A simple (and useful) exception here won't affect those fundamental driving forces, and the wider change we need to put into place to protect the integrity of our content from this long-term corrosion. Widefox; talk 11:47, 10 March 2015 (UTC)
@Widefox: Speaking personally, I want editors who (a) can edit neutrally and (b) have the option to remain anonymous if they choose. Whether or not someone is paid to edit is entirely irrelevant to me as long as the editing policies are followed. Thryduulf (talk) 09:31, 11 March 2015 (UTC)
@Thryduulf: Sure, that's our culture. The status quo.
The problem with the status quo is that we haven't fully embraced that our culture is bounded by advertising law (and TOU). (In that, we're different to other open cultures like the open source/Linux kernel) We encourage and enforce anonymous editing, incentivising non-disclosure for paid editors, driving them underground, empowering lawbreaking. This risks disenfranchising disempowered volunteers in that overlap. Balancing the needs of volunteers and paid staff is effort (both for paid WMF and paid editors). It's easier for all parties to have a clear separation. No more setting-up undisclosed paid editors and volunteers for failure. Would removing the right to anonymity for paid editing work? Well, it would set clearer guidance for all parties than currently. Simpler. It removes the desire for vigilante OUTING. These fundamentals aren't going away. [1] Widefox; talk 14:04, 13 March 2015 (UTC)
I don't understand most of that I'm sorry. Removing the right to anonymity for paid editing is both what the ToU has done and meaningless in practice - if someone wants to remain anonymous they will just say they aren't being paid, regardless of whether they are or are not and our policies against outing will rightly protect that anonymity. Outing is against both the terms of use and local policies, without exception and regardless of motivation. Harassing someone, including in an attempt to get them to out themselves, is against both the terms of use and local policies without exception and regardless of motivation. The only way to end undisclosed paid editing it to end undisclosed editing and verify the supplied credentials of every editor - that is somewhere that I am not prepared to go. Ultimately I can do little more than repeat that paid editing, disclosed or undisclosed, is not actually a problem. Biased editing (note not biased editors) is a problem, but not all paid editors engage in biased editing, and not all biased editing is done by paid editors. Thryduulf (talk) 15:26, 13 March 2015 (UTC)
Wildefox's comparison to open source software isn't quite accurate, but it is apt. I can't just go check in a patch to the linux kernel that benefits my employer at the cost of everyone else, and put the burden on other developers to find and revert it. That would be insanity, but that's what we are accepting here if we accept Thryduulf's original argument. Gigs (talk) 18:28, 13 March 2015 (UTC)
  • The outing policy was always intended to prevent harassment of good-faith editors by those who disagreed with them. In my mind it doesn't even apply to single-purpose, biased, paid editor accounts. It's only through either some weird evolution in application, or, more cynically, advocacy by the "good hand" accounts of paid editors that it ever came to apply to things like assembling evidence of organized paid editing. I, and many others, will continue to post off-wiki evidence of paid editing, with some discretion applied of course, regardless of the outcome of this discussion.
  • Using a real-sounding name on a writer-for-hire site cannot become some kind of barrier to us sharing evidence that an account is engaging in wide-scale biased paid editing. That would be a ridiculous situation, and would effectively prevent us from addressing the problem. Gigs (talk) 18:09, 13 March 2015 (UTC)
@Gigs: There are mechanism differences (e.g. all kernel edits are equivalent to "edit requests"). The similarity is that both are bazaars (See Jimbo's mention in Bazaar#Legacy), open to all for editing (paid or unpaid), with accepted edits being based on the quality of the edit, not the editor. Coming back to the point:
Fundamentally, they both rely on Linus's Law to fix bad edits.. They both need eyeballs.
When there aren't enough eyeballs in open source we get long-term bad edits resulting in Heartbleed, in WP we get long-term fake/spam/NPOV/promo/adverts/legal issues.
@Thryduulf: Does the ToU remove anonymity for paid editors? I thought it's just COI disclosure.
The practical issue with relying on eyeballs (Linus's Law) to detect bad edits, is when there's not enough eyeballs, a high burden, and we actually disempower the eyeballs. Gigs is right - this is about burden. We have automated anti-vandal edits, but no automated sock/meat anti-COI/SPI (despite automated tweeting of COI IP edits externally).
Those are all details...what matters is culture - Gigs feels they must take things off-wiki, seems in aid WP:FIVE. How can we empower editors in aid of FIVE and/or disempower the undisclosed paid editors? I agree with DGG that we have an epidemic of law-breaking, but I see no neighbourhood watch, police or laws. Widefox; talk 11:23, 14 March 2015 (UTC)
I agree we need more eyeballs to find and fix non-neutral articles and to identify edits who are not editing neutrally. However, the proposals here will not actually help this (let alone solve it) and run the very significant risk of making it worse, and of creating equal or worse problems elsewhere. Even if you identified, correctly, every undisclosed paid editor (of whatever definition of "paid editing" you choose) and removed them from Wikipedia you will not have made a dent in the number of editors with a COI who are editing non-neutrally, will not have fixed any articles, and you will have removed people who were editing neutrally directly and those who were constructively engaging on talk pages, etc. Thryduulf (talk) 13:02, 14 March 2015 (UTC)
At the risk of a reverse-Godwin, I'd suggest that the only way a smaller group (the core of experienced editors) could possibly curtail the behavior of a larger group (those who would knowingly corrupt the encyclopedia for gain), is to have swift and decisive responses. The balance of burden must be shifted. Linus' law doesn't much apply when there are more eyes wanting to deliberately corrupt something than those looking to preserve its integrity. Nearly all our major changes in recent history have been about shifting burden to a greater or lesser extent. Pending changes, Cluebot NG, a few new CSDs, no creation without autoconfirm, AfC, etc. But those mostly catch obvious stuff. Biased COI editing is much more insidious, and yet we lack a solid policy to even deal with it. At some point it's going to probably have to come down to an even stronger WMF edict, because our own consensus system is even subverted by these same editors, who can blockade any policy proposals to deal with it. Gigs (talk) 16:22, 17 March 2015 (UTC)
You have completely missed the point though that the proposed policy would not actually solve the problem it intends to, not least because biased COI editing and (undisclosed) paid editing are not the same thing. Swift and decisive responses are good, but only when they are correct - this results of this proposal would not be swift, may or may not be decisive, and will be result in a lot of both false positives and false negatives. Thryduulf (talk) 12:12, 18 March 2015 (UTC)

Self-outing

If a paid editor lists their Wikipedia user name and articles they've created on an external website, we would not be violating outing by linking to it. The editor already outed themself. Banning an editor for pointing this out is ludicrous. Oiyarbepsy (talk) 16:53, 7 March 2015 (UTC)

I would hope so. We likely need a RfC to clarify. Doc James (talk · contribs · email) 17:38, 7 March 2015 (UTC)
What if I started an account on an external site and called myself Oiyarbepsy there - and said the Oiyarbepsy account here on Wikipedia was mine too? It's entirely possible for someone on an external site to lie about their Wikipedia user name, so it's only genuine self-outing if the external site account is confirmed here on Wikipedia. Squinge (talk) 18:55, 7 March 2015 (UTC)
Squinge, similar to the case with sock puppeting, we have to look at behavioral evidence. If Oiyarbepsy has in their for-hire site job history that they accepted a request to edit a certain article, and then Oiyarbepsy here edited the same article, that's a pretty strong sign. Every situation is different, but usually the evidence is clear when this is happening. Gigs (talk) 18:14, 13 March 2015 (UTC)
So all I need to do is add Oiyarbepsy's genuine editing history to my fake freelance site page? Pretending to be someone else really is very easy, as per the comments of people here who have had it done to them. Squinge (talk) 19:05, 13 March 2015 (UTC)
It would be obvious from Oiyarbepsy's history whether it was biased paid editing or not. It's a much easier case than sockpuppeting, because you can even fake the behavioral evidence in a sock case by following someone around and agreeing with them. There's no ironclad solution for this in an anonymous environment, but on the spectrum of difficulty, I'd say a joe-job for paid editing would be more difficult than a joe-job for socking, something we already formally investigate on-wiki, and I'd like to think, we usually get it right. Gigs (talk) 16:29, 17 March 2015 (UTC)
If it "would be obvious from Oiyarbepsy's history whether it was biased paid editing or not", then we wouldn't need the external evidence, would we? If it's clearly problematic editing on-wiki then we can treat it as clearly problematic editing and don't need external evidence of being paid - and if it's not clearly problematic editing on-wiki then we're back with not being able to tell if the off-wiki evidence is really Oiyarbepsy. Essentially what you seem to be saying is that we should be allowed to do the outing, but only when we don't need to. Squinge (talk) 16:59, 17 March 2015 (UTC)
Just by debating whether that is a case of self-outing or not while considering a specific user, that user would be outed with impunity without having accepted it from his actual (Wikipedia) identity. That's a no no. But if an SPI proves someone to be another account who maybe self-outed, that's a totally different story. --lTopGunl (talk) 19:05, 7 March 2015 (UTC)
Not sure I agree with that, but I do agree that potential joe jobs need to be considered. Remember that this originated as an effort to deal with websites that freely offer Wikipedia editing services, and which sometimes freely acknowledge the user accounts they utilize. Our outing policies are gamed by these people as a shield to protect their nefarious activities, and that has to end. Coretheapple (talk) 15:38, 14 March 2015 (UTC)
If you see an advert like that, then you just need to watch their edits for bias and notability problems. If there aren't any, then their actions have benefited the encyclopaedia and everything is grand. If there are problems, then we have tools to deal with biased editing already without any need to know who they are or why they are editing in the manner they are. Thryduulf (talk) 02:55, 15 March 2015 (UTC)
The question then becomes, can we share information about who is worth watching? Whether we can do it on-wiki in such cases is I think quite uncertain. But I consider spi a good analogy: we don't ask for certainty. We certainly do share information at SPI, based on both behavioral and checkuser evidence, and share opinions also. Sometimes we're wrong. If we waited for certainty before taking action, we'd almost never be able to do anything. I can't prove it, but I suspect joe jobs are about the same likelihood of being a true explanation as "my roommate did it". If it's the case, the person should easily enough be able to disprove it, using one or another of WP's confidential facilities if necessary. We don't let the possibility of impersonation keep us from accepting copyright permission either, tho the information we normally ask for at OTRS could be faked. It's enough that it's reasonable and very probable. If challenged, there are processes. We just have to say, there's enough of a probability that this is in violation of the TOU that the editor should be blocked. (as always here, my views are my personal views only--please do not assume that anyone else on arb com agrees with me.) DGG ( talk ) 02:21, 19 March 2015 (UTC)
(also a personal view only) Do you not see the big difference between comparing two sets of on-wiki contributions and comparing statements made off-wiki (of unknown and rarely knowable reliability) with contributions on Wikipedia? If someone joe-jobs you, how do you prove your innocence - and why should you be required to disclose any information about yourself to anyone? There is nothing in the WMF terms of use or privacy policy that require anyone who is not a paid editor to disclose anything. Indeed the only disclosure a paid editor is required to make are their "employer, client, and affiliation" - how can I prove (publicly or privately) that I was not paid to make this edit? (I wasn't). If a block is made and then challenged, what processes could be used to determine if it is correct or not? With things like sockpuppetry, determinations are made on the basis of comparing behaviour - the behaviour of user X is the same as the behaviour of user Y. What you are proposing is the behaviour of user X is biased, and there is a user with a similar (or different) username off-wiki that claims something (with unknown and unknowable truthfulness, reliability or verifiability) which means user X is an undisclosed paid editor. Do you see now how different and how ridiculous this is? And ultimately, after all this effort to determine nothing reliable, you have not actually solved any problem at all while making others worse (directly or by taking resource away from improving them). Thryduulf (talk) 03:26, 19 March 2015 (UTC)
Just as aside, I have seen a number of paid editors using descriptions stolen from elsewhere. It isn't common, but it has happened in the past, and is part of why I am unwilling to link to personal details off-wiki. (Although in most cases the descriptions were taken from other paid editors).
When connection accounts off-wiki to accounts on-wiki the only reliable means is comparing jobs that they were hired to do with editing on WP. If they just say "I did X" it is open to abuse and (I assume) joe-jobbing, whatever that is. :) However if you see a WP editing job advertised, paid editor X is hired, the article is edited per the job by WP editor Y, and paid editor X is then paid for the job off wiki, it is pretty reasonable to claim a connection. I still wouldn't link to their off-wiki account, as I'm opposed to outing, but it is often possible to to reliably tell that someone is engaged in undisclosed paid editing. - Bilby (talk) 06:15, 19 March 2015 (UTC)
  • Reminder The justification of prevented undeclared paid editing is that such accounts are overwhelmingly not likely to be ones who " follows our policies faithfully". We can approach this article by article, for the many 100,000s of promotional articles, or we can try to deal with the problem at the source as well. The proposers of this initiative are, as I understand it, not proposing the automatically blocking of anyone suspects of having an account on elance (etc.), but mealy that the data be considered relevant to determining whether the person is editing in violation of our Terms of Use. I can therefore not understand Risker's statement that she would refuse to even run a checkuser in such cases. The clear suspicion of violating the TOU ought to be considered part of the justification for running checkuser, just as observed behavioral similarities are a reason. They fall under both reasons 3, and 4. of the rationales for employing checkuser given at[[{WP:Checkuser]] "3. Disruption (or potential disruption) of any Wikimedia project; 4. Legitimate concerns about bad faith editing." Of course paid editors may claim to have done more than they have in fact done, or reason to pretend to be users in good standing, and checkuser is a good rough way of helping determine this, though it has obvious and non-obviouslimitations. DGG ( talk ) 07:12, 23 March 2015 (UTC)
    • I stand by my words. It is not possible to determine a TOU violation for paid editing via checkuser. For example, the recent uproar about the NYPD-IP edits to certain articles is unprovable as a TOU violation - all kinds of people besides NYPD employees have access to certain portions of the NYPD internet portals, including but not limited to lawyers and reporters, not to mention anyone with enough technical nous to figure out how to access the system. When I was in the local shopping mall, I had nearly-unfettered access to 23 (yes, 23!) different wifi internet connections, including some from nearby non-commercial businesses, only about a third of which had sufficient security on them that I couldn't access with a few keystrokes. DGG's comments are naive in their understanding of how the internet works and how checkuser works; it's not a knock against him, but a recognition that philosophy doesn't trump reality.

      Absent another problem in editing, such as obvious bias, there is no reasonable method by which to determine that any editor is likely violating the paid editor TOU amendment. I have said this ever since the TOU amendment was first proposed. I would not perform a checkuser on anyone based on some request that centers on "I think this guy's getting paid to edit". There's got to be a lot more than that for such an accusation to stick. And if the accuser is able to provide sufficient on-wiki information to show that an editor is likely to be violating the paid editing TOU amendment, then there's also enough to show that the accused is likely to be violating any number of other longstanding policies. Risker (talk) 13:50, 24 March 2015 (UTC)

Risker. Above you wrote: "I would not perform a checkuser on anyone based on some request that centers on "I think this guy's getting paid to edit". There's got to be a lot more than that for such an accusation to stick." I realize you went on to talk about other disruption, but I would like to pause you there, and ask: Absent a declaration of paid editing by a user, what on-wiki evidence would = "a lot more than that"? And just for the sake of talking and assuming we work out some structure where it would not violate OUTING... how would off-wiki evidence (say a profile at Elance naming the WP articles that the elance user had created) help seal the "sticking" deal or not? I am really asking this - please take it as a real question. Thanks. Jytdog (talk) 02:08, 25 March 2015 (UTC)
Jytdog, I stopped taking Elance and similar services seriously once I spent some time really looking hard at a big stack of them around the time of the change to the ToU. I found a bunch of different profiles all claiming to have written the same (obviously promotional) articles. I found profiles claiming to have written articles that they obviously had not written. I found profiles that had clearly data-mined the contributions of perfectly good Wikipedians who, when asked, totally freaked out that there was an Elance profile that was claiming their edits, and at least two of them that I know of complained to Elance and the profiles were removed. I found profiles that outright claimed to be Wikipedians that they obviously weren't. Anyone who's paying people based on what they're saying in those profiles might just as well be throwing their money down a wishing well, in my opinion. There were a couple of other similar services I looked at around the same time, and found the same issues: at least half the accounts I saw on Elance and other similar sites were probably borderline fraudsters. Some of them may well have been real profiles of real Wikipedians, I will grant, but they were a small minority. So what I'd be looking for would be (a) identification of clearly promotional editing of multiple articles that has resulted in deletion or massive rewriting of articles/removal of all edits/subject of significant disruption such as edit-warring onwiki linked to (b) a profile that claims to have written that very content across (c) multiple accounts that were all (d) used within the last 3 months (the length of time that CU data is available). And the link to the profile would have to be made non-publicly. If the editing isn't viewed as sufficiently problematic that anyone is willing to do anything about it, then it shouldn't be coming to the attention of a checkuser in the first place. And to be honest, even with all that, I'm doubtful that any useful checkuser results would be obtained, in most cases. Incidentally, since a self-declaration of paid editing is precisely what is required by the ToU, that would not be a reason to checkuser the account. Risker (talk) 03:00, 25 March 2015 (UTC)
thanks for your thoughtful answer. i assume the stuff in (a) is connected by "or" not "and", yes? and the profile mentioned in (b) is off-wiki. assuming both of those, that seems very reasonable to me. In your view, would providing the profile non-publicly put the provider at risk of violating OUTING now? (in other words, do we need an exception to OUTING for that.. and if so, is language to make that OK craft-able?) Finally, you ended by saying a self-declaration of paid editing satisfies the ToU (which is true, if it is on-wiki) but I didn't see that in your (a) through (d).. did I miss it? thanks again! Jytdog (talk) 03:17, 25 March 2015 (UTC)
Well, no. I don't think it's policy or even process. It's what I might possibly consider, if I accepted private requests for CU, which I don't. (There are a few exceptions, but they are from users I know very, very well, involving possible returning socks of banned users whose cases I've already worked on.) As a general practice, I don't think it's a good idea for CUs to be working completely sub rosa. I wouldn't be working this kind of a case if it was non-public. Risker (talk) 05:02, 25 March 2015 (UTC)
thanks for replying, but you lost me. what are you saying "no" to? In my view you laid out a thoughtful case for a limited way to use off-wiki information about paid editing; we should be able to find some way to implement that. (i have stopped supporting the amendment as proposed as it is too unlimited... haven't changed my !vote yet but will do so.) Am trying to think of a process that makes sense... like a request to use off-wiki information needs to be submitted in conjunction with a solid case presented at SPI that meets your criteria above, and the off-wiki evidence can only be actually submitted (and submitted only privately) if a CU reviews that case and finds the on-wiki evidence compelling but not sufficient, and grants the request.; maybe limit the ability to present off-wiki data to auto-confirmed users, or users with more than X edits.... you see where i am going? if we can agree on a process that everybody finds OK enough, we can look at it and decide if an amendment to OUTING is needed to make that work. Jytdog (talk) 10:48, 25 March 2015 (UTC)
Risker what would you think of a process like the one I laid out above? (would need an RfC I am sure, but I am testing the waters here). Thanks. Jytdog (talk) 17:39, 29 March 2015 (UTC)
      • I think you misinterpret what I've said. Of course it is generally not possible " to determine a TOU violation for paid editing via checkuser ". The training I received since joining arb com confirms my previous suspicion that it is often not possible to determine anything much by checkuser, except for catching a naïve beginner or a rare mistake by a more experienced sock user--but both are useful. It would have an auxiliary role, as part of the evidence, and it would be run here as everywhere else not just on mere suspicion, but on some sort of plausible evidence. Not bald allegations, as you seems to think I suggest, but plausible ones, as determined by someone other than the person presenting the evidence. I think there is very firm agreement that it is not run, and never would be run, for mere fishing. The mere potential for bad-faith editing is so pervasive that this alone is not reason to run it, or we'd be running it on half the new contributors and many of the existing ones. But given reasonable suspicion, it can help unroll a net of bad-faith users. Typically in determining bad faith, multiple lines of evidence contribute. DGG ( talk ) 18:38, 24 March 2015 (UTC)
        • To make myself perfectly clear: This RFC is about whether or not to permit people to make claims of paid editing by posting links to off-wiki sites as their primary evidence that User xxxx is a paid editor. Alternately, there is a suggestion that such information should be emailed to a checkuser. We already have plenty of evidence that such linking to off-wiki sites (particularly anything that reveals underlying personal or private information about a person) is almost never acceptable to be posted on this site (unless by the user in question) and is more likely to lead to an indefinite block of the person who posts it than it is to any investigation of the other editor; and that fake webpages have been created in the names of multiple users, including at least two checkusers, with the pretty clear intent that they be used to allege inappropriate/paid/COI editing on the part of users in good standing.

          DGG, we both know that the most effective way of reducing the spam-editing/paid editing of Wikipedia would be to raise the notability bar, but you seem to be quite opposed to doing that for reasons that aren't entirely clear to me. What I do know is that the Terms of Use put the onus on the individual editor to declare their paid editing conflict of interest, and they *do not* require or even put any onus on anyone else other than perhaps the WMF to enforce it. On the other hand, we are all (to a person) responsible to the privacy policy and everyone who holds either checkuser or oversight permissions (including you) is responsible to the access to non-public information policy. Risker (talk) 20:37, 24 March 2015 (UTC)

Is suspicion of a ToU violation enough to run a checkuser?

It is claimed above that checks in these circumstances are permitted by the checkuser policy. There are four allowable reasons to run a checkuser according to policy, 1-3 do not permit such a check. 4 might do in an extremely limited set of circumstances, which will almost never occur.

detailed analysis of each of the four points
The following discussion has been closed. Please do not modify it.
  1. Vandalism.
    This can only be determined on the basis of actual edits made and does not support this proposition.
  2. Sockpuppetry.
    Whether someone is using multiple accounts in a manner contrary to the policy or not is independent of whether they are an undisclosed paid editor or not. If there is credible evidence or suspicion someone is sockpuppeting they can be CUed regardless of why that suspicion has arisen or what form the evidence takes. Simply suspecting someone of undisclosed paid editing is not sufficient (in my opinion) to justify a check under this criterion.
  3. Disruption (or potential disruption) of any Wikimedia project
  4. Paid editing, disclosed or undisclosed, is not, in and of itself, disrupting to Wikipedia. Disclosed paid editing is explicitly permitted by the terms of use and not prohibited by local policy; disclosure may explicitly be made at the time of the edit (e.g. in the edit summary). If the editor was genuinely unaware of the need to disclose, then disclosure may be made after the edit. Disclosure may also be made on the talk page or user page before or immediately after an edit or series of edits. Thus even if someone is advertising as a paid editor, they have not violated the ToU until they have actually edited without disclosing they were paid. It is not reasonable to checkuser someone simply because they _might_ cause disruption - this would allow every user and ip address to be checkusered at any time.

Therefore, unless there is credible suspicion that an editor is deliberately intending to insert biased material or other problematic *content* into articles/pages, vandalise or otherwise disrupt the functioning of the project (and this is entirely independent of whether they are being paid) then this does not permit a CU to be run.

  1. Legitimate concerns about bad faith editing.
    Again this is entirely independent of whether someone is a disclosed or undisclosed paid editor. The only point of relevance here is that a CU may (not must) run if there are legitimate concerns that someone has deliberately engaged in undisclosed paid editing, while knowing that disclosure is required, for the purpose of adding or removing material from Wikipedia in bad faith. This criterion does not permit a CU to be run based on the potential for bad-faith editing. What counts as "legitimate" is not specified and so is left to the discretion of the individual checkuser, based on the consensus of other checkusers about similar situations. Personally, I would not regard an off-site claim to be legitimate on its own in almost all cases.

Tl;DR: Simply suspecting someone of violating the terms of use does not permit a checkuser to be run. With evidence or credible suspicion one may be run, but except in a very tiny minority of cases (at most) this will also demonstrate a different, clear reason to run a checkuser and/or lead to a block without the need for a checkuser. Thryduulf (talk) 14:30, 24 March 2015 (UTC)

My reply to Risker above I think covers this. In short, The assumption running through this discussion that because we will make errors we should do nothing, is an way of thinking that is destructive to WP. We have procedures that are failing, and those who have been running them are understandably often reluctant to admit that, but I hoped they would be the first to support changes to increase their effectiveness. I seem to have been wrong there. I can only rely on the community correcting them, if not now, after further experience of our descent into a medium for promotion. DGG ( talk ) 18:38, 24 March 2015 (UTC)
As I note above, there is absolutely nothing in my reply or in Thryduulf's reply to you that even vaguely suggests that there might be errors. We are saying that your contention that we should do useless checkusers on the basis of off-wiki sleuthing that "reveals" information which is often suspect is not within policy. We're calling you out on that, DGG, and saying that you are wrong in thinking that such behaviour is within policy; in fact, your fervent insistence on this is actually kind of worrying me. Nobody has a problem with the idea of investigating the onwiki behaviour of disruptive/biased/POV users or accounts for which there is a reasonable, onwiki basis for suspecting socking. But that has nothing to do with what is being proposed in this RFC, which is off-wiki sleuthing, and which we're saying isn't okay. Risker (talk) 20:44, 24 March 2015 (UTC)
This RFC is not about off-wiki sleuthing. It is only about permitting outing via linking to off-wiki accounts, which is something that I oppose. I'm hoping that this is not read as a claim that off-wiki evidence cannot be privately used as part of a basis for investigating on-wiki violations of policy. In the past we've permitted off-wiki evidence to be sent to checkusers and arbcom as evidence for an investigation, and I can't see any valid reason for that to change. - Bilby (talk) 23:19, 24 March 2015 (UTC)
Simply put, Bilby, allowing those links serves to encourage, if not expect, off-wiki sleuthing. Linking to on-wiki behaviour should be sufficient for just about any violation that is going to involve a checkuser; the only exception might be actual harassment of other users, and that has nothing to do with paid editing, it's a run-of-the-mill behavioural problem. Risker (talk) 23:30, 24 March 2015 (UTC)
There are situations, such as large sock farms, where pointing to off-wiki evidence in private is necessary to show that it is occurring. Handled privately, with a trusted checkuser doing the investigation, has never been an issue before. Situations like Morning277, wikiexperts and similar may require some off-wiki evidence. Similarly, I can think of a number of cases where people were being hired to create a false consensus in articles, which is difficult to prove solely on on-wiki evidence, but can be a serious on-wiki problem. - Bilby (talk) 23:40, 24 March 2015 (UTC)
Well, as noted above, I'm not going to be doing any checks that don't primarily involve on-wiki disruption, although there may be other checkusers who don't mind - none of them seem to be commenting here, though. My observations of at least three of the big pushes to "get rid of" groups of paid editors or multiple accounts of the same paid editor were that the focus was almost entirely on "getting rid of" the editors but huge swaths of their edits were left untouched - which leads me back to wondering where the evidence was that the edits were even disruptive. Failure to correct those problems is another way to encourage paid editors to just keep coming back; as we have known for years, it's simple to create new accounts, and it wouldn't surprise me if half the editors in this discussion don't already know how to manage multiple accounts in a way that doesn't look like socking. Most of us are just too ethical to do that. Risker (talk) 00:02, 25 March 2015 (UTC)
I tend to agree, in that I also don't see much need to worry about paid editors if they aren't being disruptive on-wiki, and accordingly I happily ignore most of them. We do hit problems with ongoing creation of non-notable articles using socks, or disruption through whitewashing, spam links, or false consensus. In those cases I feel we need to act. Before I'd ask for a checkuser I'd want on-wiki evidence of disruption (such as running a sockfarm to evade detection, block or ban evasion, or other concern) to go with any off-wiki evidence, or I wouldn't touch it either.
I agree that part of the problem is with the need to handle old edits. If we allow editors to continue to create socks and create new paid articles, you are right in that it only encourages the behaviour. My concern is that the only way to address some cases is to use private off-wiki evidence in conjunction with on-wiki proof in order to identify the issues. I don't want to see a free-for-all where we link to suspicious off-wiki accounts or out editors in order to enforce it, but I'm comfortable with carefully managed evidence handled privately by arbcom or checkusers in addition to on-wiki evidence of disruption. That's why I'd like to keep this RFC about outing separate to the overall issue of off-wiki evidence. - Bilby (talk) 01:55, 25 March 2015 (UTC)

One needs evidence of on Wiki problems. But often it is not enough without the off Wikipedia evidence. But yes CUs do not appear to occur based on off Wiki evidence and thus the armies of sock puppets involved in undisclosed paid editing can continue without problems. It appears the solution we have taken to this problem is pretending it doesn't exist. They way these people work is they use one account per job. This makes it so that we cannot easily connect the accounts. The similarities in written styles are not enough to convince people. On Wikipedia sleuthing is a great deal of work. Doc James (talk · contribs · email) 00:05, 30 March 2015 (UTC)

Why do you think that checkuser would be of any value in resolving these cases, Doc James? It is not like these guys all sit in an office and use the same IP address. They're working from all over the world, or at least with IP addresses that locate all over the world. I'm not going to get into some of the easier ways to defeat checkuser, but let's just say anyone with half a mind to do it would likely be successful - and the really inveterate paid editors fall into this category. And the #1 way to reduce the number of "paid articles" created would be to raise the notability bar, which is far lower on enwiki than any other large project. Perhaps before looking at outing people, it might be an idea to brainstorm with some other projects on how they address this situation. How do they make it easy for editors to declare their COI? (Talk to German Wikipedia about that, they allow organizational usernames with proof of authorization to speak for the organization.) How do other projects deal with promotional articles on subjects with at best minor notability? (Hint: they're not going to PROD or AfD on most projects.) No other project would consider mass checkusering to be a solution, if for no other reason than they're not needed if you deal with promotional accounts up front. For the record - checkuser is frequently used when dealing with *advocacy*, and we frequently find socks related to that. But that's all on-wiki issues. Show us the on-wiki advocacy, across more than one account, and you'll probably get a CU to look at it when filed at SPI. Risker (talk) 03:11, 30 March 2015 (UTC)
Yes, defeating checkuser is easy (assuming I'm right in what I think it does), and any competent paid editors trying to hide themselves will be doing it for sure. I won't explain how either, but a checkuser run on me now would reveal "Mind your own business, bignose!" Squinge (talk) 09:45, 30 March 2015 (UTC)

No, suspicion of a TOU violation is not enough to run a checkuser. Because checkuser isn't very useful in detecting the TOU violation you're thinking of. -- Euryalus (talk) 09:54, 30 March 2015 (UTC)

I doubt many of them change their IP for each new account. Some of course would. If we ran some checkuser we could investigate to see if it is useful. You may be right and it may not be. Doc James (talk · contribs · email) 10:27, 30 March 2015 (UTC)
You'd be wrong. In fact, we have a hard enough time dealing with *real* sockmasters (whose effects on Wikipedia are visible) because of how common it is that people (whether intentionally or because of the way their ISP works) change IP addresses, wind up on different wifis, and use multiple devices today. Don't kid yourself, Doc James; there have been cases where there is sufficient grounds to run CU because of on-wiki evidence (i.e., paid advocacy editing might be at the root of the suspicion but there were other behavioural issues that warranted a check), and in the majority of cases it's come up as at best possible and usually inconclusive or finding no technical relationship. Meanwhile, we actively ban usernames of people who want to make their allegiances clearcut right in their username and would likely be willing to follow sensible rules about "paid advocacy" editing. For example, why would we not want organizations to update the factual information in their infoboxes (earnings, names of senior executives, etc), with appropriate links? Why would we not want a pharmaceutical company to mention that their fancy new Drug XXX now has a black box warning? Our colleagues on other projects have figured out how to make it work, and we should be able to do that without massive invasion of privacy or the dissolution of a core value of the project. Risker (talk) 13:06, 30 March 2015 (UTC)
I think Risker is right about there being other, and probably more productive, things we can do. We could start by an RfC, or RfCs, raising the notability bar. @Doc James: - how about it? I think if we make it clear why we are doing this, we might be able to stop some of the promotional articles we keep getting. Dougweller (talk) 13:52, 30 March 2015 (UTC)
Yes think that is a good idea. Doc James (talk · contribs · email) 14:02, 30 March 2015 (UTC)
Withdrawn per request of User:Jytdog, due to the confusion this has been causing. Will probably retry when the above RfC finishes. Bosstopher (talk) 17:40, 6 March 2015 (UTC)

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

I propose the addition of "Links to accounts on other websites may be posted, in cases where the content being linked relates to the editing of Wikipedia, and no undisclosed personal information would be revealed by posting the link." Would welcome a rephrase from someone better at writing concisely. The aim of such a change would be to allow for better handling of off-wiki canvassing, coordination, defamation etc. Bosstopher (talk) 21:54, 4 March 2015 (UTC)

Support 2

  • Yes if it does not disclose anyone than it is fine. Doc James (talk · contribs · email) 22:04, 4 March 2015 (UTC)
  • Support the general principle that if an editor self-discloses on an external website then it is not "outing" for editors here to point that out, if it is specifically related to Wikipedia. In other words, if I post on another site and say "Hi there, I'm User:Coretheapple on Wikipedia, and I think this and that such and such about User:PainInTheButt," then that should be allowed. But if I use the username "Coretheapple" on the "Fruit Fly Message Board" but don't link myself to Wikipedia, that's sleuthing and harassment and not allowed. Big difference between the two. In one I am trying to influence Wikipedia or I am talking about Wikipedia, possibly to skirt the rules here. In the other I'm just discussing my passion for fruit flies. Coretheapple (talk) 22:21, 4 March 2015 (UTC)
    But suppose I, not you, have the name Coretheapple on an external site, and I claim on that site that I'm also Coretheapple on Wikipedia and that I'm open to paid editing, bribery, etc? Squinge (talk) 05:13, 5 March 2015 (UTC)
    You need much more than that. We would need said user picking up a job for article X and than Coretheapple editing article X with promotional content. Doc James (talk · contribs · email) 05:25, 5 March 2015 (UTC)
    See my reply in the Discussion section below. Squinge (talk) 05:35, 5 March 2015 (UTC)

Oppose 2

  • Oppose. If I wanted to make a groundless accusation of paid editing against a Wikipedia editor, all I'd need to do is create a fake profile at Elance (or wherever) that names their Wikipedia account as if it were mine. As written, the proposed wording would then allow me to publicize that Elance account here on Wikipedia and so make the dishonest accusation. Links to external accounts should only be allowed if the link has been made here on Wikipedia and not only on the external site, so we know the external account really does belong to the named Wikipedia editor and is not a fake. Squinge (talk) 05:07, 5 March 2015 (UTC)
Ah you also need the Elance to have taken a job that pertains to a specific article. And the Wikipedia account to have edit said article with promotional content around the same time. Doc James (talk · contribs · email) 05:23, 5 March 2015 (UTC)
That's not what the proposal says - see my reply in the Discussion section below, to avoid having this dialog in three places. Squinge (talk) 05:36, 5 March 2015 (UTC) (updated Squinge (talk) 06:12, 5 March 2015 (UTC))
  • Oppose: as per my original !vote in the first discussion, this not only has privacy issues and real life concerns but also flaws such as possibilities of gaming the system (see previous oppose comment and related discussion). Any one could just run some one through this process and out them regardless of sanctions or even actual paid editing. Outing has to be dealt with privately. A bad precedent to set anyway. WP:BOOMERANG will not undo the damage done by outing or even attempted outing... this opens a Pandora's box. --lTopGunl (talk) 15:16, 5 March 2015 (UTC)
How could this have real life concerns? The wording specifically states that a link can only be given if there is no undisclosed personal information within it.Bosstopher (talk) 18:23, 5 March 2015 (UTC)
Obviously linking to an outside profile or identity is known as outing / attempted outing (and it's inherently because it's not disclosed). Any link you connect to a wikipedia account reveals something undisclosed such as real name or workplace (be it a freelance website). Secondly, the possibility of fakes and gaming the system is a greater concern on top of something mentioned by another editor above of everyone opening their junior detective bags against their wikipedia 'opponents' and throwing accusations. --lTopGunl (talk) 18:55, 5 March 2015 (UTC)
  • Oppose for several reasons:
  • The text as it stands is self-contradictory; off-wiki accounts have always been regarded as personal information, and so, unless the editor has disclosed the association of accounts on-wiki, the current proposal amounts to, "Certain types of undisclosed personal information may be revealed ... as long as no undisclosed personal information is revealed."
  • There is no clear line where "undisclosed personal information would be revealed." What counts as undisclosed personal information? It will probably be pretty clear from the contents of someone's Twitter stream which country they live in. Is that personal information? What about what city they live in? What about what football club they support, or what their hobbies are? What sort of drink they prefer? Where's the line here?
  • How much research do we expect someone to have done to see what information might be disclosed? What if someone on Twitter posted a photo of their house, or to their LinkedIn account, several years ago? There seems to me to be lots of room here to claim that the outing was done innocently, not realising what information it made available.
  • And the classic objection: If the link between accounts is disclosed off-wiki, how do we know that the accounts are actually linked? Or, if the link between accounts is assumed based on similarities, how do we know they're actually linked? There is too much room for joe jobs here. GoldenRing (talk) 22:48, 5 March 2015 (UTC)
I would note, also, that Bosstopher's amendment only makes my second point worse; in contentious issues, one person's disruptive editing is another person's good faith attempt to accurately reflect the sources. Do you really want to encourage eg. the army of politically-motivated POV-pushers editing in subcontinental topics, or similar accounts affiliated with Landmark Worldwide, to go doing opposition research on their opponents? Many editors on the side of right and good in those disputes have nonetheless edited disruptively (3RR violations, civility violations, minor personal attacks etc). Doesn't this proposal give the POV-pushers justification for digging into the off-wiki lives of their opponents? GoldenRing (talk) 22:58, 5 March 2015 (UTC)

Discussion 2

I think this might need to be adjusted a bit. Instead of "relates to the editing of Wikipedia" should be along the lines of "relates to the disruptive editing of wikipedia". Or at least add some remark that this is about linking disruptive accounts and not just people who discuss wikipedia off wikipedia. --Kyohyi (talk) 22:04, 4 March 2015 (UTC)

This seems fair are there is no real point to posting undisruptive wikibehaviour, however I'm slightly unsure as to how to go about amending such a thing. Bosstopher (talk) 18:23, 5 March 2015 (UTC)
  • This appears to overlap with the existing RfC. Is that accurate and your intention? Jytdog (talk) 22:08, 4 March 2015 (UTC)
    • This was not my intention and I dont believe it is (see the section bit above the RfCs for context). The other RfC is specifically about paid editing, which (as someone who doesnt participate at WP:COIN) I assume would usually require the disclosing of personal information when posting a link. This RfC was meant to be about more general exceptions to outing policy where real life personal details are not relevant. I meant this to be a parallel proposal to the one above. Bosstopher (talk) 22:14, 4 March 2015 (UTC)
      • hm. in my view, it does read very much on the same matter. and creates confusion. if the above RfC passes, and yours passes... then if I wanted to disclose a link to a profile at elance, and that profile had something like a nationality, or an email, i would be violating your clause. it conflicts. I think the timing is unfortunate and for what it is worth, request that you withdraw it while the other is running. They look very similar. Jytdog (talk) 22:18, 4 March 2015 (UTC)
        • for example here is someone who mistook your RfC for the one above, and then reverted. It is really unhelpful to have launched this now. It would be great if you withdrew this until the other is done. thanks. Jytdog (talk) 00:22, 5 March 2015 (UTC)
          • I doubt that sort of mistake will be made anymore now that the support and oppose sections have been retitled "Support 2, Oppose 2." Main reason I am reluctant to withdraw this is the glacial pace at which this is moving. There was a discussion about amending outing for a few months, then an RfC that lasted over a month, and now I'd have to wait another month for the RfC above to finish to re propose mine. All the while Outing policy can be freely abused by trolls, stalkers and slanderers as they wish. If both RfC's were to pass then all that has to be done is a slight rewording that accurately represents both, (you cant post personal information except on COIN for instance). As it stands I really dont want to withdraw this RfC, but would be willing to do so if asked by a few more people. Bosstopher (talk) 18:23, 5 March 2015 (UTC)

@Bosstopher: My shot at it below. Feel free to take all, none or some from what I've written: "Personal information published off-wiki can be linked to and discussed on-wiki if (a) the editor themself willfully discloses the real-world identity associated with their account on or off-wiki and (b) the information being discussed is relevant to an on-wiki investigation or discussion. Exceptions to this rule include things like discussing an editor's sexual orientation, marital status, address, religious beliefs, or other subjects of a highly personal manner."

I disclose a COI wherever I have one, but also do not disclose my real-world identity, so this is an important issue to me, as I have experienced off-wiki harassment as a result of my disclosure. I'd like to see Wikipedia empowered to chase down the bad guys despite WMF's lethargy (probably not related to a lack of resources btw), but without endorsing harassment.

On the other issue, I do believe this RFC is different and @Jytdog:, you probably shouldn't change the RFC after editors have already voted, because it gives the appearance of looking like they voted for something they didn't. CorporateM (Talk) 22:27, 4 March 2015 (UTC)

you are commenting on the one above. I dated the amendment so any closer will clearly see and I notified every person who commented or voted, including you. I realize it is kind of messy but did the best i could.... i do hear you tho. Jytdog (talk) 22:40, 4 March 2015 (UTC)
Oh, I see. I did not !vote, though I'm not sure I actually have a COI per se. However, I added an amended comment I guess. CorporateM (Talk) 22:59, 4 March 2015 (UTC)
  • All of these proposals are ignoring the elephant in the room - that it is impossible to tell if someone on an external site with the same username as a Wikipedia editor is actually the same person, unless the link is made here on Wikipedia too. Squinge (talk) 05:16, 5 March 2015 (UTC)
Actually it is fairly easy. They state they have written article X on Elance. They were paid to write article X on Elance. And than you find an account that creates article X right at that time in a very promotional tone. The account is "brand new" but has great template abilities and formats references perfect. Content however reads like spam. Doc James (talk · contribs · email) 05:27, 5 March 2015 (UTC)

An example or two

All the same user. Doc James (talk · contribs · email) 05:34, 5 March 2015 (UTC)

That's not what the proposal says - the proposal says nothing about having to have independent evidence first before the external site could be mentioned. And once you start adding "We would need said user picking up a job for article X... and than you find an account that creates article X right at that time in a very promotional tone..." conditionals to the proposed wording and trying to define what constitutes sufficient evidence before the external site can be disclosed, you end up with a nightmare of complication. Once actual evidence of undisclosed paid editing is available, why not just get ArbCom on it and send them the evidence? If actual hard evidence is needed first, I don't see that the proposal actually brings us any benefit. Squinge (talk) 05:32, 5 March 2015 (UTC)
Arbcom states that dealing with paid editing is not part of what they do at this point in time.Doc James (talk · contribs · email) 05:38, 5 March 2015 (UTC)
OK, I'll strike that bit - perhaps we need some other body or method for dealing with undisclosed paid editing, but that doesn't answer the rest of my points about this actual proposal. Maybe a more open question of how to deal with undisclosed paid editing needs to be discussed? Squinge (talk) 05:41, 5 March 2015 (UTC)
This is part of the question. Are we allowed to link to other websites on which undisclosed paid editing is being transacted? This has been linked in the past http://www.wikilinkpro.com/ This is a business that spams Wikipedia is a specific manner. A lot of work is going into trying to deal with this. Are we allowed to link stuff like this? Without such links it is hard to even talk about the issues. Doc James (talk · contribs · email) 05:49, 5 March 2015 (UTC)
Yes, I agree this proposal is part of the question, but I think it needs to be discussed in a wider context rather than as a proposal in its own right. Squinge (talk) 05:52, 5 March 2015 (UTC)
I also agree that we likely need a body to deal with enforcement of the TOU. The only way to move forwards is try to get some consensus on little question. I guess we could have a RfC regarding if we should bother enforcing the TOU at all. There was one on meta that received 80% support. Doc James (talk · contribs · email) 05:52, 5 March 2015 (UTC)
Well, we can't say it's the only way, just one possibly fruitful way. So on the little question do you have any thoughts on my suggestion that "once you start adding [...] conditionals to the proposed wording and trying to define what constitutes sufficient evidence before the external site can be disclosed, you end up with a nightmare of complication"? (I'm awake too early, so I'm going to try to get back to sleep for a couple of hours before I have to start work - will hopefully catch up tomorrow). Squinge (talk) 06:01, 5 March 2015 (UTC)
I do not think it is that much of an issue. Basically do we want people to be able to link to information that may contribute to already existing information that an account is a sock puppet involved with paid editing. Doc James (talk · contribs · email) 19:51, 5 March 2015 (UTC)
So how, precisely, would you word the proposal to allow what you want but prevent "joe jobs" from causing damage, in a way that would be broad enough to work but sufficiently narrow to prevent harm to innocent people? Squinge (talk) 09:20, 6 March 2015 (UTC)
  • Comment The way the RfC is written now may not even be a policy violation. I've had an editor link to an offsite discussion that simply mentioned my WP username making a false statement. Since it did not reveal any real world identification, it wasn't eligible for oversight. I've seen other editors link a WP account to a Twitter account without repercussion since it didn't contain real world info. The admin that reviewed it said it was fine as long as the editor didn't say they were the same person, only that they had the same username (yes, we have admins that are obtuse and anal at the same time). Personally, I'd like it to be against policy and when I likened such links as "opposition research" (which is exactly what it was), the anally obtuse admin said that was a stretch. It seems the proposal may already be allowed. If it's already forbidden, a pointer would be appreciated. --DHeyward (talk) 10:58, 5 March 2015 (UTC)
I think this shows why we need more clear language on this policy. I've seen an editor indef blocked for posting private information when the editor isn't even remotely named in the post. --Kyohyi (talk) 15:07, 5 March 2015 (UTC)
There is clearly a lot of uncertainty around this issue, see this enforcement request for an example of when it went the other way. Bosstopher (talk) 18:23, 5 March 2015 (UTC)
Yes this is why we are trying to develop greater clarity about what is and is not allowed. Doc James (talk · contribs · email) 19:52, 5 March 2015 (UTC)
@CorporateM:Surely this a different issue altogether? I meant this amendment to refer to situations in which no personal information is disclosed. Bosstopher (talk) 18:23, 5 March 2015 (UTC)
I would say that "User:X has a freelance account at eLance" is personal information. Squinge (talk) 09:21, 6 March 2015 (UTC)
  • General comment regarding all of this: The fact that Arbcom does not believe the ToU are policy or that they are bound by such a policy is extremely frustrating for those of us trying to deal with paid editing. However, I don't believe the solution to their refusal is to throw away a protection that has been very important to both our users and the WMF for pretty much as long as Wikipedia has existed. It is not merely a hypothetical that publishing a user's details can lead to them being harassed irl. It has happened, repeatedly, and that's a very significant reason why we have reacted by not allowing personal details to be thrown around in public. If Arbcom won't privately investigate paid editing cases, I suppose we can't force them to do it (short of passing an arbcom recall procedure and then recalling the whole committee at once, which is about as likely to happen as me being made the next Queen of England), but we can create another committee to investigate it, or hand it over to the oversight team, or start talks with the WMF about them globally enforcing it. We could even empower individual admins to do one-person, private investigations-and-blockings, if we want to err on the side of false positives rather than catching nothing at all. All of these are better options than shrugging and saying, "Well, I guess it's open season on the personal details of anyone we think might be getting paid, even if we turn out to be wrong!" A fluffernutter is a sandwich! (talk) 16:39, 6 March 2015 (UTC)
Fluffernutter is your comment about the 1st RfC above? This one has nothing to do with the ToU. More confusion from the inaptly launched 2nd one, i fear.Jytdog (talk) 17:07, 6 March 2015 (UTC)
I'm responding generally to the idea that "we have to be able to link people to offsite identities because Arbcom won't handle them" is a good reason to drastically change a protective policy, and this seemed as relevant a place as anywhere in the discussion to plunk commentary on that. I guess that applies a bit more to paid editing since that's the thing Arbcom explicitly disavowed, but my overarching point is that we shouldn't be responding to "We don't have a mechanism to do this in private right now" with "ok, let's throw away the entire idea of privacy, then!" A fluffernutter is a sandwich! (talk) 17:27, 6 March 2015 (UTC)
what is happening is exactly what i feared. i asked the poser of this 2nd RfC to withdraw it because it appears to me overlap the 1st one. He wrote>: "This was not my intention and I dont believe it is (see the section bit above the RfCs for context). The other RfC is specifically about paid editing, which (as someone who doesnt participate at WP:COIN) I assume would usually require the disclosing of personal information when posting a link. This RfC was meant to be about more general exceptions to outing policy where real life personal details are not relevant. I meant this to be a parallel proposal to the one above." Nonetheless, there is lots of discussion in this RfC about paid editing and what Arbcom will or will not do. I am very unhappy - the exact confusion I was afraid of is happening. Bosstopher I ask you again to withdraw this RfC for now, while the other runs. Jytdog (talk) 17:33, 6 March 2015 (UTC)
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:Harassment/Archive_3&oldid=1089739676"