Jump to content

Wikipedia talk:Conflict of interest/Archive 27

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 20←Archive 25Archive 26Archive 27Archive 28Archive 29Archive 30

Investigating COI policy

It strikes me that we have three broad options for investigating suspected COI (whether it be paid editing or other advocacy), and we now have a discrepancy between outing policy, and the WMF's new essay on paid editing. We really need to gain an idea of what the broad consensus is on this. To whit, this RfC. Cas Liber (talk · contribs) 03:17, 27 January 2017 (UTC)

Milieu 1

MILIEU 1 CLOSED:

Opposed by consensus; 31–9 against. (non-admin closure) George Ho (talk) 00:20, 3 April 2017 (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.
  • "Wikipedia Outing and (Harassment) policies always trump investigating paid (or COI) editing. An editor's contributions should be judged solely on their merit.

Support milieu 1

  1. As stated, I can only support this but have to oppose it as well. Of course OUTING trumps COI investigating in day-to-day Wikipedia. Of course it does. And COI/paid editing investigations always' and I mean always - begin with the quality of edits. A paid editor who makes excellent edits (very well sourced, neutrally worded, reasonable WEIGHT to all sides of issues) would never be noticed. Edits are judged solely on their merit. However once a pattern of promotional edits is established, then the editor starts to be looked at. Which isn't even mentioned here. So of course I support this. Everything always starts with content Jytdog (talk) 05:13, 27 January 2017 (UTC) (added qualification; i did oppose as well Jytdog (talk) 20:29, 28 January 2017 (UTC))
    Excellent point. Yes we nearly always start with looking at the quality of the edits. Doc James (talk · contribs · email) 05:24, 27 January 2017 (UTC)
    Agree with Jytdog's reasoning here. • • • Peter (Southwood) (talk): 06:22, 27 January 2017 (UTC)
  2. I support this without reservation. Even if someone is a paid editor damaging the wiki, the most necessary is only to ban them. Outing them as well will not benefit anybody and may cause real world harm. Thryduulf (talk) 12:28, 27 January 2017 (UTC)
  3. I don't see option 1 and option 3 as being in conflict, and I support a combination of the two scenarios. IMO Outing trumps COI investigations, however outing doesn't prevent private channel communication, nor does it prevent COI investigations that are limited to content already posted on the wiki. --Kyohyi (talk) 14:32, 27 January 2017 (UTC)
  4. IMO support goes without saying: WP:HA is a policy, while WP:COI is a guideline, and AFAIU, policy always trumps a guideline. I would suggest that "what harasssment in not" could benefit from a clause that an investigation of abuse is not harassment. BTW, I didn't read the WMF essay carefully, but IMO it did take WP:HA under consideration. Staszek Lem (talk) 17:15, 27 January 2017 (UTC)
  5. Support the first phrase. But as indicated below I think that the wording is fatally flawed. North8000 (talk) 13:20, 6 February 2017 (UTC)
  6. I agree with the concept that the focus should be on the merit of editors' contributions, not on conducting private investigations of people. wbm1058 (talk) 14:41, 6 February 2017 (UTC)
  7. More or less with Jytdog on this one in both supporting and opposing. If it were possible to have any information which might OUT a person used in such a way as to not make that material publicly available, I could support that. Having said that, I don't know how popular that might be, and whether "regular editors" would use it as a basis to belittle the project. John Carter (talk) 16:43, 6 February 2017 (UTC)
  8. Bilby (talk) 08:05, 30 January 2017 (UTC)
  9. Support providing by "Wikipedia Outing and (Harassment) policies" is meant "Banning all outing (and harassment)", not some subsequently weakened policy. Yngvadottir (talk) 06:39, 15 February 2017 (UTC)

Oppose milieu 1

  1. Cas Liber (talk · contribs) 03:17, 27 January 2017 (UTC)
  2. DGG ( talk ) 04:04, 27 January 2017 (UTC)
  3. Iazyges Consermonor Opus meum 04:35, 27 January 2017 (UTC)
  4. Beyond My Ken (talk) 04:46, 27 January 2017 (UTC)
  5. Doc James (talk · contribs · email) 04:47, 27 January 2017 (UTC)
  6. MER-C 05:06, 27 January 2017 (UTC)
  7. Of course I oppose this, if what is meant is "mark the COI guideline as historic" or "off-wiki evidence shall never, ever be used in any way in WP". Jytdog (talk) 05:18, 27 January 2017 (UTC)
  8. As someone who regularly works WP:COIN issues, the use of outing policy to protect COI editors is problematical. That said, the identity of COI editors is rarely needed in practice. See my Wikipedia:Hints on dealing with conflict of interest problems, which doesn't indicate much need for knowing who is behind an edit. That issue comes up mostly when socking is involved. We already have procedures for dealing with sockpuppets. As an example, right now, there's a issue with someone who is putting up vast amounts of marketing detail for Sony smartphones. These come in about 60 models. The editors involved have tried to create an article for each model, and other editors have condensed those to a table in a single article. We have no idea whether the editor involved is connected with the manufacturer or merely very interested in the product line. In practice, it doesn't matter. There's edit warring and possible IP socking, which can be dealt with via the normal sanctions. We don't need to know if it's someone from a PR firm to deal with the problem. John Nagle (talk) 06:10, 27 January 2017 (UTC)
  9. Smallbones(smalltalk) 10:22, 27 January 2017 (UTC)
  10. Ozzie10aaaa (talk) 10:27, 27 January 2017 (UTC)
  11. Absolutely not — we've been through this so many times. This entirely undermines Wikipedia — it would actively make content worse and more biased. Carl Fredrik 💌 📧 10:47, 27 January 2017 (UTC)
  12. BethNaught (talk) 11:22, 27 January 2017 (UTC)
  13. ThePlatypusofDoom (talk) 13:03, 27 January 2017 (UTC)
  14. The Big Bad Wolfowitz (aka Hullaballoo). Treated like dirt by many administrators since 2006. (talk) 14:14, 27 January 2017 (UTC)
  15. Alanscottwalker (talk) 15:03, 27 January 2017 (UTC)
  16. Coretheapple (talk) 18:01, 27 January 2017 (UTC)
  17. No, no, no. We have WP:CRYBLP, and with all due respect to the very real issue of harassment, I'm tempted to make WP:CRYOUT blue. --Tryptofish (talk) 23:49, 27 January 2017 (UTC)
  18. We can do better than unconditional surrender. Shock Brigade Harvester Boris (talk) 04:44, 28 January 2017 (UTC)
  19. Oppose to first phrase. Second phrase is good. We should tell somewhere that if contributions by editor X are good, no one should conduct COI investigations. My very best wishes (talk) 17:09, 28 January 2017 (UTC)
  20. WMF Legal's statement allows ToU toh trump OUTING. Chris Troutman (talk) 22:49, 28 January 2017 (UTC)
    Outing is also prohibited by the ToU (although not using that word). The terms of use cannot trump themselves. Thryduulf (talk) 03:31, 29 January 2017 (UTC)
  21. Figureofnine (talk • contribs) 22:53, 28 January 2017 (UTC)
  22. Heck no. Wikipedia is not a suicide pact. We are entitled to reasonable defensive measures. Herostratus (talk) 17:28, 29 January 2017 (UTC)
  23. COI and Paid advocacy are almost without exception done to promote a person, product, group, or service of some kind. The actual quality is high, which makes it extremely difficult for the newbies and inexperienced users who are still allowed to patrol new pages to recognise advertorial or other promotion masquerading as articles. Kudpung กุดผึ้ง (talk) 02:01, 30 January 2017 (UTC)
  24. Useles and not compliant with either current practice or policy as written. Changing policy to support this would leave Wikipedia wide open. Only in death does duty end (talk) 09:37, 30 January 2017 (UTC)
  25. Oppose as the outing is not necessarily harassment. TOU violators do not retain all their rights. Graeme Bartlett (talk) 02:31, 31 January 2017 (UTC)
  26. Absolutely oppose. Fylbecatulous talk 13:40, 6 February 2017 (UTC)
  27. As per my comment in the last section above. John Carter (talk) 16:43, 6 February 2017 (UTC)
  28. This is not going to work. QuackGuru (talk) 03:12, 10 February 2017 (UTC)
  29. Oppose. Inexcusably overbroad. James (talk/contribs) 22:46, 14 February 2017 (UTC)
  30. Oppose. Due to wording problem discussed below, harmful as currently written. North8000 (talk) 19:09, 18 February 2017 (UTC)
  31. Oppose per Graeme Bartlett – users who have violated the Terms of Use no longer enjoy the same protection as other editors. Unless stopped, paid editing will within a very few years have completely taken over this project, and dealing with it will absorb all the time and energy of volunteer editors. This applies regardless of the quality of edits: we don't allow promotion of any kind, and that includes well-written promotion. Justlettersandnumbers (talk) 19:35, 2 March 2017 (UTC)

Discussion of milieu 1

clarification - In milieu 1, a concerned editor can only examine the content of the edits - any speculation on accounts is off-limits. Cas Liber (talk · contribs) 05:38, 27 January 2017 (UTC)

Not really a clarification. And it is way down here. Jytdog (talk) 05:55, 27 January 2017 (UTC)

The wording is really defective in a way that plants a poison pill in milieu 1. There are two different statements in milieu 1. The first (which is probably the main operative statement) is a basic one against outing, which due to the searchable data (e.g. every edit that they ever made and exactly when they made it) available to the public on an individual Wikipedian can do immense real world damage to someone. The second statement is farther reaching statement which fewer would agree with. Most would take the question logically literally in which case one would need to agree with BOTH statements to agree with milieu 1, which is a poison pill for milieu 1North8000 (talk) 00:08, 29 January 2017 (UTC)

Actually I find the second sentence far easier to agree with than the first. Comment on the content not on the contributor is a highly related principle that I've seen almost nobody object to. Thryduulf (talk) 03:33, 29 January 2017 (UTC)
If you take it completely seperately I think that you're right. But as a codicil to the statement that it is coupled with it seems to say that COI should be ignored and instead just judge the content on it's merits. IMO such is too far reaching and thus likely to be a poison pill. North8000 (talk) 21:37, 29 January 2017 (UTC)
  • I was pinged to attend this discussion but am having some trouble making sense of it. What's a milieu and, if it's not a concrete proposal, why are we talking about it? This first one contains two sentences. The second one is a general statement of principle and I'm ok with that. The first sentence begs a lots more questions and fails because it uses the word "always". There's no "always" on Wikipedia because of logical conflicts, capricious enforcement and subversive principles like WP:IAR. I'm starting to think the word should be melée rather than mileu – "a confused or heated debate among many participants" :) Andrew D. (talk) 08:20, 6 February 2017 (UTC)
  • I've put in my vote my one caveat - we need to ban outing and harassment unequivocally. Of course edits should be judged on their merits. This is "the encyclopedia anyone can edit"; all limitations on indivduals' ability to edit are supposed to be protective, not punitive; and contributors need to feel safe here. Yngvadottir (talk) 06:42, 15 February 2017 (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.

Milieu 2

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section.

Support milieu 2

  1. Some degree of transparency regarding the follow up of undisclosed paid editing is required. We had degrees of transparency for example during the Wiki-PR investigation and follow up. One should still have mechanisms in place for people to defend themselves. And one needs some publication of results once the investigation is complete. Doc James (talk · contribs · email) 04:49, 27 January 2017 (UTC)
  2. Responding to the exact statement here, I can only respond in part with support, and in part with oppose. For the support part, ToU violations have been discussed openly at ANI and people have been blocked and banned based on these open discussions. See Wikipedia:Administrators'_noticeboard/IncidentArchive942#Earflaps and the section under that, as well as Wikipedia:Administrators'_noticeboard/IncidentArchive943#Proposal_to_ban_FoCuSandLeArN_due_to_undisclosed_paid_editing. Jytdog (talk) 05:22, 27 January 2017 (UTC)
  3. Support transparency is one of our core principles. I do not find the need to secretly observe an editor's edits to determine if they are COI convincing enough to usurp this basic principle. --Tom (LT) (talk) 06:46, 27 January 2017 (UTC)
  4. Support even though it mis-states the WMF essay, which says that that links to the public web can be made without violating WMF policy. Let's get concrete here folks, ArbCom seems to want to ban linking to websites where companies post paid editing jobs. That doesn't out an editor, it outs the company who is hiring a paid editor who is simply not allowed to edit if he does not disclose. Are we supposed to only allow Arbs and admins to discuss this secretely in a star chamber? Smallbones(smalltalk) 10:36, 27 January 2017 (UTC)
  5. support Ozzie10aaaa (talk) 10:42, 27 January 2017 (UTC)
  6. Certainly this does not mean that one can issue baseless complaints in order to OUT people, but rather that when there are strong arguments that an editor is engaged in COI-editing that this be made public. Often uninvolved administrators are hard to come by, and when they do engage — if they don't reveal information that was found in an investigation — the damage can't be fixed. Editors need to be made aware of potential COI-editors in order to address the issues they cause. Carl Fredrik 💌 📧 10:50, 27 January 2017 (UTC)
  7. Support - COI investigations need to be open and transparent for everyone involved. Karst (talk) 12:39, 27 January 2017 (UTC)
    Support per Doc James. ThePlatypusofDoom (talk) 13:06, 27 January 2017 (UTC) I've reconsidered, and I'm not sure about this one.
  8. Agree that COI investigations must not be conducted privately, lest abuses occur. Coretheapple (talk) 18:05, 27 January 2017 (UTC)
  9. Support No question - the "investigation" should be open and transparent, but that does not mean every piece of data has to be revealed - transparent, ethical processes in the real world do this all the time, and the only unethical position is to oppose openness (entirely secret proceedings have been seen as grossly abusive since the 16th century, at least) . Alanscottwalker (talk) 18:38, 27 January 2017 (UTC)
  10. Certainly. Chris Troutman (talk) 22:50, 28 January 2017 (UTC)
  11. It's complicated. Any solution chosen will have flaws. I guess the question is on which side do we want to err. PR editing is a big problem and a serious threat to the project. Open discussion about it a reasonable expectation for a project like this. Herostratus (talk) 17:37, 29 January 2017 (UTC)
  12. Support per Doc. Only in death does duty end (talk) 09:39, 30 January 2017 (UTC)
  13. Support people above have made good statements, if public information is available in that people have already outed themselves, then it should be usable on Wikipedia. Blocking or punishing those who have tried to expose paid editing has caused far more damage to Wikipedia and people than leaving them alone. Graeme Bartlett (talk) 01:47, 31 January 2017 (UTC)
  14. Support – Per discussion of the proposal at Wikipedia:Village pump (proposals)/Archive 134#Proposal for a confidential COI mailing list from four months ago, and the "clear consensus against private evaluation of COI-related conduct, as opposed to open discussion on-site" cited in the closing comment. However I agree with Alanscottwalker that not "every piece of data has to be revealed". Mojoworker (talk) 20:14, 31 January 2017 (UTC)
  15. Support per Smallbones. Arbcom must not be allowed to usurp a power it is ill-prepared to use. James (talk/contribs) 22:52, 14 February 2017 (UTC)
  16. Tentative support. The principle is right, but care and safeguards are needed. Again, Graeme Bartlett is exactly right – blocks and bans of those who've tried to limit paid and COI editing have been, on balance, damaging to the project. Justlettersandnumbers (talk) 19:44, 2 March 2017 (UTC)

Oppose milieu 2

  1. Cas Liber (talk · contribs) 03:17, 27 January 2017 (UTC)
  2. Very dangerous. --Rschen7754 03:21, 27 January 2017 (UTC)
  3. Compromises our harassment policies, which we already have difficulty discussing as a community and applying evenly and fairly. This approach would worsen that. I JethroBT drop me a line 03:52, 27 January 2017 (UTC)
  4. For the reasons outlined here. -- Euryalus (talk) 04:25, 27 January 2017 (UTC)
  5. Iazyges Consermonor Opus meum 04:35, 27 January 2017 (UTC)
  6. Absolutely not (or at least not when such concerns involve the discussion of private information), per my comments in the Arbitration Committee's statement and elsewhere on this topic. GorillaWarfare (talk) 04:40, 27 January 2017 (UTC)
  7. Beyond My Ken (talk) 04:46, 27 January 2017 (UTC)
  8. the way this is worded, I have to oppose it. There needs to be clear limits on what is openly discussed, and there cannot be pure transparency. But there are parts I support per the above. The statement doesn't bring any clarity to the discussion. Jytdog (talk) 05:23, 27 January 2017 (UTC)
  9. Oppose as written. There should be as much transparency as possible, but enough privacy to protect innocent editors from abuse of the system. BethNaught (talk) 11:22, 27 January 2017 (UTC)
  10. Strongly oppose. There is never any cause for releasing private information about a user on wiki without their consent. Thryduulf (talk) 12:30, 27 January 2017 (UTC)
  11. Oppose, I don't see a clear advantage to this, it gives a lot of leeway to fake information and potential witch hunts. --Kyohyi (talk) 14:34, 27 January 2017 (UTC)
  12. Strong Oppose - The proposal strips away the protection of privacy for every single editor. It allows anyone to be dragged into a UPE investigation with the allowance that your public personal information can be posted. The problem is that the damage is done as soon as the information is posted. If it turns out that the person who posted your personal information did so in bad faith, or falsely accuses you of UPE, your personal information has already been posted to the fourth most visited website in the world and copied by mirrors making it impossible to permanently delete. We need only look at the track record at ANI to show how that works (but now with personal information). Moving to this model would be an irresponsible and unethical practice that would have lasting consequences for otherwise innocent people. Mkdw talk 16:38, 27 January 2017 (UTC)
  13. Oppose. per Mkdw. If an only if the investigation proved abuse beyond doubt, then the investigation results may be disclosed, and probably only to a trusted group (OTRS or something). There always a chance of error, an we had this kind of scandal already. Staszek Lem (talk) 17:24, 27 January 2017 (UTC)
  14. Oppose. Who would edit this project on a volunteer basis if we allow outing and harassment by any editor any time someone "suspects" paid editing? Definite no. This would be abused. ~ Rob13Talk 19:16, 27 January 2017 (UTC)
  15. Strong Oppose per Mkdw (and really everyone else in this section). Privacy is sacrosanct. - MrX 19:35, 27 January 2017 (UTC)
  16. This contradicts current privacy policy on using anonymous accounts, and I assume we would like to keep this policy. My very best wishes (talk) 19:37, 27 January 2017 (UTC)
  17. I oppose this, because there is information that needs to be kept private, but I also want to remind everyone of the very strong negative consensus at Wikipedia:Village pump (proposals)/Archive 134#Proposal for a confidential COI mailing list. It wasn't just consensus against the proposed list, but also consensus against any kind of COI investigation off-site, because of the need for transparency. Methinks we have consensus that COI needs to be investigated, but it must not be investigated on-wiki, and it must not be investigated off-wiki. Facepalm. --Tryptofish (talk) 23:55, 27 January 2017 (UTC)
  18. feminist 10:39, 28 January 2017 (UTC)
  19. Oppose - we don't need outing fishing expeditions. Ealdgyth - Talk 14:47, 28 January 2017 (UTC)
  20. Oppose Outing, which due to the searchable data (e.g. every edit that they ever made and exactly when they made it) available to the public on an individual Wikipedian can cause Wikipedia to do immense real world damage to someone. North8000 (talk) 00:12, 29 January 2017 (UTC)
  21. Oppose We've always respected people's privacy beyond simply their IP address. To discard that in order to detect a relatively small number of paid editors is a lot to surrender, especially given the risks of abuse. - Bilby (talk) 08:06, 30 January 2017 (UTC)
  22. Nope. "I suspect you of paid editing or a COI" is not sufficient grounds to out someone, due to the extremely vague nature of the statement and its potential for abuse. Sam Walton (talk) 11:20, 30 January 2017 (UTC)
  23. Oppose as Samwalton9 said "I suspect you of paid editing" should never be enough to expose a link between someone and their on wiki account. HUGE amount of room for abuse, this may have good intentions but as they say the path to hell is often paved with good intentions. --Cameron11598 (Talk) 17:35, 1 February 2017 (UTC)
  24. Oppose too many ways this could go wrong. John Carter (talk) 16:45, 6 February 2017 (UTC)
  25. Oppose, strongly. No. Fylbecatulous talk 01:23, 7 February 2017 (UTC)

Discussion of milieu 2

  1. I am somewhere between 2 & 3. I an not convinced that open discussion in at least some such cases would lead to danger, and I do not think that those doing deliberate undeclared paid editing on an extensive scale should have their privacy protected. My disagreement here is why I did not sign the letter. DGG ( talk ) 04:15, 27 January 2017 (UTC)
  • My !vote sounds closer to "I think you'll find that it's more complicated than that" than to either support or opposition. We could benefit from "some transparency into investigations" (e.g., the level of transparency that CheckUsers provide for socks), but the project will be harmed by "extensive transparency", especially if anti-COI warriors are allowed to pass off blatant disrespect for privacy as "just trying to be transparent". No matter where each of us fall on the spectrum of aversion to paid editing, and even if we were naïve enough to believe that COI charges were never used as a tactical attack in a content dispute, we all make mistakes of judgment as well as fact. Our rules should therefore minimize the risks that our honest mistakes create for other people. WhatamIdoing (talk) 05:26, 27 January 2017 (UTC)
  • There certainly could be some limits on what is discussed openly, e.g. no real names of individuals. I don't think we want to say "no employers can be discussed", the company paying the paid editor is very relevant, but we could have something like "no employers who are not involved in paid editing" can be disclosed. Smallbones(smalltalk) 11:10, 27 January 2017 (UTC)
    • I can accept that only on occasions were it is as close to 100% certain it is possible to be that all the following are true:
      1. There is/was an employer-employee or contractor-contractee relationship (not a "my dad works for this company", "I like this company's products", "I work for a competitor and want to discredit them", or even simple mistaken identity or incorrect guess, etc)
      2. The employer/contractor asked the person to edit Wikipedia on their behalf (just as much as people are not responsible for actions taken in their name they had no knowledge of nor reasonable reason to suspect it, we must not accuse companies of breaking our terms and conditions unless they actually did so and not because a random employee acted on their own initiative without their manager's knowledge)
      3. Both parties knew that it was against our terms and conditions (on more than one occasion I have explained to someone that paying somebody to write a Wikipedia article on their behalf is not allowed without disclosure, etc - they genuinely did not know.)
      4. That a breach of the terms of use has occurred and no good-faith attempt to comply, including retrospectively, has been made..
      5. The person wanting to disclose the link between editor and company can show, without outing anybody, that all of the above are true. Thryduulf (talk) 17:17, 27 January 2017 (UTC)
Ay, [[ User:Thryduulf it is untrue that paying somebody to write a Wikipedia article on their behalf is not allowed. What is against the ToU and WP:PAID, is for the paid editor to work in WP without disclosing their "employer, client, and affiliation". Nothing bars people from paying other people to edit WP. Please correct what you wrote. Jytdog (talk) 00:15, 28 January 2017 (UTC)
Italicised portions added for (hopefully) increased clarity. Thryduulf (talk) 00:42, 28 January 2017 (UTC)
Thanks! There is a big difference before/after. :) Jytdog (talk) 02:05, 28 January 2017 (UTC)
The above discussion is closed. Please do not modify it. Subsequent comments should be made in a new section.

Milieu 3

There is consensus for the proposal with the obvious caveat, that this approach needs a lot more details and clarification.Many have clarified that other bodies shall only refer to editors who have been vetted by the community to handle sensitive and personally identifying information.There has been concerns about the use of the word only as it seems to nullify on-wiki processes based on CU and behaviorial evidence.Winged Blades Godric 11:50, 18 April 2017 (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.


  • "The balancing COI and privacy/outing means that the only option is that people investigating COI must submit information in private to the relevant people. Currently this is the arbitration committee and/or the WMF, but other bodies could be considered if there is consensus for this.

Support milieu 3

  1. Cas Liber (talk · contribs) 03:17, 27 January 2017 (UTC)
  2. Rschen7754 03:20, 27 January 2017 (UTC)
  3. ~ Rob13Talk 03:28, 27 January 2017 (UTC)
  4. My support is contingent on the idea that "other bodies" only refers to editors who have been vetted by the community to handle sensitive and personally identifying information. Consensus supporting access to that info for all confirmed editors, for example, is not a acceptable approach to handling cases of paid editing or advocacy. I JethroBT drop me a line 03:49, 27 January 2017 (UTC)
  5. As per JethroBT, this should be restricted to users who have been vetted and have signed the non-disclosure policy for private information. NorthBySouthBaranof (talk) 04:06, 27 January 2017 (UTC)
  6. Only real option here. Iazyges Consermonor Opus meum 04:35, 27 January 2017 (UTC)
  7. Support, though I think "information" and "the relevant people" are really the crux of the matter. I know that (at least during my tenure) the ArbCom and the functionaries have often been resistant to opening these investigations. I think there is little understanding or consensus on what kind of evidence is necessary to incriminate an editor as an undisclosed paid editor. GorillaWarfare (talk) 04:40, 27 January 2017 (UTC)
  8. Support a new group of individuals. Arbcom does not have the numbers / desire. We also need mechanisms to involved all admins and some public disclosure of certain details to help the entire community with detection. This is not exclusive of some public efforts. Doc James (talk · contribs · email) 04:58, 27 January 2017 (UTC) Not the only option. It is one option that is potentially useful among many. UPE is not a simple problem and will not be solved by one simple change. Doc James (talk · contribs · email) 23:18, 29 January 2017 (UTC)
  9. If the "information" referenced here refers only to "off-wiki personally identifying information" then this the first part of this is accurate and I support it. I cannot give unequivocal support because this doesn't specify the kind of information, and the arbcom-only option doesn't reflect community practice nor WP:BLOCKEVIDENCE. Jytdog (talk) 05:31, 27 January 2017 (UTC)
  10. Ozzie10aaaa (talk) 10:30, 27 January 2017 (UTC)
  11. Certainly there should be as transparent an investigation as possible given certain limits, but in the cases where a transparent investigation is limited a more private body should investigate. ArbCom would be the worst possible body to do this. It sometimes seems like they want to veto the Terms of Use change, which they are not allowed to do. Smallbones(smalltalk) 11:18, 27 January 2017 (UTC)
  12. Support. There needs to be further discussion to define specifics, but this protects the users privacy. I disagree with Doc James that there should be public disclosure of some details, if by "some details" they mean private information about editors. Thryduulf (talk) 12:34, 27 January 2017 (UTC)
  13. Support - If we have someone that's an undisclosed paid editor, the only way to prove it, is to share information that would be considered outing, and since openly sharing that information would or could lead to harrasment, it would need to be shared privately with trusted individuals , as this proposal suggests. I fully support it! ƘƟ SĦ 12:59, 27 January 2017 (UTC)
  14. Support - As I said in 1, I don't see a conflict between these two positions. Information that is already on Wikipedia can be used in a COI investigation, and information that is off-Wikipedia can be communicated privately. --Kyohyi (talk) 14:36, 27 January 2017 (UTC)
  15. Support. My very best wishes (talk) 19:38, 27 January 2017 (UTC)
  16. Qualified Support - I interpret "... must submit information in private..." to mean "... must submit undisclosed personally identifying information in private...". This is roughly the status quo, but it also allows for a separate body to be formed to help with such investigations, which would be a step forward toward addressing the paid editing/COI problem.- MrX 19:42, 27 January 2017 (UTC)
  17. Support. The type of undisclosed paid editing that I identify as being the greatest threat to the English Wikipedia are sophisticated attempts to willfully undermine and avoid our policies and guidelines. They work beyond most of our detection methods and often involve sock puppetry and promotional editing across dozens of articles. I think a subcommittee comprising of functionaries and administrators (who have experience at OTRS; and thereby effectively has signed the WMF 'confidentiality agreement for nonpublic information') could conduct these investigations and report their findings without revealing any personal information about the editor. A 'UPE block' could only be unblocked by an individual on the subcommittee who has access to any privately submitted information -- much how our OS and CU blocks work. Mkdw talk 21:19, 27 January 2017 (UTC)
  18. Or something approximately like this – I'm not convinced that we have a user-group that is willing to do this and also has community support to do this. Please see also my oppose at M#2. --Tryptofish (talk) 23:58, 27 January 2017 (UTC)
  19. Support with the obvious caveat, that this approach needs a lot more details and clarification. Considering ArbComs recent response, I'd prefer a separate "other body" of less-involved Wikipedians to handle such issues. As an aside, we should also intensify our dialogue with WMF Legal to resolve such issues, instead of criticizing procedural technicalities of their good-faith input. GermanJoe (talk) 12:43, 28 January 2017 (UTC)
  20. Support, with the proviso that another body should be created. Preferably one that is large enough to effectively handle the issue — we're not talking about minor cases, but what is a full-blown industry serving up paid editing. ArbCom is underequiped to handle these issues, and considering the inteheavynse burden already placed on ArbCom and the high degree of absense of their members it seems very unwise to place any more tasks on their backs. Carl Fredrik 💌 📧 13:29, 28 January 2017 (UTC)
  21. Support, but qualified. I'd rather not have a new group, simply because the more we expand the number of people with access to what we regard as private information, the greater the risk of potential misuse. I'd rather see it as a role that an existing group of trusted editors (be that ArbCom, CUs, or whatever) can take on. I also want to note that this proposal has not had a lot of support in the past among the wider WP community, so getting broad consensus to the approach might be difficult. - Bilby (talk) 08:09, 30 January 2017 (UTC)
  22. Support a new group of editors. This would be both more private and, possibly, more effective. Sam Walton (talk) 11:19, 30 January 2017 (UTC)
  23. Support per GorillaWarfare, Mkdw although I do share I JethroBT's concerns. --Cameron11598 (Talk) 17:40, 1 February 2017 (UTC)
  24. Perhaps not perfectly worded, but seems to encapsulate the right spirit. Samsara 23:48, 1 February 2017 (UTC)
  25. Necessary, but not sufficient. MER-C 06:52, 2 February 2017 (UTC)
  26. Wording needs help, though. John Carter (talk) 16:46, 6 February 2017 (UTC)
  27. This balances 2 significant needs - the right of all users for privacy and the protection of Wikipedia from undisclosed COI issues. עוד מישהו Od Mishehu 04:08, 14 February 2017 (UTC)
  28. Support as written. Since the devil is in the details, "be considered" is an important part of this proposal. North8000 (talk) 19:13, 18 February 2017 (UTC)
  29. Support Not all current arbs will want to do this, but the group can be expanded if necessary to include Arb + Checkusers + Oversighters. There would of course be a number of details to be worked out. DGG ( talk ) 17:40, 27 February 2017 (UTC)
  30. Support, partially. If this is the first part of an investigation/discussion process, then well and good – it may indeed be the best way forward; but all definitive findings absolutely need to made public/ fully transparent per Milieu 2. Justlettersandnumbers (talk) 19:58, 2 March 2017 (UTC)
  31. Primefac, thank you for reverting the close. I suppose I agree with this particular proposal, knowing fully well it's not really new--the clarification may be useful, though. However, and I think this was the problem with the close, maybe, "Support a new group of individuals", as slipped in by Doc James, will have to be the subject of a separate RfC because it's huge (I appreciate NorthBySouthBaranof's comment on extreme vetting--you like that term, NBSB? :)). And speaking as an arb who has dealt with many COI cases, both before and during my Arbness, I find Doc James's suggestion that ArbCom has no desire to tackle COI cases to be in poor taste, and the proposal for a "task force" to be little more than a runaround. We/I cannot stress enough that the problem is and has always been not the lack of desire to tackle COI cases, but the ease with which some editors think we should fling around private information in violation of OUTING. If a case/query is presented properly, ArbCom or individual arbs will be more than happy to investigate, and I think those of you who have sought the advice of ArbCom/individual arbs know this. Drmies (talk) 15:10, 17 March 2017 (UTC)
  • User:Drmies If members of arbcom have changed their position on this would be happy to be informed differently. My statement more pertains to before the most recent election so things may have changed somewhat but many of members of arbcom do remain the same. Doc James (talk · contribs · email) 15:33, 17 March 2017 (UTC)
  • Some may have remained the same (one hopes) but that doesn't mean that they do or did not have the "desire" to handle COI matters. News flash: I just handled one. Drmies (talk) 16:42, 17 March 2017 (UTC)
  • @Drmies: Over the last few months, I have understood ArbCom's feelings much as you just described them here. But, as someone who probably has been involved in these discussions for far too long, I clearly remember that, as of a year or two ago, it was being widely stated by some functionaries and, I think, some arbs, that ArbCom had discussed taking on COI investigation, and had decided that it was simply something that ArbCom (of that time) did not want to do. It's definitely something that has been repeated a lot, so I think it may be understandable that some editors would still think that this was the situation. --Tryptofish (talk) 22:04, 17 March 2017 (UTC)

Oppose milieu 3

  1. This one must be opposed as it is proposed, as it ignores the fact that COI/ToU are investigated every day using purely on-wiki evidence, and with regard to off-wiki personally identifying information, the existing WP:BLOCKEVIDENCE policy offers CU and oversighters as well as Arbcom as vehicles. And in any case Arbcom has said it is unwilling to address paid editing. Perhaps Arbcom would do so if an RfC empowering them passed, but this option as stated should not be it. Jytdog (talk) 05:26, 27 January 2017 (UTC)
    Comment This is a good option but NOT the ONLY option. Agree that COI/ToU is investigated all the time based on only Wiki evidence per WP:DUCK Doc James (talk · contribs · email) 05:37, 27 January 2017 (UTC)
  2. I don't see a convincing need as to why COI edits need to be investigated in secret. We investigate other processes openly and I see no reason why more eyes won't help here. --Tom (LT) (talk) 06:48, 27 January 2017 (UTC)
  3. Oppose, but only because badly worded. COI editing is discussed on-wiki in various ways. For example, Focusandlearn was banned at ANI based on behavioural evidence, and sockfarms are checkusered with account associations published publicly. The aim of this proposal, that information which would constitute outing on-wiki should be submitted for private consideration to a suitably empowered body, is something I generally agree with, subject to details. BethNaught (talk) 11:22, 27 January 2017 (UTC)
  4. Oppose: What is ignored here is that we are not interested in name, address and phone number for paid editors, we only request and require what entity is paying for the edits, and this is already circumscribed in policy (so any evidence beyond paymaster evidence ("employer, client, and affiliation" ) should not be onwiki). Alanscottwalker (talk) 12:07, 28 January 2017 (UTC)
  5. Oppose. Outside of Arbcom's remit, which is to settle unresolved disputes. Not investigate potential COI's. Given Arbcom's recent history of taking unilateral action after closed door proceedings, I oppose anything that would increase this trend. Only in death does duty end (talk) 09:41, 30 January 2017 (UTC)
  6. Oppose we can't expect that these few people have to capacity to deal with the problems. Graeme Bartlett (talk) 02:33, 31 January 2017 (UTC)
  7. Oppose – The wording of milieu 3 as the "only option" is painting with far too broad a brush. See the concerns over secret Star Chamber proceedings in the discussion at Wikipedia:Village pump (proposals)/Archive 134#Proposal for a confidential COI mailing list from four months ago, and the "clear consensus against private evaluation of COI-related conduct, as opposed to open discussion on-site" cited in the closing comment. I'd support establishment of such a group of "relevant people" but only for clearly defined private data, which would be reported back in redacted form for on-wiki discussion. Mojoworker (talk) 20:50, 31 January 2017 (UTC)
  8. Strongly oppose. We need less kangaroo courts, not more. James (talk/contribs) 22:49, 14 February 2017 (UTC)
  9. Oppose. The close of this was overturned, and I assume will stay that way, but in case not I'm adding an oppose.
    The introductory text said: "people investigating COI must submit information in private to the relevant people". Yes, and that can be admins and functionaries. Then it says: "Currently this is the arbitration committee and/or the WMF, but other bodies could be considered if there is consensus for this." But currently it isn't the arbitration committee or WMF. Almost all COI work is done by volunteers at COIN and elsewhere, and it works fine; personal information is almost never needed. The big problem is that we don't have enough volunteers, and COI/paid editing is ubiquitous, but that's not something a dedicated group can change.
    In the overturned close, this section of the RfC was interpreted as having shown consensus for a task force to "detect, deter and prevent" COI editing, which might publicly name paid editors. But this August–September 2016 RfC proposed something similar, not a new body but a group of functionaries to set up a separate mailing list for COI (and nothing about naming people in public). The response was a clear no: 4 in support, 21 against. Any RfC that aims to overturn that should be held at the village pump; should refer back to the 2016 RfC and make sure everyone who commented knows about the new one; and the proposal needs to be very clear, preferably in a stand-alone RfC so that it isn't mixed up with other issues. SarahSV (talk) 23:28, 17 March 2017 (UTC)
    Primefac has indicated that he's going to re-close this. That would be a mistake, because it's malformed, unless the close suggests another RfC on this particular point (creation of a new group). I don't want to ping the 25 people who commented in the August 2016 RfC, but if there's a chance that that consensus will be overturned here, someone ought to. SarahSV (talk) 21:12, 18 March 2017 (UTC)
  10. Oppose a "task force" for the reasons stated by myself and SlimVirgin [1]: in particular, the COI and/or paid-editors (disclosed or undisclosed) might end up on or running the task force, e.g. regulatory capture. Also because of the previous RfC described above by SlimVirgin. --David Tornheim (talk) 04:12, 18 March 2017 (UTC)
  11. Oppose. This just sort of slipped on by me when it first was introduced, and I didn't consider its implications. No, I don't see what value would be added by some kind of body or task force. My main concern is that it appears to be just another bureaucratic mechanism that can be subject to abuse and is unlikely to have a significant impact. Indeed, COI or paid editors could wind up on such a panel. Or, given the maturity level I've seen displayed by administrators on various occasions, the reverse could happen and innocent editors could wind up unfairly dealt with. Makes far more sense to improve the standards for notability for companies and people, That would provide an even playing field less conducive to abuse by paid and COI editors. Coretheapple (talk) 21:21, 18 March 2017 (UTC)

Discussion of milieu 3

  • Only possible option - I have been on the committee and I have not seen this to be overly onerous to date. Cas Liber (talk · contribs) 03:17, 27 January 2017 (UTC)
  • I suppose we are making some progress, because 2 years ago the arb com as it was then constituted overwhelmingly refused to consider the possibility that it would have a role in such investigations. DGG ( talk ) 04:12, 27 January 2017 (UTC)
    • @DGG: What do you think about an alternative group of editors, though? I support this with the hope that this will not fall on the Arbitration Committee. See the proposal I laid out at the ArbCom noticeboard talk section. ~ Rob13Talk 04:24, 27 January 2017 (UTC)
      • BU Rob13 I think the situation is adequately covered by "We may also disclose your personal information if we reasonably believe it necessary to detect, prevent, or otherwise assess and address potential spam, malware, fraud, abuse, unlawful activity, and security or technical concerns." Creating new user groups is to invite even more bureaucracy creep, particularly where you are at odds with the opinions you voiced on other related discussions. A new group pf New Page reviewers has been created who could fulfill this role as they come under heavy vetting (at least by me) before they are accorded the right. What the community shpuld be looking at are ways to plug the gap left which under your (and others' ) insistence, still allows patrolling to be done by the least likely and least qualified users, namely the newbies and youngsters for whom maintenance tasks are a magnet. Kudpung กุดผึ้ง (talk) 02:47, 30 January 2017 (UTC)
        • @Kudpung: Actually, new page patrolling is not a barrier to the most malicious forms of paid editing in any sense. Malicious paid editing in modern times takes two forms: article hijacking for non-notable subjects and non-neutral editing on notable subjects. Article hijacking involves moving a little-watched existing stub to a new page title to elude new page patrollers when creating a new article. Non-neutral editing on existing articles for notable subjects is outside what new page patrollers will see. Disclosure on-wiki under that sentence of the Terms of Use would be great if everyone who tries to out another editor does so with good intentions. They do not. If it would not violate my privacy, I would send you some details about past instances where people have tried to maliciously out me to make me the target of harassment. Under the proposal to conduct paid editing witchhunts on-wiki, those who would maliciously out another editor have a defense that will likely stand up at ANI if they find some obscure reason to cry paid editing before doing the outing. Sure, we could block the editors who do that, but the damage is done, often in a permanent manner. If we need bureaucratic creep to keep editors from being harmed by outing and harassed in real life, I'll gladly take the creep. ~ Rob13Talk 03:14, 30 January 2017 (UTC)
  • Asking a group of volunteers to address a problem that many do not appear to think is a problem and do not want to be involved with even if they though it was means one of two things 1) they will simple not do it 2) they will do a poor job. Handing the work to arbcom is not a solution. Creating another group of individuals to address it might be. Doc James (talk · contribs · email) 04:55, 27 January 2017 (UTC)
Doc James, see my response to BU Rob13 above. Kudpung กุดผึ้ง (talk) 02:47, 30 January 2017 (UTC)
  • @Jytdog: I don't quite see your objection. Mi3 does not ban any private investigation. It is that the results are not to be widely disseminated. Of course, investigations would pus a huge burden on any committee. But A reasonable judgement (biased towards presumption of innocence) IMO would not that burdensome. Staszek Lem (talk) 17:29, 27 January 2017 (UTC)
User:Staszek Lem - The biggest problem is that as this is worded, it makes no distinction between evidence that is purely on-wiki and evidence that is off-wiki. It basically means closing down COIN and immediately killing any ANI thread where on-wiki diffs providing evidence of COI/paid editing are brought; this is horrible. (and no, this is not a tendentious reading -- there are people in the community who oppose any discussion of COI and believe that "content not contributor" should be rigorously followed). With regard to off-wiki evidence, the second problem is "Currently this is the arbitration committee and/or the WMF" which is not accurate per WP:BLOCKEVIDENCE and the practice of some admins who already block for ToU violations. And there are arbs who have said that they will not look at this stuff due the legal risks. See my remarks about that way below. I am pretty confused that anybody has supported this without reservation. The details matter. Jytdog (talk) 20:16, 27 January 2017 (UTC)
User:Casliber the key bit of WP:BLOCKEVIDENCE is if a "user needs to be blocked based on information that will not be made available to all administrators" I am happy to share off wiki details with all other admins by email. We are not using CU data in this or private WMF data, we are using publicly avaliable data elsewhere on the internet. So in other words this does not need to restricted to abrcom/CUs Doc James (talk · contribs · email) 08:00, 30 January 2017 (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.

Milieu 4

There is consensus for the above proposal, with a condition that the proposal must be clarified to remove vaugeness, and that any information released must be limited to "employer, client, and affiliation".

More specifically, the information that is to be clarified is:

  1. Who is doing the investigating? (this looks like it's covered by Milieu 3)
  2. What information is to be released? The proposer has stated in the discussion below (and other editors agreed) that the information that is released is to be limited to "employer, client, and affiliation". This renders the argument of wp:outing invalid, which really was the only argument brought up on the oppose side.


Gamebuster19901 (Talk║Contributions) 08:15, 2 June 2017 (UTC) (non-admin closure)

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.

We need to balance privacy provided to those editing in good faith against the requirements of addressing undisclosed paid promotional editing. To do so can be achieved with a private investigation with some release of results publicly to help with the detection of further related accounts. These details may include the name of the Wikipedia editing company with which the account is associated (such as for example the connections drawn here)

Support milieu 4

  1. Doc James (talk · contribs · email) 05:04, 27 January 2017 (UTC)
  2. Ozzie10aaaa (talk) 10:32, 27 January 2017 (UTC)
  3. Carl Fredrik 💌 📧 10:51, 27 January 2017 (UTC)
  4. Sure - but needs details. Smallbones(smalltalk) 10:59, 27 January 2017 (UTC)
    ThePlatypusofDoom (talk) 13:06, 27 January 2017 (UTC) too vague.
  5. Staszek Lem (talk) 17:40, 27 January 2017 (UTC)
  6. The wording is precise and good. My very best wishes (talk) 19:39, 27 January 2017 (UTC)
  7. DGG ( talk ) 19:57, 27 January 2017 (UTC)
  8. Beyond My Ken (talk) 20:11, 27 January 2017 (UTC)
  9. But the devil is in the details. --Tryptofish (talk) 23:59, 27 January 2017 (UTC)
  10. support in part. it would be good if we had, for our own use and for the use of people who hire paid editors, a public list of persistent violators of the ToU that listed Employer, client, affiliation, and the relevant user accounts. But there are serious problems with this proposal, as described in my oppose. Jytdog (talk) 03:31, 28 January 2017 (UTC)
  11. Support, assuming it is limited to the paymaster information that is required and requested by policy (and to onwiki evidence), we have no interest in other information, except paid editor "employer, client, and affiliation" . Alanscottwalker (talk) 12:14, 28 January 2017 (UTC)
  12. WMF's statement would allow an even looser interpretation, but this is OK. Chris Troutman (talk) 22:51, 28 January 2017 (UTC)
  13. Noting that the WMF since 2011 has deliberately refused to implement or develop any editing restrictions (or help) that would address this issue at its core. The Foundation staff are notorious at polarising, or unilateral action, or even doing nothing when it suits them. Issues of this kind should be entirely devolved to the local Wikipedia communities for interpretation and implementation. In short: there is no reason to believe that the WMF staff are any more competent than we are ourselves. Kudpung กุดผึ้ง (talk) 02:54, 30 January 2017 (UTC)
  14. Only in death does duty end (talk) 09:43, 30 January 2017 (UTC)
  15. Support, it should be as private as possible to avoid harming the innocent, but people that fail to disclose required information can expect that information to be disclosed about them. Graeme Bartlett (talk) 02:36, 31 January 2017 (UTC)
  16. Qualified support if the details for how this would be done are developed and provided. Even then, however, there might well be serious unanticipated problems. John Carter (talk) 16:49, 6 February 2017 (UTC)
  17. Good faith editors are allowed to have alterative accounts. If admins disagree then we should make policy more clear. This is a step in the right direction. QuackGuru (talk) 03:15, 10 February 2017 (UTC)
  18. Support per Graeme Bartlett and based on my support of Milieux 2 and 3 above. This is the right direction, but will need to be carefully developed and refined. Justlettersandnumbers (talk) 20:03, 2 March 2017 (UTC)

Oppose milieu 4

  1. The right to privacy of your personal information is not lost simply because you were not editing in good faith. That results in the loss of your editing privileges, it does not and should not entail the possibility of real world harm. Thryduulf (talk) 12:35, 27 January 2017 (UTC)
  2. Pretty much the same reason as 2, I don't see a real advantage to such a list. --Kyohyi (talk) 14:38, 27 January 2017 (UTC)
  3. Private investigation by whom? Random editor #5128490? What results are released? Must strongly oppose due to vagueness. This could be abused to the point of ruining someone's life. ~ Rob13Talk 19:17, 27 January 2017 (UTC)
  4. I can't support this because "with some release of results publicly to help with the detect of further related accounts" is too open ended for something that will be enshrined in policy. I agree with BU Rob13 that by whom needs to be defined, as does "some release of results". In principle, this milieu seems reasonable, but I would want to know what the boundaries are.- MrX 19:49, 27 January 2017 (UTC)
  5. have to oppose this as it is too vague about what off-wiki information could be released (it should be limited only to "employer, client, and affiliation" no ifs and or buts about that.), and also about who decides if the evidence is "good enough", as well as controlling what is released. These are three essential things and this proposal is silent on them. Jytdog (talk) 03:31, 28 January 2017 (UTC)
  6. Oppose Outing, which due to the searchable data (e.g. every edit that they ever made and exactly when they made it) available to the public on an individual Wikipedian can cause Wikipedia to do immense real world damage to someone. North8000 (talk) 00:15, 29 January 2017 (UTC) First it is dangerously vague on the carefulness of the process and on what would be disclosed. If the investigatoin process were very careful and thorough, and the disclosure very narrow (probably just that they are a paid editor) I would support this. North8000 (talk) 19:17, 18 February 2017 (UTC)
  7. Far too vague, per Thryduulf and MrX. Sam Walton (talk) 11:15, 30 January 2017 (UTC)
  8. Oppose, per BU Rob13, MrX, and Jytdog --Cameron11598 (Talk) 17:43, 1 February 2017 (UTC)
  9. Oppose due to vagueness. Samsara 23:46, 1 February 2017 (UTC)
  10. Oppose, without precision, this proposal allows for way too much ambiguity on how much personal information can be posted publicly. Questions that need to be answered are: investigation by whom? Results "may include" but not not limited to? Who determines which parts of the results are released publicly? When it comes to personal information, do we trust investigators on the grounds of good-faith and presume they will exercise good judgement in outing people? Again, ANI and SPI show a much different track record, but now with personal information at stake. Mkdw talk 00:07, 4 February 2017 (UTC)

Discussion of milieu 4

Could you be more specific about the circumstances of such release? GorillaWarfare (talk) 05:27, 27 January 2017 (UTC)
I see a process of (1) evidence is submitted (2) evidence is privately examined by a group given authority to do so (3) if the concerns are founded the account is blocked (4) if there is strong evidence that the account is related to a Wikipedia editing PR firm those details are disclosed to help with further follow up (such as was sort of done in the Wiki PR case). Doc James (talk · contribs · email) 05:35, 27 January 2017 (UTC)
Doc James this would be much stronger if it said something like "results confirmed by CU, Oversight, or Arbcom". Additionally one of the very hard things, is even if we identify someone as violating the ToU, is whether we can connect that account to a specific paid editing company. What is evidence for the "employer", even handled privately? Probably multiple independent lines, like a) the paid editor says so (game-able); b) the paid editing company lists the client on its official website; c) somebody we know is from the company says that they hired paid editing company X; d) a CU shows the account is a SOCK of somebody we know worked for the paid editing company.  ?? Not sure what would allow certainty on this. Jytdog (talk) 05:49, 27 January 2017 (UTC)
Yes 100% certainty is impossible for basically everything besides "death". 98 to 99% certainty is sufficient. Trump proved even taxes are not for sure. Doc James (talk · contribs · email) 06:31, 27 January 2017 (UTC)
Please Doc James I mean reasonable certainty. Would you please name the ways? Jytdog (talk) 06:36, 27 January 2017 (UTC)
IMO it is a number of things rather than one in isolation. Some of these include (1) their edits are very promotional (2) this applies to most rather than just a few edits (3) their clients complain by email to me when the articles they paid to have created are deleted (4) they become "technically excellent" at WP really quickly (5) the refs they use are puff peices (6) they copy and paste from press releases and than say the company in question has given them permission to use the details (7) the topics they write about are non notable and they continue over and over (8) a bunch of brand new accounts show up to "defend" their work from deletion (9) they get professional quality images and typically claim them as their own (10) they upload multiple images from many people when their is no way they could have been in all these places to take the pictures themselves. These are some. Doc James (talk · contribs · email) 07:05, 27 January 2017 (UTC)
Take this photo for example File:Nathan James Sykes.jpg, it is amazingly professional. Perfect lighting. Done during a photo shoot. Do I think the up-loader is the actual author? No, but it is not yet enough for me to state unequivocally that this person is paid. Only at 80% on this one currently. It is enough for me to tag the article with COI and remove the image from EN WP though. Doc James (talk · contribs · email) 07:15, 27 January 2017 (UTC)
Okay/wow. I have not been involved in a lot of this and digesting it now. Cas Liber (talk · contribs) 11:39, 27 January 2017 (UTC)
@Doc James: I see there are other issues on the Nathan Sikes article, but please note that uploading a flattering photograph of oneself or even, hypothetically, of a paying client does not skew the content of Wikipedia and cannot reasonably be considered a conflict of interest or a paid editing concern. We specifically invite BLP subjects who dislike their photos in their articles to "send us a better one" (see, Wikipedia:FAQ/Article subjects#I hate the photo in the article about me). It can be expected that some celebrities may want their photographs to be well-posed and well-lit, as opposed to a candid shot taken in the street. In a given case, if a photograph appears unduly posed and artificial, editors on the article may choose to use a different one, but that does not mean that the article subject or his or her photographer has violated the conflict-of-interest guidelines, much less the terms of use. As it happens, in this particular instance, the editor in question has openly admitted that he or she is editing on behalf of the BLP subject. He or she is may be a bit too assertive in doing so, but edits such as deleting unwanted information about the subject's personal/romantic life and updating the infobox content to emphasize an entertainer's current rather than former activities, can absolutely be dealt with in the ordinary course and without convening a "paid editing investigation" or overturning the anonymity of editors. Newyorkbrad (talk) 17:12, 27 January 2017 (UTC)
User:Newyorkbrad agree, that was my point. There is not enough concern here to require any mechanisms other than the ones we currently have. Doc James (talk · contribs · email) 21:46, 27 January 2017 (UTC)
@Doc James: Actually, as I look more at the specific instance you've described, I don't see much if any evidence of paid editing, and I do see undue escalation against a new editor. In fact, if this is how new editors who may or may not be editing with a COI are treated, I think we have a broader WP:BITE issue to think about. I don't want to get off-topic on this page so I'll follow up on this elsewhere. Newyorkbrad (talk) 23:50, 29 January 2017 (UTC)
Well I am now 99% sure. Anyway helping them navigate releasing material under an open licence. Normal measures work fine for these cases as they are mostly acting in goodish faith. We only need extra measure for long term UPE who are purposely trying to mislead (like Wikipediawriters and LegalMorning, etc). Doc James (talk · contribs · email) 14:11, 30 January 2017 (UTC)
It is a conflict of interest and if that account edits the article it should be "disclosed" to all readers on the talk page using the standard template - it is quite odd that some argue that readers should have concealed from them such coi in articles, when in the real world reliable and ethical publications do disclose potential conflict of interest - now, perhaps the argument is that Wikipedia is neither reliable nor ethical - but that does not mean it should not strive to be. Alanscottwalker (talk) 17:27, 27 January 2017 (UTC)
Doc James, the criteria you identified, are simply the signs of an editor with a COI and I of course agree that those are valid signs. And it would be clear who the client is. I asked a different question - namely, how do you connect the account with an intermediary paid editing company, WikiWriters for example. Jytdog (talk) 17:22, 27 January 2017 (UTC)
While if the client emails you who they hired or a CU connects the account with another account known to be a paid editor. You need to combine that with a high degree of evidence of UPE of course. Doc James (talk · contribs · email) 21:46, 27 January 2017 (UTC)
James, I am focusing on what information would be released. The only legitimate information I could see releasing would be "employer, client, and affiliation". In my view determining employer is very, very difficult. I think your proposal is fatally flawed because you don't specify what off-wiki information would be made public, and people can read that as (for example) the town someone lives in. The lack of specificity around who makes the decision about what information to release will also lead to rejection of this option. The proposal doesn't address community concerns about OUTING adequately on these two key issues. Jytdog (talk) 03:18, 28 January 2017 (UTC)
Yes agree "employer, client, and affiliation" would be all that would be released. Agree it is difficult to determine who the employer is. We probably have 50 ish fair sized Wikipedia editing companies. So not a super-unreasonable burden.
Release of details were sort of in this sentence "These details may include the name of the Wikipedia editing company"Doc James (talk · contribs · email) 04:41, 28 January 2017 (UTC)
While I can see the intent of this, we almost never manage to identify the name of a Wikipeida editing company. This is such an unusual occurrence compared to normal undisclosed paid editing, that I'm not sure that it is worth making a specific exception. - Bilby (talk) 22:05, 28 January 2017 (UTC)
Here are a couple [2] User:Bilby. These are accounts that have made 10s of thousands of paid undisclosed edits and created hundreds of pages. We are talking about having a special prevision to deal with the worst of the worst to prevent the issue from becoming more common. These organizations are purposely breaching Wikipedia rules, they know it, and they continue to edit even though these socks are blocked. Doc James (talk · contribs · email) 03:58, 29 January 2017 (UTC)
How would releasing the name of the company assist in those cases? I'm not overly worried about this provision, I just can't see how it would assist with identifying paid editing. - Bilby (talk) 04:21, 29 January 2017 (UTC)
Well lets say their are 50 companies involved in undisclosed paid editing at a moderate or greater scale, having a list of socks associated with each would help in the building of AI to detect further socks from these companies. It would also help editors learn their patterns.
While we can ban accounts for undisclosed paid editing, to help speed up clean up it is useful to know if they are a sock of a previous account as this allows G5
This would also help people who may be hiring the company to realize that the company is banned from editing Wikipedia before they hire them. Than when people complain to us that they article they bought was deleted we can point them to this list. Doc James (talk · contribs · email) 05:49, 29 January 2017 (UTC)
I agree that recognising that socks belong to the same individual is useful. But I'm still lost on how naming the organisation they work for would be practical, in that it isn't something we would know - if we know anything, it will be that the account is a sock of a blocked/banned user, and we'll block them on that basis. Under what circumstances would we recognise that a sock account belongs to an organisation without also identifying the prior account? In regards to naming, can't we name banned companies under existing policy? - Bilby (talk) 07:43, 29 January 2017 (UTC)
People frequently report who they have paid to write their articles.
How do we create a list of banned companies if we do not know which companies are associated with what banned account? Doc James (talk · contribs · email) 09:18, 29 January 2017 (UTC)
How often do people report who they paid? Where do these reports occur? I've rarely seen that occur, but I may well be wrong, and that might be a decent case for blocking an account on the assumption that they belong to the reported company, although it does lend itself to being gamed. - Bilby (talk) 09:33, 29 January 2017 (UTC)
I have received a few well working on clean up of copyright issues. These are judgement that one does not make on a single piece of evidence. Basically one company worked by writing the articles well they got the subjects themselves to upload the images. The subjects; however, do not know copyright issues well (they think just because they paid a place to take photos of them that means that they own the copyright and are the author of said photos).
During discussions to solve that issue they mention who they paid which was an already blocked account. Ocne you have that confirmation the others can be connected by CU if one can find a CU willing to run these sorts of cases. They become easier to connect though as you have known accounts to connect them with. For example lets say you have a "new account" and you delete some stuff and receive an email from the client saying they paid company X. As CU will not accept off WP evidence you can find what other accounts X used in the past and look to build a case based on behavior evidence for SPI. Doc James (talk · contribs · email) 09:47, 29 January 2017 (UTC)
I'm looking for a reason to go further than option 3 (I'm not inclined to go with "milieu" which seems a very odd choice of words). At this stage, I don't see it. I agree that we want to tackle the paid editing companies, if they act outside of the ToU, but it isn't clear from what I've seen that we need anything stronger than the previous option. I've only seen one case where company can be identified independent of the editor, so investigating the editor as a possible sock seems sufficient in almost all situations, as per the cases you describe. But thank you for helping to explain the situation to me, as it helps a lot. - Bilby (talk) 12:35, 29 January 2017 (UTC)
@Doc James: having a list of socks associated with each would help in the building of AI to detect further socks from these companies Is this something that has been discussed in any detail? I would be curious to see how this would be implemented. GorillaWarfare (talk) 16:36, 29 January 2017 (UTC)
Yes I have had discussions with those who build AI tools and this is something that is being worked on. Would also be useful for addressing other long term socks. Will publish something in the Signpost on this eventually. Doc James (talk · contribs · email) 22:51, 29 January 2017 (UTC)
I look forward to reading it. GorillaWarfare (talk) 23:11, 29 January 2017 (UTC)
What I don't see is how an AI that is presumably looking at behavioral issues would identify different editors working for the same company, as opposed to being limited (as the case would be now) to identifying socks of a common editor. But I'm interested in this as well. - Bilby (talk) 23:54, 29 January 2017 (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.

Milieu 5

MILIEU 5 CLOSED:

Opposed by consensus; 16–7 against. (non-admin closure) --George Ho (talk) 00:23, 3 April 2017 (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.

No one wants to see Wikipedia overrun by promotional paid editing, much less concealed paid editing. However, combating it should not come at the expense of abandoning our longstanding and widely supported policy that editors may choose how much of their personal identifying information they wish to disclose.

Most concerns about paid editing can be addressed by focusing primarily on on-wiki article content and editor behavior. For example, if an article fails to establish the notability of its subject, it can be deleted. If an article is edited in an unduly promotional way, the content can be reverted or modified, and the editor counseled on proper editing practices. If an editor repeatedly posts inappropriate content and persists after being warned, he or she can be blocked. If an editor appears to be posting promotional content under multiple accounts, an SSI investigation can be commenced and a checkuser run, if warranted. All of these actions (other than checkuser, which of course is confidential) can be taken without focusing on editors' private identifying information, much less disclosing it on-wiki.

The most serious instances of "undisclosed paid editing" have involved entities advertising paid editing services as part of a large-scale, for-profit business model. In egregious cases, it may be necessary to coordinate action to identify these entities and put a halt to their activities, especially where dealing with their editing is consuming inordinate amounts of editors' and administrators' time or where false and misleading content is being generated. It may be appropriate to develop an updated protocol and procedures for investigating and dealing with these most serious violations of the terms of use. However, it will be difficult to make progress in this arena unless it is recognized that such special procedures are indeed reserved for the most serious situations only and not for everyday use.

Support milieu 5

  1. Proposed for consideration. Newyorkbrad (talk) 17:47, 27 January 2017 (UTC)
  2. I broadly support the final part of this and firmly support the rest. I think it is important to note that even in the very exceptional cases, that editor privacy is important and only the absolute minimum disclosure of private information necessary is made and only when linkings are agreed by multiple people to be certain or extremely likely, it is not a "joe job" or similar, and that doing so is (not just may) required to the protect the project. Thryduulf (talk) 18:20, 27 January 2017 (UTC)
  3. --Kyohyi (talk) 19:07, 27 January 2017 (UTC)
  4. kelapstick(bainuu) 00:49, 28 January 2017 (UTC)
  5. support this in part, and oppose in part. This proposal is the one that is most in tune with what happens every day. Most cases of advocacy editing are handled, on the ground, with the standard content policies, shifting over into addressing the editor if and only if the editor establishes a pattern of DISRUPTIVE behavior (consistently violating content policies including PROMO, edit warring, etc.) Problematic in the last paragraph, though. We already handle things like Wiki-PR through SPI. The last paragraph does make a nod to the steadily expressed desire for a better way to manage OUTING concerns to allow the community to better address ToU violations that are not so drastic but still not trivial one-offs. So it falls down there. Jytdog (talk) 03:24, 28 January 2017 (UTC)
  6. Support Although it might not go quite far enough. It looks like most of the opposes are not against it per se, but due to "not far enough". North8000 (talk) 19:21, 18 February 2017 (UTC)
  7. Support but only as a starting point. Again, the details will matter. DGG ( talk ) 17:45, 27 February 2017 (UTC)

Oppose milieu 5

  1. The first part is just an old saw that just never cut wood. "We'll crack down on advertising, non-notable companies, POV pushing, etc. etc." it just hasn't worked before. Exert yourselves and see if you can make it work, but don't expect anybody to be satisfied at this late date that this will stop UPE . As stated below by Stazek Lem undisclosed paid editors are not Wikipedians. They are specifically prohibited from editing Wikipedia by the Terms of Use, which trump all policies and guidelines. We don't need to protect the "personal" information that they are required to disclose. Moreover that information is not an individual's name, address, phone number, etc. It is publicly disclosed information such as the paid-editor's on-Wiki employer. We don't need the personal information, and if it pops up - just keep that private. But we do need to link to the paid editing job sites. Smallbones(smalltalk) 18:52, 27 January 2017 (UTC)
  2. No, an editor does not get to "choose" how much info they disclose. If you are a paid editor you are required to disclose certain things. Agree that most cases can be dealt with using on WP evidence. We are discussing what to do about egregious cases, which are frequent, and which we are failing on. Doc James (talk · contribs · email) 21:50, 27 January 2017 (UTC)
  3. Oppose, per Doc James and Smallbones above. Paid editors are a special category of editors, and should not be permitted to hide behind the rules that govern editors who are here in good faith, and not here in the service of employers. We have an entire apparatus and a culture that protects, shields and generally makes excuses for paid editors. That's the problem, as is the namby-pamby mindset reflected by this "milieu." Why are we calling these "milieus" anyway? Coretheapple (talk) 23:02, 27 January 2017 (UTC)
  4. I don't entirely oppose this, and I appreciate the effort to distinguish between mild COI where privacy should be protected, versus extreme COI where privacy stops being the primary consideration. That aspect, I agree with. But I feel that the balance of the statement overall, is too cautious about recognizing the threat posed by the more extreme scenarios. --Tryptofish (talk) 00:03, 28 January 2017 (UTC)
  5. I mostly support this as a good expression of the status quo, but that is all it does. This doesn't provide any new paths forward. The last paragraph does make a nod to the steadily expressed desire for a better way to manage OUTING concerns, to allow the community to better address ToU violations that are not so drastic but still not trivial one-offs. So it falls down there. It also doesn't classify what personal information is relevant to COI/PAID concerns. Needs to be limited to "employer, client, and affiliation" and no more than that. Jytdog (talk) 03:26, 28 January 2017 (UTC)
  6. I oppose this because it misses the really critical thing about paid for article information, readers deserve the honesty of being able to know who paid for the article edits, they are reading and relying on. Alanscottwalker (talk) 12:36, 28 January 2017 (UTC)
  7. Strong oppose This entirely misses the fact that identifying serious cases relies on looking into more "everyday cases". Nearly all of our major cases of COI or sock editing have started as minor cases that have later bloomed into massive cases involving many accounts. Carl Fredrik 💌 📧 13:26, 28 January 2017 (UTC)
  8. Oppose per Alanscottwalker. Figureofnine (talk • contribs) 22:52, 28 January 2017 (UTC)
  9. Only because it's poorly-written. Chris Troutman (talk) 22:53, 28 January 2017 (UTC)
  10. oppose --Ozzie10aaaa (talk) 23:53, 28 January 2017 (UTC)
  11. Oppose To the extent I understand what this is trying to say, it reduces mostly to a statement of current practice. That's not a terrible thing in itself, but it offers little to help us move ahead on this issue. The most substantive part of this proposal is the last paragraph, the best response to which is the comment by Carl Fredrik above that "nearly all of our major cases of COI or sock editing have started as minor cases that have later bloomed into massive cases involving many accounts." Shock Brigade Harvester Boris (talk) 04:06, 29 January 2017 (UTC)
  12. Oppose per Carl Fredrik. BethNaught (talk) 09:37, 29 January 2017 (UTC)
  13. Per Doc and CF. Only in death does duty end (talk) 09:44, 30 January 2017 (UTC)
  14. People should be following the terms of use, so we don't need to add "you don't really have to do this bit of the TOU" to our policy. Graeme Bartlett (talk) 02:40, 31 January 2017 (UTC)
  15. Oppose Quite simply, inadequate to the situation. Any approach that depends only at the article level or only at the contributor level will give inadequate results. It's just like sockpupettry, where the most secure results are one where the checkuser data and the manner of editing complement each other. Having detected a probably UPE article, the necessary next step is to see what other articles and editors are involved. If we go only by the articles, we'll have to go one at a time--and the reality of things at WP is that by the chance of who comments at AfD we can not count on article review processes to stop all UPE articles.. But if we can go after the editors we can stop many articles. Any one approach can be defeated by any sufficient clever UPE, but by using both, we have a chance. Most of the rings were discovered by following up chance findings. DGG ( talk ) 06:00, 31 January 2017 (UTC)
  16. Feeling kinda guilty opposing it for the most part - Honestly, I wish this proposal could work. But I don't necessarily see how it would deal with the worst cases effectively, with the provision proposed. If in the complexities of the internet there were some way to create a cloud which could be created and destroyed as required to present to qualified individuals given a certain user right to see it for an ANI or similar discussion on a proposed ban of some sort, with the information removed afterwards, that might work. I doubt it is necessarily possible though, and the information in the cloud could perhaps be copied and reused later by unscrupulous persons, maybe. John Carter (talk) 16:57, 6 February 2017 (UTC)

Discussion of milieu 5

  1. A person in an egregious violation of our fundamental rules forfeits his rights as a wikipedian. The right for anonymity within wikipedia is a right of a wikipedian, right? Why not treat it as such? Staszek Lem (talk) 18:31, 27 January 2017 (UTC)
    There are several possible responses to that. Perhaps the most obvious is that that logic could apply equally, for example, to a repeat vandal, yet we don't combat vandalism by tracking down vandals and publicizing their identities. Newyorkbrad (talk) 18:34, 27 January 2017 (UTC)
    Are you sure you want to take this position? Harassing and Abusing Others, Violating the Privacy of Others, Engaging in False Statements, Impersonation, or Fraud, Committing Infringement, Misusing Our Services for Other Illegal Purposes, Engaging in Disruptive and Illegal Misuse of Facilities are all things that are prohibited by the TOU. Would engaging in these prohibited actions also forfeit their rights as a wikipedian? --Kyohyi (talk) 19:11, 27 January 2017 (UTC)
I don't see how that is related. Outing is actually not part of the TOU, only a policy on Wikipedia. That does not make it unimportant, but makes it subject to the terms and conditions we as a community impose, not what the WMF imposes. Carl Fredrik 💌 📧 13:36, 28 January 2017 (UTC)
@CFCF: See wmf:Terms of Use#4. Refraining from Certain Activities. "Harassing and Abusing Others" and "Violating the Privacy of Others" are both prohibited activities. Thryduulf (talk) 01:08, 29 January 2017 (UTC)
Thryduulf — the privacy policy and the outing policy are absolutely related, but not synonymous. What I objected to above was the comparison of investigating paid editing and engaging in false statements, impersonation or fraud, etc. — those do not seem related to me. Carl Fredrik 💌 📧 01:15, 29 January 2017 (UTC)
  • It's not true the "no one" supports paid editing mills. There's a wide spectrum of opinion on wikipedia that opposes the very concept of COI, and is indifferent or supportive of paid editing. We have an "articles for creation" mechanism that is exploited by paid and COI editors. Milieu 5 is both mealy-mouthed and not terribly honest, to be blunt about it. Coretheapple (talk) 19:06, 27 January 2017 (UTC)
  • User:Newyorkbrad in my view this describes the current situation accurately and I appreciate you posting this. There will be some disagreement about what constitutes a "big enough" problem, and it doesn't really move ball forward about what the new procedures would be. But trying to establish consensus on the background is important.
The one place where this falls a bit short is clarity on what "personal information" is relevant. I don't think anybody who wants more off-wiki information to be used, wants to post people's names and home addresses. I understand that those opposed to any use of off-wiki information on-wiki, is the risk that it can be. In my view, the relevant information for the community, is what ToU calls for: employer, client, and affiliation. These pieces of information are (in my view) important to the community.
Linking a user account to any one of those three, can perhaps be considered a) (and this is key) not itself "personal information" like RW name and home address; b) may not be linkable at all to any actual personal information, depending on how well someone has managed their online identity; c) and more controversially - to the extent that one of those three pieces of information can be connected further to someone's personal identity, this ~might not~ be our concern nor our problem. One of the things that has bedeviled these discussions is the lack of classification of kinds of information that exist, whether any distinction can be made at all and if it should be, and if meaningful distinctions can be made, how to handle them. This is (in my view) one of the really foundational disagreements here. It even brings in issues of legal risk, as I understand it.
Thoughts? Jytdog (talk) 19:47, 27 January 2017 (UTC)
The problem with this proposal like many is that it includes many things. I disagree with the first part that undisclosed paid editors have a right not to disclose who is paying them. I agree that most concerns can be dealt with with on WP evidence and they are. I also agree that special procedures are only required for egregious cases. Doc James (talk · contribs · email) 21:55, 27 January 2017 (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.

Concrete proposal 1

There is a strong consensus to insert the proposed statement at the desired place.Winged Blades Godric 11:59, 18 April 2017 (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.

The following shall be inserted at the end of Wikipedia:Harassment#Posting of personal_information

"There are job posting sites where employers publicly post advertisements to recruit paid Wikipedia editors. Linking to such an ad in a forum such as WP:Conflict of interest noticeboard is not a violation of this policy."


Support concrete proposal 1

  1. As proposer. There's no personal information posted here unless the poster actively chooses to post it publicly. The paid editors are prohibited from editing Wikipedia unless they disclose the employer, client, and other relevant affiliations onWiki, so if they disclose this information in response to the ad, it's nothing that isn't already required. So far there have only been very vague proposals - this simple rule would resolve at least half of the vague problems discussed here. I've posted a notice at WT:Harassment because it directly effects that policy. Smallbones(smalltalk) 15:28, 29 January 2017 (UTC)
  2. Clear and direct. Figureofnine (talk • contribs) 15:42, 29 January 2017 (UTC)
  3. Strong support, with the understanding that WP:Outing also needs a lot more revision. I might also want to see this sentence modified to specify that it is about editing that is contrary to policy – I'm not sure that I would want to apply this to some hypothetical site where editors are hired to edit with on-wiki disclosure and in an NPOV manner. But yes, this is exactly what we need to make possible, even if the linked site contains personal information. By the time someone gets to this level of deception, worrying about their personal privacy is a fool's errand. --Tryptofish (talk) 21:30, 29 January 2017 (UTC)
    Seeing some objections raised by other editors, I want to add something. I realize that these sites can contain personal information, as personal information is described in the outing policy. But given the nature of these sites, as extremely far from benign, I still support linking to them. In other words, I am explicitly supporting an exception to the outing policy, which is what I understand this proposal to be. Protection of personal information is not a suicide pact. --Tryptofish (talk) 21:21, 5 February 2017 (UTC)
  4. Strong support unambiguous and definitely a good idea. Doc James (talk · contribs · email) 23:12, 29 January 2017 (UTC)
  5. Doesnt violate the policy now, however if it needs to be stated explicitly. Only in death does duty end (talk) 09:46, 30 January 2017 (UTC)
  6. Support If someone is advertising paid editing then it's utterly perverse to consider it harassment or outing to link to the advertisements. Alsee (talk) 12:45, 30 January 2017 (UTC)
  7. Support, even if an editor's PII is included in the advertisement. My only worry is that this will drive the advertisements underground. We need to stamp out unethical paid editing, and this is a good compromise between paid editing and outing. I have no problem with a strong recommendation that the version linked to should be one with as much PII redacted as possible, if this is technically practical to implement.Tazerdadog (talk) 22:24, 30 January 2017 (UTC)
  8. Support this is completely OK to draw attention to advertisements directly affecting Wikipedia. This also permission also includes ads that reveal real name, occupation or companies. If these are advertised, linking from Wikipedia is not any outing violation, as the information is already out there. Graeme Bartlett (talk) 02:41, 31 January 2017 (UTC)
  9. Support This is certainly a first step. Clever UPE of course advertise in such a way that the posting is of no use to us, but a surprising percentage of them are not that clever-- or are so contemptuous of us that they don't even think it matters. DGG ( talk ) 05:48, 31 January 2017 (UTC)
  10. Support but it can be broadened to include any public website. Coretheapple (talk) 20:48, 31 January 2017 (UTC)
  11. Support – Might as well make it explicit as a first small step in tackling this contentious issue. Mojoworker (talk) 20:58, 31 January 2017 (UTC)
  12. Support Good first step. It targets the request and can help identify articles where paid editors are likely to show up in the near term. JbhTalk 23:15, 31 January 2017 (UTC)
  13. Support merely linking to a publicly viewable "editor wanted" ad in good faith is not an invasion of privacy, because anyone posting such an add has no expectation of privacy. This can be done abusively or in bad faith, just like everything else, and that's where admin discretion comes into play. Geogene (talk) 00:25, 2 February 2017 (UTC)
  14. Support. I disagree with the opposers per Geogene, though discretion is advised. MER-C 06:56, 2 February 2017 (UTC)
  15. Support since undisclosed paid editing is a no-no even if the content is beautiful, content alone cannot be the deciding factor. If paid editing were a norm, WP would cease to be what it is and turn into a PR wire with some celebrity gossip here and there. Carlossuarez46 (talk) 02:48, 3 February 2017 (UTC)
  16. Support Clearly a good idea. BU Rob13 makes a good point, however. ThePlatypusofDoom (talk) 15:08, 3 February 2017 (UTC)
    @ThePlatypusofDoom: I have to ask, why support then? Outing is something that, when done wrong, causes irreparable harm to an editor. If we accidentally out someone who isn't really a paid editor, we open up that person to harassment in real-life, stalking, etc. Even if we out the "correct" person, these are risks. These are real problems that experienced editors face on Wikipedia. Can we afford to acknowledge the issues with this language, approve this anyway, and then wait until someone's life is ruined to fix the problems? These proposals have already cost the site at least one administrator. It would be a damn shame if they cost us more. Policy writing is hard, but we've got to get it right when the safety and livelihoods of real living people are on the line. There are few issues on Wikipedia that are actually that important, but this is one of them. ~ Rob13Talk 04:28, 5 February 2017 (UTC)
    @BU Rob13: This is the closest to what I think is acceptable. I'm considering making a new proposal, and not supporting this one. ThePlatypusofDoom (talk) 14:19, 5 February 2017 (UTC)
    Obviously, your vote is yours, but given the significance of outing an editor, I don't think it's unreasonable to demand a well-written proposal. It's perfectly acceptable to oppose everything, go back to the drawing board, and find something better. ~ Rob13Talk 14:22, 5 February 2017 (UTC)
    User:BU Rob13 I am curious regarding what wording or measures you would support? Doc James (talk · contribs · email) 07:25, 15 February 2017 (UTC)
  17. support per Smallbones--Ozzie10aaaa (talk) 02:18, 5 February 2017 (UTC)
  18. Support—as a step forward, but not sufficient. Implementation has got to start somewhere. Harassment is a real issue, but more harm to individuals by on-wiki treatment occurs than can be expected from an effective anti-undisclosed-COI-paid-editing process. There is an irreducible risk associated with editing Wikipedia. We have some effective tools for reducing outing, but can't eliminate it. The process we need to reduce undisclosed-paid-COI-editing can reduce the risk of outing by handling all information other than user name, client, and wiki-editing service group privately. We can not eliminate undisclosed-paid-COI editing, but it must be reduced from existential proportions. Balancing the two risks is the first task of the process. — User:Neonorange (Phil) — 06:25, 5 February 2017 (UTC)
  19. Support, basically per Phil. Alanscottwalker (talk) 14:27, 5 February 2017 (UTC)
  20. Support. I realize this doesn't protect against "joe jobs" but Wikipedia is full of false accusations about policy violations, and this will be by no means the worst of them. Wnt (talk) 14:55, 5 February 2017 (UTC)
  21. Support although there is the question as to whether such a site is necessarily related to the edits in question even with such a link. John Carter (talk) 16:58, 6 February 2017 (UTC)
  22. Strong support — and indicate this overrides WP:OUTING. Carl Fredrik 💌 📧 19:27, 6 February 2017 (UTC)
  23. Support. I have enough trouble dealing with true believers. Paid editors gaming the system is asking for more trouble. Stop the madness. QuackGuru (talk) 03:18, 10 February 2017 (UTC)
  24. Strongly support. WP is not equipped to deal with UPE on merits alone. Paid editors are incentivized to create content that appears to meet NPOV criteria while still being promotional. This is the smallest of many steps that need to be taken. James (talk/contribs) 22:56, 14 February 2017 (UTC)
  25. Strong support per Smallbones and a worthy idea indeed. GSS (talk|c|em) 17:21, 18 February 2017 (UTC)
  26. Support I can't see how simply linking to a public site per se could be a problem. North8000 (talk) 19:25, 18 February 2017 (UTC)
  27. Support This does not come near solving all our problems, but it's a reasonable move forward. DGG ( talk ) 17:09, 27 February 2017 (UTC)
  28. Support. This should be self-evident, but a clear statement of it does no harm. Justlettersandnumbers (talk) 20:32, 2 March 2017 (UTC)

Oppose concrete proposal 1

  1. Oppose marginally, but this is close to being fine. When an article subject (company or person) is advertising off-site for paid edits to be made, I'd be fine with that being posted on-wiki. We need to clarify that we should not post ads posted by a paid editor if they reveal real name, occupation, etc. however. That's just plain outing. ~ Rob13Talk 03:17, 30 January 2017 (UTC)
  2. Oppose as written. This would negate portions of the first paragraph and fourth paragraphs, undermining the effectiveness of WP:OUTING. Generally I agree with BU Rob13's comments. - MrX 12:41, 30 January 2017 (UTC)
  3. Oppose I think this needs to be more specific as to how the posting can be made. If it amounts to outing, it should exclusively go the private ArbCom/WMF route and be suppressed elsewhere. Samsara 23:53, 1 February 2017 (UTC)
  4. Oppose. Does not protect people from joe jobs, etc. Does not restrict the amount of information that can be posted. Does not distinguish between those who will comply with the TOU and those who will not. Thryduulf (talk) 12:05, 5 February 2017 (UTC)
  5. Oppose Those job ads contain extensive personal information about the people who apply for them. While it makes sense to highlight the employer, what this does is out - in some detail - the people who then apply for the job. At best, this is outing a paid editor who failed to meet the ToU, but at worst, this will be outing people who have done nothing wrong. We haven't accepted that risk before, and accepting it to catch a few (maybe a dozen or so) paid editors is not worth the risk. - Bilby (talk) 21:11, 5 February 2017 (UTC)
  6. Oppose per Samsara and Bilby. Gamaliel (talk) 03:56, 6 February 2017 (UTC)

Discussion of concrete proposal 1

@Tryptofish: I'd also want to revise WP:Outing, just to clear up the bad writing and poor organization. There are probably some self-contradictions in there too. Perhaps just ask two good writers to each clean the whole thing up, retaining the current meaning (with the above added of course) and then have an RfC to decide which is better. As far as "outing" somebody replying to a site where they are going to pay for NPOV text and proper ToU disclosure. Well, if there is proper disclosure there'd be no reason to link to it Smallbones(smalltalk) 22:48, 29 January 2017 (UTC)

Sounds good, thanks. --Tryptofish (talk) 00:48, 31 January 2017 (UTC)
Some feel that all the paid editor would need to do is add a fake name and personal details and then the ad would be "safe". Would need to be set up so that it cannot be gamed. Doc James (talk · contribs · email) 03:58, 30 January 2017 (UTC)
@BU Rob13 and Doc James: I don't think individual's real names are usually posted, and I don't see a reason why we would want to mention them at all on Wiki. I wouldn't be totally against not linking if there was a real world name in the ad, as long as the ad could be mentioned onWiki. Presumably, however, there will be a group set up for a private hearing/investigation (as discussed below). Hopefully there will be rules set up for what kind of information can or should be submitted privately, and what shouldn't be posted, e.g. on WP:COIN, before a private hearing/investigation. I'd like to leave some leeway for how it's set up. Smallbones(smalltalk) 06:23, 30 January 2017 (UTC)
@Smallbones: The problem is that the damage caused by even one abuse of the rules set down here is potentially enormous. "I don't see a reason to do X" isn't sufficient. We need "Do not do X" in the policy if we're going to weaken it to allow some limited outing. We need to very carefully tailor what is released on-wiki to be beyond potential abuse. If we throw this to a free-for-all where all editors can post any information on-wiki with "use your judgement" as the guideline so long as they yell "paid editing!" first, that's obviously ripe for abuse. You would exercise good judgement and limit the information posted to what's critically necessary, but what about an editor with less-than-pure intentions whose goal is to drive someone off the project? Growing pains are not acceptable for something this serious. If we set up a group to confidentially handle paid editing investigations involving any information not posted on-wiki, then what's the actual benefit of posting anything on-wiki beyond "It's paid editing" or "It's not paid editing"? I've yet to see this case. Is it just transparency? Because we have many other processes that aren't transparent to protect editors (even the "bad" ones) from harassment and outing. As far as I'm concerned, how we handle private information is an existential question. Given certain realities of my own personal situation, I simply cannot edit on the English Wikipedia if the policies on outing and harassment are less than perfectly robust. I assume that's true for many others as well. ~ Rob13Talk 07:16, 30 January 2017 (UTC)
Not true for me. In fact the severe interpretations that some take of the "outing policies" make me feel less safe rather than more safe, as they protect those who harass others. Doc James (talk · contribs · email) 14:21, 30 January 2017 (UTC)
I concur with Doc James here. IMHO based on experience, paid editors are among the worst harassers on WikipediaSmallbones(smalltalk) 15:01, 1 February 2017 (UTC)
@Casliber: Could you close this, or find somebody to close it. I think that the result is clear now, but wouldn't be against waiting a few days; or even asking for more broad based input, if it was truly broad based. Any help appreciated. Smallbones(smalltalk) 15:01, 1 February 2017 (UTC)
@Smallbones: what is the result? There seem to be about five milieus (whatever that means) and one concrete proposal here. I've been trying to make sense of it all, in order to cast some votes, but haven't so far been able to do so. Is the proposal just to rubberstamp the "concrete proposal 1" or to do more than that?  — Amakuru (talk) 15:11, 1 February 2017 (UTC)
Just for Concrete proposal 1. My experience is that, after 3-4 days, a 12-2 !vote usually stands up after waiting a few more days, unless more broad based input is sought. So it's a choice between doing the obvious, waiting for the inevitable, or soliciting more input. I'd prefer Casliber solicit the input if he thinks that that's the way to go. Smallbones(smalltalk) 17:29, 1 February 2017 (UTC)

Would this replace the case-by-case sentence in WP:OUTING? Or would we expect to see this exercised in addition to that sentence allowing for that job posting site and other personal information? Mkdw talk 00:10, 4 February 2017 (UTC)

@Smallbones: - my concern here is tat linking to the ad will also entail linking to the person hired, and they are likely to post their real names. - Bilby (talk) 21:13, 5 February 2017 (UTC)
Not replace but supplement. This is an specific example of one case. Doc James (talk · contribs · email) 06:35, 4 February 2017 (UTC)
  • @Doc James: Question for you. One of the many sockmasters I've annoyed decides to get back at me by posting an ad for paid editing on a site hosting such advertisements including my personal information as if I'm the paid editor. Under your proposal, what happens? ~ Rob13Talk 04:29, 5 February 2017 (UTC)
    • If you are doing paid editing, somebody will probably notice that, look for some confirmation, notice the ad and say "It looks like he's a paid editor."
    • If you are not doing anything that looks like paid editing, nobody will investigate you for paid editing, and nothing will happen. BTW if your enemies have your personal info, they wouldn't need to go through this rigamorol to out you. Smallbones(smalltalk) 05:46, 5 February 2017 (UTC)
With respect to outing they could just use socks here and post you personal information in a bunch of other places on the Internet were the outing policy does not apply. Would be a lot simpler and cheaper than doing what you describe.
But basically, let say the job post is for the creation of an article on "company X". Job posting is awarded to an account with your name. Next a check would occur to see if you created said article about said company after the job was awarded. Would be very very unlikely for that to happen unless someone had access to your account. Doc James (talk · contribs · email) 08:16, 5 February 2017 (UTC)

I'm wondering if it has been made clear what linking to job ads would reveal. In most cases, there is little information about the employer - the employer typically doesn't have to reveal the company's name, location, or contact details, as those are managed by the job site. In about 20-30% of cases (depending on the site), there is enough to identify the employer, but this is due to either previous advertisements where the employer has identified themselves either deliberately or as a side effect of providing sample materials, or they have stated in the job ad who they are. However, the editor who is hired for the job can typically reveal (depending on the site):

  • Real name
  • Location (generally - not a specific address)
  • Photo
  • Education history
  • Employment history (away from the job site)
  • Employment history (on the job site)

This isn't always the case, but that is a lot for us to be linking to. We're talking about linking to a lot more than just the job, as we will be providing access to a high level of detail about the editor. If we want to take this path then ok, but revealing personal information was not supported in Milieu 2, and in the past the community has been very reluctant about linking to this level of personal information. - Bilby (talk) 23:41, 5 February 2017 (UTC)

@Bilby: As written, this proposal covers only postings where employers are doing the recruiting, so the types of advertisements you're describing would not be affected. ~ Rob13Talk 18:16, 6 February 2017 (UTC)
@BU Rob13: That's not the case. It would be if we were targeting Craigslist, but that isn't what this is about. The proposal is targeting sites such as Upwork and the freelancer sites. Employers place the ad, but you need to be registered to apply for the job. That registration is public, and contains extensive personal information. Far more than what we would normally allow to be linked to. - Bilby (talk) 21:51, 6 February 2017 (UTC)
@Bilby: The proposal states "There are job posting sites where employers publicly post advertisements to recruit paid Wikipedia editors. Linking to such an ad...". I would interpret that to mean that linking to the ad itself is fine, but any information about the person who answers the ad is not. Are you saying the person that answers the ad is publicly disclosed on the ad itself? ~ Rob13Talk 03:07, 7 February 2017 (UTC)
Yes, if they are hired, the ad then contains their name and a link to their full personal details. Some also contain links to people who have applied, but not Upwork, which now hides applications. - Bilby (talk) 03:34, 7 February 2017 (UTC)
  • I've been thinking hard about the kinds of issues that Bilby raises here, about how this would be a substantive change to the current consensus about outing, and here are my personal thoughts. I agree that there is a solid consensus in the community that the posting of personal information of good-faith editors is a truly horrible thing that must not be allowed. And I agree that it applies just as much to the dumb kid who posts vandalism, the editor whose personal opinions are repellent, the editor who is constantly tendentious, and the editor who does a small bit of undisclosed paid editing as a bit of low-level spam, out of a lack of sophistication about how Wikipedia works. Those kinds of users are entitled to privacy too, and their personal information must not be outed. But here, we are entering into a whole new category of disruptive users, and sites like Upwork are the least of it. There are deep-pocketed interests that will want to use us to distort reality, and now a US government that is apparently dedicated to "alternate facts". We are going to be facing large, dedicated, and well-supported groups of editors who are deliberately committed to subverting Wikipedia and to utilizing clever methods to avoid detection. I am willing to allow outing of those users. I'm not saying that as a sort of punishment, as a perception that they "deserve it", but rather as something where I just don't care about their well-being that much, whereas I think it is an existential threat to Wikipedia (no, I am not being dramatic or exaggerating) in which we need to be able to use more tools to enforce community norms. --Tryptofish (talk) 22:20, 6 February 2017 (UTC)
I don't feel that the freelancers are our major problem - the biggest problem, as you describe it, comes mostly from the professional PR groups editing Wikipedia, or other vested interests who can sustain a long campaign. But with that said, if the community wants to take this direction so be it. :) I just don't want to take this by stealth, which while not the intent, is the risk attached to this proposal. - Bilby (talk) 23:34, 6 February 2017 (UTC)

@Smallbones: "There are job posting sites where employers publicly post advertisements to recruit paid Wikipedia editors." Do we have a page that lists an at least semi-comprehensive list of the homepages of such sites? This proposal, while touted as "concrete", seems somewhat abstract to me, as I'm not familiar with the specific content of these sites. Would be nice to be able to review such sites before taking a position on the proposal. wbm1058 (talk) 14:08, 11 February 2017 (UTC)

@Wbm1058:. Thanks for the question. While I could post some specific links here and answer your question directly, I don't think that I can according to the interpretation of the current rules by some arbcom members. Kinda puts me on the spot, doesn't it? But that's the situation now that anybody faces if they see that a job is offered on a public posting board that seems to refer to a specific type of business. Note that I don't want to mention any company's name here, or any editor's user name, real name, place of employment, etc. What I do want to do is link to a publicly viewable site that might link to another page that somebody has publicly posted their own information that is usually very nonspecific. These folks, if they have already done paid editing on Wikipedia, have either declared similar info already on-Wiki, or are prohibited from editing Wikipedia as undeclared paid editors. I think the best I can do is give general directions on how to find one of these sites. Do a search for something like "Wikipedia freelancing jobs offered" (I haven't done this specific search - maybe I'm being paranoid about the current rules?)
Smallbones(smalltalk) 15:05, 11 February 2017 (UTC)
Smallbones, I took you up on your search suggestion, and Upwork (upwork.com) was the first one I noticed at the top of the list. It's not immediately apparent to me how one would apply for one of these jobs, but I don't see any publicly posted responses to these offers. Are there sites where responses are publicly posted, and should we make any sort of distinction between sites where responses are posted and those where they're not? Someone could be innocently responding in the belief that these offers are legitimate and not against any Wikipedia policy that they're aware of. I also note that most of these don't name a specific company; I don't see how linking to an ad that doesn't name anything specific would be helpful. Though the latest ad at the moment does mention Inbox Pros. It's not hard to locate their homepage. I assume your intention is to "name and shame" companies like this one (they're in the businesses of email deliverability and SEO optimization). We could watchlist for the creation of Inbox Pros. Perhaps a bot could crawl these sites and automatically generate such watchlists. Though I see reference to "joe jobs" in the discussion above. How can we be sure that this company actually placed the ad, and the ad is not a "joe job"? Notable e-lancing websites include Fiverr, Freelancer.com, Guru.com, Upwork, and 99designs. See also Category:Freelance marketplace websites. wbm1058 (talk) 16:34, 11 February 2017 (UTC)
Oh, I don't think just showing up on one of these sites and seeing company names on 4 or 5 of the 30+ job listed is going to do a whole lot by itself, but it is a tool that combined with other (mostly on-Wiki) data is going to help a lot. For example, if the name of one of the companies shows up as a newly created article, with somebody saying that it looks like extreme COI editing, it looks like we can delete the article and let the editor know that they need to declare. This will effectively warn off the lower levels of companies and editors and let them know that if they want to break our rules they are at least going to have to make a plan to do it right. That might scare them off permanently. Some people are like "since nobody is watching, why not?" but if they would have to actively lie, they will stop. BTW, I can't see how a joe job would work with something like this. Who would pull the joe job and why?
A big thing for more complicated cases would be to follow the boards over time. While catching the simple mistakes as above, a fuller knowledge of how the boards work will develop, e.g. the writing styles of certain employers who use multiple names. All this will just point to articles on Wikipedia, and the articles will confirm by themselves whether paid editing is going on. Now when you ask an obvious paid editor about why his text was so promotional, they can just deny everything and you can't point out the job posting. If they know that we can link to the posting, it can be much simpler. So all-in-all, just a tool. Smallbones(smalltalk) 21:12, 11 February 2017 (UTC)
BTW, naming and shaming is a limited tool that should never be overused. I don't intend to do that at all, but if somebody simply denies the obvious, I don't see any problem in stating the obvious (or at least linking to it on an appropriate page such as WP:COIN. Smallbones(smalltalk) 21:30, 11 February 2017 (UTC)
An additional benefit occurs to me. Editors accused of COI editing may well ask: what evidence do you have for saying this about me? It can be an entirely reasonable question, and making an accusation without evidence can be considered a personal attack. It can also be a clever strategy by an intentionally deceptive editor, if they know that we are unable to link to such a site. In any of these situations, such linking is beneficial. --Tryptofish (talk) 00:56, 13 February 2017 (UTC)
My problem is publicly linking to their personal information. Other proposals didn't require public outing - this one does. Traditionally, the wider community has been very reluctant (at best) to approve anything which results in publicly revealing the personal information of any editors. - Bilby (talk) 01:53, 13 February 2017 (UTC)
And thus, my comment a little ways above. Sometimes, there can be compelling reasons to change from tradition (with consensus, of course). --Tryptofish (talk) 02:04, 13 February 2017 (UTC)
Then convince the community. What seems to be happening here is that we're dropping a core policy, that the community has repeatedly reinforced, on teh basis of a couple of dozen editors in order to (at best) limit the activities of a small number of paid editors. It isn't a great way forward. - Bilby (talk) 02:14, 13 February 2017 (UTC)
I agree with you very much on that. As I see it, the RfC here is going to turn out to be more like an early step, than something that will result directly in action. --Tryptofish (talk) 02:21, 13 February 2017 (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.

Concrete proposal 2

No consensus.There are views that the wording/proposed clarification is not up to the mark.Winged Blades Godric 12:04, 18 April 2017 (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.

Banned accounts can be connected with the Wikipedia editing companies with which they are involved if a number of specific criteria apply. This only applies to banned accounts and not blocked accounts. The accounts may be banned as a result of a CU finding sock-puppetry or through a community process. The companies that are connected can only be those that are involved primarily in paid editing of Wikipedia such as Wikipediawriters and Legalmorning and not companies whose primary business does not relate to editing Wikipedia for pay (such as universities or NGOs).

Evidence for making the connection should be shareable with admins or functionaries / WMF legal and need not be postable on Wikipedia. Such evidence may include a combination of the following along with common sense 1) banned account in which the majority of the edits are promotional (always a must) 2) emails from subjects of articles mentioning the company which they hired 3) urls from websites such as Fivver which mentions the article paid for and show which company was awarded the contract.

Support concrete proposal 2

  • "need not be postable" means that evidence used to justify the linking of the banned WP account to the Wikipedia editing company need not be of the type that can be posted to Wikipedia. If such evidence does exist than of course it can be posted. Not sure if that clarifies your concerns. Best Doc James (talk · contribs · email) 09:14, 6 February 2017 (UTC)

Oppose concrete proposal 2

  • Oppose. If I have understood the sentiment correctly, then it's OK, but after a few readings I'm still not sure that I do understand it. Anything that makes exceptions to the privacy policy MUST be clear and unambiguous. Specifically, I think that the evidence must be presented to a group of people who have been explicitly trusted with this sort of information (currently only checkusers and arbcom), and only after that group reaches a consensus that that (a) the link is accurate, AND (b) that posting the link publicly will benefit Wikipedia, AND (c) will (as far as can be determined) not harm an innocent person, can the information be posted publicly. Thryduulf (talk) 12:11, 5 February 2017 (UTC)
  • Oppose. I'm getting the feeling that people have decided that "milieu" is bad branding and "concrete proposal" is good branding. This proposal twists itself into knots by introducing so many conditions that it stops really being a concrete proposal. I do not in fact oppose posting that kind of information about banned accounts. But this proposal does not go far enough. It implies that we can't post anything about blocked accounts or accused accounts, and I'm not willing to restrict things that narrowly. --Tryptofish (talk) 21:02, 5 February 2017 (UTC)
  • Well, it does use language that specifically distinguishes between bans and blocks, and no matter the intent, I can confidently predict that it would be wikilawyered that way. --Tryptofish (talk) 21:58, 6 February 2017 (UTC)
  • Sorry, no, I understand what you are saying, but I disagree. My reading of the community is that passage of a narrow case will be used as an excuse for not considering the wider case. --Tryptofish (talk) 22:46, 9 February 2017 (UTC)
@Doc James:The short answer is that if you took the spirit of the proposal, and explicitly stated various things that are only implied in it, IMHO it would be a good proposal which I'd support. I'll try to expand on this. North8000 (talk) 13:07, 19 February 2017 (UTC)
Clear wording that this applies (only) to account banned for undisclosed paid editing. Also that it is for flat-out paid editing. ("flat-out"= not considered "paid" if they received some token like a thank-you card.). Next that the evidence is only disclosed to a limited group on a need-to-know basis. The proposal can be read to say that the individual (not just the connection of a user name to the company) may be outed to the entire admin list.North8000 (talk) 13:19, 19 February 2017 (UTC)

Discussion of concrete proposal 2

Two things 1) to help with follow up of future socks 2) to help future clients of these companies realize that they do not act in compliance with WP rules. Doc James (talk · contribs · email) 07:20, 15 February 2017 (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.

Concrete proposal 3

Closed as no consensus. There does not appear to be clear consensus for this proposal. Gamebuster19901 (Talk║Contributions) 06:21, 4 June 2017 (UTC) (non-admin closure)

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.

This is a relisted proposal, which was originally listed at WT:Admin in September, 2015 . See:Wikipedia_talk:Administrators/Archive 15#Proposed change - "No paid editing" for admins. The !vote there, by my count, was 15 supporting, and 13 opposing. I've made some very minor cosmetic changes.

In the Orangemoody sockpuppet/paid editing fiasco, socks were approaching AfC participants and claiming to be admins

Consider the following paragraph in the Guardian:

"The Wikipedia Foundation said the accounts were blocked over “black hat” editing – charging money for the creation of promotional articles – amid allegations that hundreds of businesses and minor celebrities have been blackmailed by scammers posing as Wikipedia administrators."

It would be very good if we could tell everybody - including participants at WP:AFC, all our readers, and any journalists - that "admins are *not* allowed to accept payment for any services on Wikipedia. Do not believe those who claim to be admins and ask for money."

This would be a bright line rule that would protect everybody involved, admins, the WMF, and the scam targets.

I am not accusing any admins of being paid editors, suggesting that admins were responsible for the Orangemoody fiasco, or singling out admins for criticism. This is just a simple way to stop future Orangemoodies.

Proposed text to be inserted at the bottom of the "Involved admins" section at WP:Admins

Payments not allowed

"No administrator may accept payment to edit articles or to perform any administrative function on Wikipedia. Admins who work for the WMF or Wikimedia chapters are exempt when performing their duties for these organizations. Admins who work for other organizations which have missions that are aligned with the WMF mission, such as GLAMs (galleries, libraries, archives, and museums), universities and not-for-profit research centers, should declare a potential conflict of interest on their user pages and follow the policy WP:Paid editing disclosure and the conflict of interest guideline closely, but are also exempt when performing their duties for these organizations."

BTW, this proposal was developed at User:Doc James/Paid editing#16. Ban all Admins from paid editing and I will invite the participants there to comment here (as well as the participants of the original RfC). I'll add that, to the best of my knowledge, since the change in the ToU no admin has declared that they are paid editors, so the benefit from being able to tell the public "No administrator may accept payment to edit articles or to perform any administrative function on Wikipedia" far out-weights any actual loss to any admins. Smallbones(smalltalk) 03:03, 5 February 2017 (UTC)

Support concrete proposal 3

  • Support as proposer. Smallbones(smalltalk) 03:05, 5 February 2017 (UTC)
  • Support Another important measure. An important clarification for the wider world. Unlikely to affect any current admins. Doc James (talk · contribs · email) 03:16, 5 February 2017 (UTC)
  • Support If it's good enough for Doc James it's good enough for me. I see nothing compelling or even logical in the oppose votes to date. Nitpicking perhaps, but then, we're used to that. Kudpung กุดผึ้ง (talk) 06:25, 5 February 2017 (UTC)
  • Support. Admin is a position of privilege and trust, and responsibility. The COI with paid editing is extreme. --SmokeyJoe (talk) 10:58, 5 February 2017 (UTC)
  • Support last time there was consensus for something like this (but wording issues hung it up), let us just say it (I suppose, to allay whatever objections there are, let's also say that Arbcom can offer pre-clearance for things not inconsistent with the spirit of this direction, and also maintains it's usual post-hoc discretion such as in the Wilifone paid admin case). Alanscottwalker (talk) 14:20, 5 February 2017 (UTC)
  • Support. The language isn't perfect, but it is close enough. I am beginning to think that perhaps a shorter text might work better: "No administrator may accept payment for Wikipedia editing or administrative functions of any kind, except when authorized by the WMF or community consensus and prominently disclosed on the admin's user page." As for the WikiCup... screw the WikiCup, it's the sort of gamification that tends to sharpen elbows; the community could authorize it for admins, but shouldn't, because admins should be mopping, not counting points. Even banned, the admins could still participate, just not cash the check. Wnt (talk) 14:43, 5 February 2017 (UTC)
The way I look at it, if there is disclosure, the article will be watched like a hawk by most of the parties to this discussion, and the admin editing it will be under a very bright spotlight. If it is undisclosed paid editing, then there is already a problem under the Terms of Use. So I really don't see the whole point.--Wehwalt (talk) 20:58, 7 February 2017 (UTC)
  • Support - Admins should never, in any case, be paid for their editing. Paid editors should not be eligible to be admins. The WikiCup could disappear tomorrow and the encyclopedia would hardly feel its passage, but if we must have it, make the "prize" non-monetary. There is no possibly way to reconcile the standards expected from admins with their being paid for their edits, and any admin discovered having taken money for edits should be desysopped immediately, or give up the bit voluntarily, and register as a paid editor. Beyond My Ken (talk) 14:59, 5 February 2017 (UTC)
  • Support but with broader language - prohibiting administrators from being paid by subjects or potential subjects, in relation to the article about the subject. So for instance I can't be hired for the purpose of persuading another editor to create an article, or to weigh in at an AfD or AfC. (User:Jbhunley frames the issue perfectly below.) Coretheapple (talk) 15:04, 5 February 2017 (UTC)
  • support per Smallbones(again)--Ozzie10aaaa (talk) 19:22, 5 February 2017 (UTC)
  • Support I have no issue with forcing admins to choose between performing trusted work for en.wp or getting paid to edit. The issue here is monetizing their admin bit, an administrator can easily charge a premium for their services and would use theirs status as an administrator as a selling point. This is not acceptable on any level. Jbh Talk 03:27, 6 February 2017 (UTC)
  • Support on the condition that it is reworded to exclude contests, Wikimedia scholarships, and the like. Gamaliel (talk) 04:00, 6 February 2017 (UTC)
  • Support—I find it hard to believe that Wikipedia should have lower standards of conduct for administrators (or any editor, for that matter) than most major news organizations in the U.S. Ultimately, the only protection editors have here is that we show in every way possible that all of our work is governed by Wikipedia is written from a neutral point of view. Anything that that weakens that pillar weakens protection for individual editors. — User:Neonorange (Phil) 18:26, 7 February 2017 (UTC)
  • Support. Paid admins is a bad idea. There should be rules to prevent this from happening. QuackGuru (talk) 03:23, 10 February 2017 (UTC)
  • Support This is a good deal more than a symbolic gesture, since it will be widely known that admins hypothetically could do a lot more for a paymaster than other editors can. "Let me look after your page - $50 per week semiprotection, $1500 for indef full protection, with my groundbreaking Gold Package any criticism is removed and revdel'd and the contributors blocked as 'vandals'...": Noyster (talk), 11:04, 10 February 2017 (UTC)
  • Support paid admin hurtful for reputation, regardless actual misdeeds. Want money - relinquish mop. Of course, as with all "paid editing" it goes without saying that monies from WMF, wikipedian-in-residence, wikicups, and other "wiki-legal" monies are OK. Staszek Lem (talk) 21:03, 10 February 2017 (UTC)
  • Support. Transparent and prevents scams. I don't see any good reason why not. James (talk/contribs) 23:05, 14 February 2017 (UTC)
  • Support per proposer. This should be a no-brainer. Justlettersandnumbers (talk) 20:38, 2 March 2017 (UTC)
  • support--Ozzie10aaaa (talk) 20:47, 9 March 2017 (UTC)
  • support Having an undeclared COI should be grounds for immediate removal of adminship at the very least IMO. All previous COI or current employment that is directly related to Wikipedia should be declared prior to RfA and any changes in that status should result in a review of adminship IMO. I agree with these changes. Endercase (talk) 17:41, 30 April 2017 (UTC)

Oppose concrete proposal 3

  • Oppose. This needs better language. Must admins drop out of the WikiCup, which includes monetary prizes this year? Must they decline travel grants or other grants from the WMF intended to help them develop anything related to Wikipedia that involves making an edit on-wiki? I'm not opposed to banning paid editing in some sense for administrators, but we need to actually get it right, as with all of these proposals related to paid editing. I seriously question what "gain" there is here, by the way. In what conversation with an external stakeholder does "our admins can't edit for pay" even come up? ~ Rob13Talk 04:24, 5 February 2017 (UTC)
  • Wait so paid editors will be banned from becoming admins? Even one user who might've only done one job to write a single article? I also agree with the opposers in the last vote, linked above, who said things like ""It wouldn't do anything but at least we'd have a policy" is a bad reason to have a policy." 63.139.207.2 (talk) 04:53, 5 February 2017 (UTC)
  • Oppose Sounds to me like what is needed is a way of busting these scams, not restricting our volunteers who act in good faith. What this is saying is, even if you act honorably and follow the rules to the letter, that still isn't good enough.--Wehwalt (talk) 05:53, 5 February 2017 (UTC)
    As I mentioned in my support above, the issue is people monitizing their admin bit. Jbh Talk 03:38, 6 February 2017 (UTC)
Then amend it so it only applies to administrative actions and not content contributing. Do you actually think that the usual suspects above would recognize any nuance? No, anything that involves an admin and compensation would be the subject of a a torches and pitchforks brigade. No.--Wehwalt (talk) 08:57, 16 February 2017 (UTC)
  • Oppose per Wehwalt. If people are running scams, the solution is to find a way to bust those scams, not to vent your frustration by harassing the people who devote significant chunks of their time to actually writing and maintaining Wikipedia and who accept the occasional gift in appreciation of their doing so. The most significant payer of paid editors is the WMF itself and various WMF local chapters, who regularly give editors grants for the purpose of writing a specific article or group of articles; there are also significant numbers of Wikipedians in Residence (formal or otherwise) who accept either payment or payment-in-kind for editing articles relating to a particular organization. While I'm not a great fan of this practice, were admins to be banned from accepting payment of gifts or forced to display mark-of-Cain badges of shame every time someone who'd read one of their articles bought them a drink or they got a £10 grant from WMUK to order an obscure book, the result wouldn't be a sudden uptick in Wikipedia's neutrality, it would be a mass resignation of admins. ‑ Iridescent 10:06, 5 February 2017 (UTC)
    I am sorry, the $10 cup of coffee is not and has never been the target of this ban but I hear it, or similar, brought up in every paid editing discussion. It is and always has been a straw man. People brought this type of arguement up when we banned UPE. We banned UPE and no one has been banned for violating the ToU for coffee, books or anything similar. This proposal is targeting people who make money editing Wikipedia - specifically those who make money by exploiting their position of community trust to make money. Jbh Talk 03:38, 6 February 2017 (UTC)
    @Jbhunley: I trust the community to look at a coffee and realize that's not the intent behind the policy. It's common sense. But consider someone malicious and consider what happens under some of these proposals when they try to wikilawyer some nonsense about paid editing regarding an editor they wish to out, possibly fabricating evidence, lying about it, or just pushing something as paid editing which isn't the serious type of abuse we're trying to stop here. In the past, the administrative response is simple; outing is never tolerated, so we immediately oversight the attempts to out an editor. Under some of these proposals, the response is less-than-simple, with an oversighter having to evaluate and make a judgement call about whether something is reasonably paid editing before doing anything, including reviewing all the evidence and diffs, evaluating the claims, and possibly even reading discussions like these to gauge the intent behind the policy. Sure, it will eventually be removed, but in the meantime, someone's life is potentially being ruined by opening them up to real-life harassment, stalking, or violence. I've never said that we shouldn't be more aggressive toward paid editing; I think we should. But we have to get this right. Collateral damage isn't acceptable here. (As for this particular proposal, there are some larger gray areas than a cup of coffee up for discussion.) ~ Rob13Talk 13:31, 6 February 2017 (UTC)
    @BU Rob13: I think you may be replying about the wrong thing here. There is no outing issue here. It is simply a blanket statement that admins can not edit for pay.

    As to the question of outing brought up in other sections I, personally. think it is rather simple for the admins and experienced members of the community to tell the difference between a legitimate COI investigation and an attempt at harassment and outing - a big part of it would have to do with the reputation of the person making the COI claim and whether they were the equivalent of 'involved' with the other person. Mostly what I have seen you worrying about are situations which are both worst case and assuming that the admin staff are idiotic martinets who apply the rules pedanticly with no discression. That is not the case we would ever have with the realistic application of these proposals. I do realize that you may, for your own reasons, have a very low tolerance for risk but, in my opinion, the ability for some bad actor to exploit the COI posting proposals here are minimal. If someone is the kind of person who is going to dox you, they are going to dox you no matter what and we have processes in place to address that.

    Paid editing, both disclosed and undisclosed, is already an existential threat to Wikipedia and it will only get worse until the community takes a firm stand that we will not be used as a promotional, marketing and SEO platform and to do that we must have rules and tools to combat the promotionalism. We wanted to tighten NPP and that got watered down to uselessness - it takes some serious skill to identify UPE and remove it before it gets indexed. Now people are trying to find ways to address the matter of paid promotional content which has been in the encyclopedia for some time and people are more concerned about their own black swan risks than the existential risk to the project.

    I learned long ago to assess and balance the risks of anything I want to do and simply not do those things which I found too risky - assuming it was something I had a choice about. I respect your decision to drop your tools because you found things too risky for your taste quite a bit. People should understand that one must tailor one's activities to the risk environment not attempt to change the environment to meet one's personal willingness to accept risk. Jbh Talk 14:11, 6 February 2017 (UTC)

    I was responding generally to this attitude of "the details don't matter" which has been clearly demonstrated in response to all proposals on this page. You speak about our processes to "deal with" doxxing. The process now is that we remove it from the site immediately, no second guessing, no questions asked. The process after some of these proposals pass (if they pass) is to create large gray areas where we wind up leaving up personal information while admins consider whether it's ok for it to be posted or not. In the meantime, it's potentially copied onto mirrors, propagated around the web, and found by other malicious users on-site. I agree that people will doxx if they want to no matter what, but usually, we don't give them one of the top 10 websites by traffic in the world as their platform to do it. I'm not trying to change the environment; I'm resisting a change. As for dealing with paid editing, I of course agree with you, which is why I've implemented edit filters to detect some variants of abusive paid editing, was looking into working with the WMF to implement machine learning algorithms to help identify paid editing (before my withdrawal from administrative areas of the project, that is), and have floated the idea of implementing a group of functionaries tasked with investigating paid editing and actioning on it without opening things up to a "free-for-all" where anyone can out someone they believe (possibly incorrectly) is a paid editor. It's not like I'm not addressing the issue, and the repeated attempts to characterize anyone opposed to a particular proposal as someone aiding and abetting paid editors is downright insulting. ~ Rob13Talk 18:13, 6 February 2017 (UTC)
    I do not think your particular opposition is 'aiding and abetting paid editors' however I do feel that there is, in general in the community and in particular at the WMF, a massive failure to understand that Wikipedia is not some idealistic social experiment where "anyone can edit" and a total right to anonymity trumps the obligation we have to our readers to provide them with content that is NPOV and as free as possible from promotionalism and PR.

    I do see your concern about rapid propagation of dox to mirror sights and I had not thought about that. Do you have any ideas about how that can be stoped or reasonably slowed? Also, I know that articles propagate rapidly but do our discussion forums/back of house pages also propagate so rapidly. How would NOINDEXing the COI forum, if it is not already, affect the propagation?

    I like the idea of a specialized group of editors, as you suggest, who can manage COI off-wiki - say maybe a specialized OTRS queue where the information provided would fall under the non-disclosure agreement OTRS volunteers sign. Jbh Talk 19:15, 6 February 2017 (UTC)

    @Jbhunley: NOINDEX would theoretically work, but I've seen many userpages (which are supposed to be NOINDEX'd) show up in Google search results. There have been some well-documented cases of search engines ignoring NOINDEX, which is unfortunate. Mirrors can ignore NOINDEX at their discretion, as far as I know, so they may well still host the page. At least some of them. Keep in mind also that, if we host a page like this on Wikipedia, the crazies are unlikely to do nothing. They could setup a web crawler to take a snapshot of the page at rapid intervals and propagate information that way. My idea of the ideal solution would be to have a group of functionaries to investigate paid editing. All information funnels through them, and they make a determination of whether paid editing has occurred or not. If it has, they also make the determination of what information (if any) is necessary to be released to the community to prevent ongoing disruption to the encyclopedia. There should be a clear rationale for why any release of information will prevent harm; more likely to release in cases of long-term or widespread abuse, etc. We could definitely do an OTRS queue, although they'd probably also have to sign the same agreement CheckUsers sign so they can discuss technical data amongst themselves openly (even if not all members are CheckUsers themselves). Basically, allow release of information, but set up a gate keeper so this doesn't turn into a free-for-all. Thoughts? ~ Rob13Talk 13:36, 8 February 2017 (UTC)
  • Oppose per Wehwalt. The problem we have is with undisclosed paid editing, which is already banned, for admins or anyone else, and which this restriction doesn't assist with enforcing at all. In fact, if anyone is going to edit for payment, (and discloses that fact publicly per the proper procedure) then isn't an admin actually a better choice than some jonny-come-lately newbie, since the admin has already been vetted by the RfA process, and been shown to understand and implement Wikipedia's policies, including neutrality.  — Amakuru (talk) 10:59, 5 February 2017 (UTC)
  • Oppose per Amakuru and Wehwalt. This will actually just increase the amount of undisclosed paid editing as people will still get grants, be bought drinks, accept scholoarships, etc, etc, but there is an added disadvantage to declaring this. It will also decrease the incentives for people who are engaging in the type of editing you are attempting to reduce to become legitimate. Thryduulf (talk) 12:00, 5 February 2017 (UTC)
  • Oppose. First, this proposal fails to distinguish between disclosed and undisclosed paid editing. With disclosure, the danger of corrupt use of administrative permissions becomes a moot point. Secondly, I oppose per WP:DEAL. What, so we want to hold admins to a higher standard in this regard than other editors? Not me. Undisclosed paid editing is unacceptable for any user. --Tryptofish (talk) 21:06, 5 February 2017 (UTC)
  • Oppose as instruction creep, and per User:Amakuru. Lankiveil (speak to me) 23:50, 5 February 2017 (UTC).
  • Oppose this is not actually any better an idea than it was last time around. It was and still is a non-solution to a barely-existent problem. There are more known cases of admins being harassed, threatened, or intimidated off-wiki in an attempt to manipulate their on-wiki behavior than there are probable-admin-paid-editing cases, so if your priority is making sure admin decisions are on the level, you should be concentrating your efforts elsewhere. There are serious problems with the drafting of this proposal, as others have already pointed out; unlike some supporters, I do not think these problems are repairable. There is no robust distinction to be made differentiating payment we'd be happy to see (contest awards, [most] Wikipedian-in-Residence positions, a beer from the author of the FA you copyedited, etc) from those we'd be skeptical of. Also, the claimed PR benefits are unproven and sound mostly imaginary; 99.99% of readers give no shits about our internal policies, while editors sure do care about these artificial social distinctions we keep trying to confer on the people who have no special social authority beyond a few extra buttons on their interface. We'd be potentially constraining admins from doing useful work while artificially perpetuating toxic social-class distinctions for the sake of a symbolic gesture whose audience won't know or care about the symbolism, and getting no actual difference in the rate or disruptive effects of paid editing for our trouble. I'm starting to feel like a broken record, but there are lots of things we can try, that are actually within our control, to manage the effects of paid editing: Notability and sourcing requirements, new-editor onboarding, technically simpler disclosure processes, improved spam detection, etc. Start there. Opabinia regalis (talk) 07:02, 6 February 2017 (UTC)
  • @Opabinia regalis:—PR benefits? It's rather a question of reliability and neutral point of view We don't have to reinvent the wheel. Mainstream news organizations have strict, enforced ethical practices manuals. The New York Times and NPR have publically available ethical practice guides on their websites. Other U.S. news organizations also have ethical practices guides—some publically available, some not. It's a question of 'content'—is it free of COI influenced articles to the best of our ability. The reputation of Wikipedia is primary. Reduce the reputation of Wikipedia enough and the problem of undisclosed-paid-COI editing will go away—it won't be worth paying to put an article into this encyclopedia. We protect editors from outing or other harassment to build the encyclopedia, not because it's the right thing to do, but because we have decided it's the only way to build the encyclopedia. We have a code of ethics that prevents outing—that is enforceable. We should have a code of ethics that prevents enabling undisclosed-paid-COI editing—that is enforceable. Especially for admins.
(I'm sorry to have to write this using an explicitly U.S. example, but that's what I know.) — User:Neonorange (Phil 02:58, 13 February 2017 (UTC)
  • Weak oppose because of it targeting specifically admins and because of the fact that it could be used as a form of harassment. In those cases where admins are paid editors, and that information becomes publicly available, I think it might be possible to deal with things in such a way as to not necessarily involve labelling the admin. I think most admins caught would probably step down voluntarily if someone e-mailed them with good enough information anyway. John Carter (talk) 17:02, 6 February 2017 (UTC)
  • Strong oppose — This de facto means that no administrator may engage as a Wikipedian in Residence, which I presume is not the intended goal. We need to learn how to better handle the distinction between good paid editing and bad paid editing — something this proposal fails miserably at doing. Carl Fredrik 💌 📧 19:23, 6 February 2017 (UTC)Edited Carl Fredrik 💌 📧 09:57, 9 February 2017 (UTC)
    User:CFCF does "Admins who work for other organizations which have missions that are aligned with the WMF mission, such as GLAMs (galleries, libraries, archives, and museums), universities and not-for-profit research centers, should declare a potential conflict of interest on their user pages and follow the policy WP:Paid editing disclosure and the conflict of interest guideline closely, but are also exempt when performing their duties for these organizations." not cover your concerns? Doc James (talk · contribs · email) 09:12, 9 February 2017 (UTC)
  • Oppose. What you really mean to say is that admins are not allowed to tale bribes, unlike elected government officials who are (aka "campaign contributions"). Unfortunately, your proposal doesn't say that, so I have to oppose. Admins should be allowed to accept payment for performing any policy-supported function on Wikipedia, including reverting and deleting contributions that violate copyright, etc. wbm1058 (talk) 23:29, 6 February 2017 (UTC)
  • Oppose You don't fix a scamming problem by outlawing otherwise-lawful behavior. Which is DISCLOSED paid editing. North8000 (talk) 19:34, 18 February 2017 (UTC)
  • Oppose per North8000 which has been my opinion for a long time. This is about principals: new rules won't stop bad faith editors. It only burdens the good faith editors. It no longer surprises me how many fanatical activists will keep pushing more and more rules that aren't effective and with no evidence that more rules will be any more effective than the old ones. These proposals only serve to give us a false sense of security.--v/r - TP 23:29, 24 February 2017 (UTC)
  • Oppose - In the case that an admin would take money for editing, I have confidence that one, the admin would make the edits in accordance with our policy, and that two, if they don't, that the community will be able to handle it. This is just burdening the good faith editors, and not actually solving the problem. And anyways, if the admins did take money for editing and their contributions were against policy, then what will new policy due to fix it? Will they suddenly start following policy? I doubt it. RileyBugzYell at me | Edits 14:51, 18 March 2017 (UTC)
  • Oppose. I see no problem with admins, which are supposed to be ones of the most trusted members of our community, writing payed articles once they have disclosed that and follow the rules as long as they do not use their admin flag on the articles with CoI. --Lingveno (talk) 18:15, 3 April 2017 (UTC)

Discussion of concrete proposal 3

@63.139.207.2: There's nothing in this that says "If you've ever been a paid editor, you may not become an administrator". Rather it says, paraphased "If you are an admin *now* you may not do paid editing *now*." Also, this is not "having a policy, just to have a policy." Its having a policy so that we can say loudly and clearly to our readers, to potential scam victims, to ordinary editors, to admins:

  • "No administrator may accept payment to edit articles or to perform any administrative function on Wikipedia."

That simple communication should work wonders.Smallbones(smalltalk) 05:31, 5 February 2017 (UTC)

@BU Rob13: This has nothing to do with travel grants, WMF grants, etc. It has to do with the Terms of Use which clearly describe paid editing and have an extensive FAQ as well. Payment for Wikicup? I don't know how much they are paying, but it doesn't look like anything that comes under the ToU to me. If Admins, or any opposers, want to make this work, they can. Just propose your own text. I just would suggest, please keep it simple, so that we can say it loud and clear to potential scam victims. When has this come up in external conversations? One example is when the OrangeMoody scammers told their potential victims that they had paid admins on staff. We can put this up at AfC, in the press, etc. so that everybody will know

  • "No administrator may accept payment to edit articles or to perform any administrative function on Wikipedia."

Smallbones(smalltalk) 05:31, 5 February 2017 (UTC)

@Smallbones: WP:PAID is pretty clear that grants for Wikipedia-related work and prizes for community contests are paid editing, though. From the terms of use, "As part of these obligations, you must disclose your employer, client, and affiliation with respect to any contribution for which you receive, or expect to receive, compensation." That seems to restrict things to employment situations, until you look at the definition of employer, which is simply "the person or organization that pays a user to contribute to Wikipedia" (with notes that this does not need to involve being a salaried employee, etc). Contribution is similarly defined quite broadly as "any text added to Wikipedia, including talk-page and sandbox contributions, and material added to articles by others at the behest of paid editors". Under those definitions, if I accept a grant from the WMF to develop a training program for OTRS agents on the permissions queue, I would become a paid editor (cue spooky music) as soon as I document the training process on-wiki, even on talk pages, or when the people I'm training add a permissions template to any file at my direction. That's not your intent, and I fully understand that, but this is the issue that's plaguing all of these paid editing proposals. I agree with you we need to find a way to make this work, but the devil's in the details. As for proposed language, I would go with something like "No administrator may accept payment to perform any administrative action on Wikipedia. Further, no administrator may accept payment from any person or organization to make contributions related to that person or organization." (Plus the exemptions for WMF and GLAM, etc.) This eliminates admin-for-pay situations, which I agree would be problematic. It further eliminates clear COI paid editing, but stops short of preventing all paid editing. Thoughts on that? ~ Rob13Talk 07:36, 5 February 2017 (UTC)
You are just missing this somehow. You think that if you get a grant from the WMF, you are employed by the WMF. The FAQs on the ToU don't support that reading. But, in any case the 2nd sentence here, "Admins who work for the the WMF or Wikimedia chapters are exempt when performing their duties for these organizations," would mean you are exempt. Smallbones(smalltalk) 10:29, 5 February 2017 (UTC) @Iridescent: this paragraph applies to your comments as well. Smallbones(smalltalk) 10:35, 5 February 2017 (UTC)
@Smallbones: Our current definition of "employer" does not actually require one to be employed (see WP:PAID). WP:PAID is local policy, and even if the Terms of Use are less restrictive, we're presumably required to comply with the more restrictive language in the local policy, adopted without a consensus-seeking process. But let's set all that to the side for a second. What about the WikiCup prizes, which is direct money for edits, with heavy encouragement to contribute on certain articles which provide more points? What about a grant to create a bot or script to do some task, where the exact edits the bot will do are clearly spelled out? What about an educational program that pays someone to lead a workshop, including setting up a specific course page on-wiki? Those are all examples of compensation for specific edits. Alternatively, say we perfectly accept the FAQ of the terms of use, which states it's only paid editing if you're directed to make specific edits or contributions. What if Evil POV-Pushing Organization #82 contacts me and tells me to promote their interests on Wikipedia, without giving me specific articles to edit? What if a multi-level marketing scheme pays me to make edits which are biased toward multi-level marketing in general, while telling me specifically not to edit their own article and giving me no further instructions? If we interpret that FAQ the way you're saying we should, then that isn't paid editing which must be disclosed, but we clearly would want it to be. Our definitions of paid editing are contradictory and imprecise. Depending on how you apply them, they can be too broad or too narrow. This is one of the main reasons I worry about the outing business. Right now, if someone gets the definition "wrong" when you compare it to common sense, we can educate them, correct the situation as necessary, and move on. If we establish outing of paid editors as policy and get the definition "wrong", we may have opened up someone to harassment and stalking. In this specific example, we may have opened up someone to desysopping or forced them to give up a good opportunity both for themselves and the project due to vague language that is difficult to interpret. The details matter. ~ Rob13Talk 12:48, 5 February 2017 (UTC)
Let's please confine ourselves to real world possibilities. It does not move things forward to talk about unemployed employees or any other bizarre scenarios that you outline above. Perhaps you can address the issue of why admins want to have a right to accept pay for editing, even though none of them have declared that they are paid in the last 2.5 years? Smallbones(smalltalk) 15:25, 5 February 2017 (UTC)
Unemployed employees are in our policy. Our definitions state an employer is the person who pays an editor even if the editor "has no contract". I've disclosed on-wiki that I received a travel grant from Wikimedia NYC to attend Wikipedia Day NYC with the understanding I'd make edits/contributions related to a few things I specifically mentioned in my travel grant application. So there's one admin paid editor right there, according to the definitions at WP:PAID. (Technically, I'm not currently an administrator. I handed in my tools due to how concerned I am about the potential for sockmasters I've pissed off to maliciously out me under some of these proposals, but I was an administrator at the time.) You requested I advance my preferred wording instead of just opposing yours, which was a fair critique. I've done so, but it's received no response. What do you want to prevent that's supported by your wording but not mine? ~ Rob13Talk 15:35, 5 February 2017 (UTC)
Please stop talking nonsense. If somebody pays you to do a job, you are a paid editor and his employee. If you get a travel grant from the WMF, paid editing doesn't apply to you, and this proposed rule simply doesn't apply to you. I'd like to have a discussion with somebody who is not trying to pull the wool over people eyes. Smallbones(smalltalk) 16:33, 5 February 2017 (UTC)
You can say whatever you want about me. All I've tried to do is hammer out the wording so the policy actually does what you want it to do without being interpretable as something broader. Our definition of "paid editing" at WP:PAID is rather horrible, and we should either fix that or be specific here. I still await any explanation of what you're trying to prevent that wouldn't be prevented by my text. ~ Rob13Talk 13:19, 6 February 2017 (UTC)

I'm not so sure that "No administrator may accept payment to perform any administrative function on Wikipedia." is going to be compatible with what we want without exception. For example, if $controversial_company has an article that is frequently the target of vandalism, POV-pushers, etc, would it be problematic for that company to offer money to an admin to keep the page vandalism free? Even if they explicitly told that admin to disclose this on the talk page and explicitly told them not make any edits to content related to the company? Obviously such vandalism reversion etc should happen as a matter of course, but even so paying someone specifically to ensure their article is vandalism free may not be unattractive to companies with large budgets who are not au fait with the inner workings of Wikipedia. Thryduulf (talk) 13:00, 5 February 2017 (UTC)

@Thryduulf: Personally, I think such actions should be excluded. We don't want to establish a "fast lane" to counter-vandalism which involves payment for many, many reasons. For one, it gives the appearance that there's a "slow lane" (see net neutrality). Second, it's fundamentally at odds with our values to accept payment for basic maintenance tasks on our free encyclopedia. Third, it may carry legal implications to accept payment for a "fast lane" of counter-vandalism when we consider libelous edits. Imagine someone arguing in court that we didn't handle libelous edits on their page but did handle the vandalism on the page of someone who paid us in the same time frame. That doesn't look so great. ~ Rob13Talk 14:27, 5 February 2017 (UTC)
Are you saying there are admins undisclosedly doing this now? If not, it seems fantastical, a weird hypothetical that makes little sense. And yes, is problematic not just because it creates COI, but it means article edits are privileged to the highest bidder. Alanscottwalker (talk) 14:33, 5 February 2017 (UTC)
@Thryduulf: I would find such a situation extremely distasteful and a massive ethical failing on the part of the administrator. This is precicely the type of thing this should prevent and I would go so far as to say that if any administrator is doing this they should be desysoped, and possibly banned, for violations of the ToU for UPE. They are exploiting their position of community trust to make money. Jbh Talk 03:49, 6 February 2017 (UTC)
I do not know if anyone is doing this now (I've certainly never been approached), but I cannot see how it would be a violation of the terms of use regarding undisclosed paid editing if everything is disclosed. Whether it is ethical, and if it is not what the consequences should be, are different questions but if you desysop and ban someone for disclosing a COI then there is an incentive to hide it which is the exact opposite of what we want. Thryduulf (talk) 10:01, 6 February 2017 (UTC)

@Thryduulf, Wehwalt, and Amakuru: I understand this proposal does not address undisclosed paid editing, but instead a kind of disclosed paid editing we don't see admins doing at this time. But if they are not doing it now that is precisely why this is a good time to ban the practice! We know full well that if an admin starts doing it, there is going to be a huge brouhaha, a lot of personal arguments and recriminations, a lot of little details about whether a COI working on the article of, say, one coal company means that you can't block someone for vandalizing Clean coal or even declare a short term for a block of someone vandalizing Global warming. Pandora's box is reputedly much easier to lock when it is closed. Nor should we underestimate the danger here --- just look up Sourceforge. Even the smallest amount of commercialism in a free-culture enterprise can rapidly lead to its complete humiliation. Wnt (talk) 14:52, 5 February 2017 (UTC)

@Wnt: Under our broad definition of paid editing, they are doing it now, actually, in the form of the WikiCup, accepting grants to work on specific projects related to Wikipedia's mission, receiving travel grants to edit-a-thons, etc. Under our definition of paid editing, I'm potentially a paid editor because I received a travel grant to Wikipedia Day NYC with the accompanying expectation that I would make edits specifically related to activities discussed there, which were used to support my travel grant application. Must I cease that type of activity under this proposal? As written, yes. We're throwing out all the good stuff with zero evidence any bad has ever occurred. ~ Rob13Talk 15:31, 5 February 2017 (UTC)
@BU Rob13: Given the text as written, admins working for the WMF are exempt, and the travel grant is paid for work done for the WMF. So there would be no change or action required on your part. For WikiCup, the version proposed for this vote might exempt you as a Wikipedia allied activity but it isn't totally clear to me - presumably some manner of community consensus on whether it is or not can be arrived at, and it is not mission critical. (I mean seriously, how diligent an admin can you be and also be trying to win a cash prize as the "best" content editor?) Wnt (talk) 16:38, 5 February 2017 (UTC)
Rob, I'm concerned about this part of your post at 12:48, 5 February:

WP:PAID is local policy, and even if the Terms of Use are less restrictive, we're presumably required to comply with the more restrictive language in the local policy, adopted without a consensus-seeking process.

WP:PAID isn't more restrictive than the terms of use. It's supposed to reflect them, which is the only reason it's policy. Can you say which part you see as implying that it's more restrictive? SarahSV (talk) 15:43, 5 February 2017 (UTC)
@SlimVirgin: Smallbones has pointed to the FAQ on the terms of use, which states an edit is paid editing only when the edit or contribution is specifically made at the direction of an organization that's paying the editor. That definition would not include things like "a museum employee who is contributing to projects generally without more specific instruction" (from the FAQ) or an editor who accepts a grant to improve coverage of some broad topic area without further direction on what to edit. But WP:PAID states that "Wikipedians in residence who are paid must disclose which organization (GLAM or similar) pays them", which seems to contradict the specific museum employee example above, and it contains no language granting any exception when the direction received by an editor receiving compensation is minimal. Since PAID is local policy (as claimed by many editors on this page - I personally see its status as more dubious), editors must follow it regardless of what the TOU states. If it's intended merely to document the TOU, it should be an information page, not policy. ~ Rob13Talk 15:51, 5 February 2017 (UTC)
Rob, the ToU FAQ state that this is policy until an RfC replaces it with an alternative disclosure policy. It's a policy for that reason only, but whether we call it policy or an information page, it amounts to the same thing, namely that disclosure is mandatory if you're being compensated for edits.
Re: the museum example, the FAQ page says: "you are only required to comply with the disclosure provision when you are compensated by your employer or by a client specifically for edits and uploads to a Wikimedia project." A WiR who is being paid is being paid in relation to Wikipedia, so that does fall under the ToU. There's no need for someone to be telling you "add these words", "upload this image". SarahSV (talk) 16:51, 5 February 2017 (UTC)
Arbitrary break
I haven't yet weighed in with a formal Support of Oppose. I am very sympathetic to the goal, and likely to be supportive, but a number of issues have been raised that aren't yet satisfactorily resolved.
  • Can a paid editor become an admin? I accept Smallbones's argument that this is not in the proposal. I view the proposal as covering actions after one gets the bit. Yet I see @Beyond My Ken: including “Paid editors should not be eligible to be admins.” in their support statement. What is this, a condition for supporting? A belief that the proposal does prohibit this? Some clarification form BMK would be nice
  • Can an admin temporarily give up the bit to do paid editing? We generally allow admins to voluntarily relinquish, then ask for restoration of the bit. Can they engage in paid editing in the interim?
  • What constitutes “payment'? Cash, of course, but travel grants? Free products from companies? Free access behind paywalls? Meals? A salary or stipend not tied to editing?
  • Nexus between payment and edit. Obviously, “we;ll pay you $X to include this information” is covered. What about “Please improve Article Y, we'll leave it up to you to determine what should be included”? How about, “We'll provide a stipend, name you WIR for our organization, ask you to teach our staff how to edit, and occasionally edit, if you see something that needs improvement”? Is the answer the same if the organization is not a museum, but a consortium of coal companies? If you give different answers, where do you draw the line? Non-profits OK, but for profit verboten?--S Philbrick(Talk) 19:42, 5 February 2017 (UTC)
The payment question is also interesting, because OTRS volunteers often have stories about receiving a free lunch or other gift from companies or notable individuals after helping them out with a particularly difficult issue. It doesn't happen often, but most long-time OTRS volunteers who answer Quality tickets have had it happen at least once. That's directly compensation as a result of editing on-wiki, in most cases, but the compensation isn't asked for and occurs after the editing. Is that paid editing? Our definition definitely doesn't make that clear. ~ Rob13Talk 13:22, 6 February 2017 (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.

RfC discussion

It is never entirely possible to have perfect representations of any viewpoint, and I think we're letting perfect be the enemy of good here. The point is that COI is a massive problem that we have pretty much been sweeping under the rug — we cannot continue doing so. Carl Fredrik 💌 📧 13:41, 28 January 2017 (UTC)
  • (edit conflict) This RfC is unfortunately malformed and overly simplistic. Reasons:
    • The word "investigating" is apparently meant to mean "handling offwiki information". Yet it doesn't say that, and in the absence of that specificy, this RfC makes no sense. There are investigations of COI/paid editing that happen every day here in WP that only look at on-wiki behavior and edits, and surely this RfC is not addressing that. Yet the broad, simple wording pulls that perfectly legitimate behavior into it.
Milieu 1 - I don't understand what this even means and I am (still) one of the biggest contributors to COIN. What does this even mean when "investigating" is used in its normal English sense?
Milieu #2 has the same problem; COI/Paid editing is addressed "openly and transparently" at COIN every day, without coming near the OUTING line. What does this mean? We have had two recent ANIs that resulted in blocks without coming near violating OUTING: Wikipedia:Administrators'_noticeboard/IncidentArchive942#Earflaps and the section under that, as well as Wikipedia:Administrators'_noticeboard/IncidentArchive943#Proposal_to_ban_FoCuSandLeArN_due_to_undisclosed_paid_editing. I can bring more ANI cases based on ToU violations if anyone needs more
Even Milieu #3 does not match community practice There are admins who block based on ToU violations and even for those admins who want to be very cautious, per WP:BLOCKEVIDENCE the evidence can be brought to a CU or oversighter. Not just arbcom. So the arbcom-only option in #3 is false with this too broad umbrella of "investigating" as the key term.
And to be frank, the arbcom-only option in #3 is unfotunate, especially when Arbcom has made it clear that it will not address paid editing (Wifione case Principle, and additionally when several admins and editors have told me that they have sent off-wiki evidence to Arbcom and gotten back no reply. At all. With that background, Milieu 3 (which is false in any case per WP:BLOCKEVIDENCE) doesn't seem feasible.
    • There are aspects to all three that I support, and aspects that I don't. It is just too simplistic and distorted.
    • The WMF statement is not authoritative in WP. It doesn't need to be "refuted" in this way, or in any way. It is just advice and stated as such.
    • The question of how to handle off-wiki evidence in COI/paid editing investigations is difficult and takes nuance and care to discuss. And the community has a hard time talking about it, as it evokes lots of emotion and many people only half-understand the policy background and even fewer understand the actual practices used on the ground at COIN and elsewhere to address it responsibly (including minding OUTING). A misleadingly worded and simplistic RfC like this, will not help resolve anything.
User:Casliber would you please withdraw this and open a discussion about how to frame the RfC? . Jytdog (talk) 05:06, 27 January 2017 (UTC)
Agree that one cannot codify common sense or good faith. Both these are needed in the follow up of UPE and harassment. A step wise algorithmic approach to this is no easier than creating one for what is a good a reference or a good article. Judgement is required. Doc James (talk · contribs · email) 05:32, 27 January 2017 (UTC)

I initially framed this as examples about what an editor can/is allowed to do if s/he has suspicions about COI editing:

  • In milieu 1, they can only examine the content of the edits - any speculation on accounts is off-limits.
  • In milieu 2, they can investigate publicly
  • In Milieu 3, they can only investigate privately and communicate concerns to an identified group (Arbcom, WMF, maybe checkusers or whoever we can figure out to fit here).

I don't think there are any other options (can you think of any?). I will slot these examples above. And no, I don't want to restart as I can't see other options. We can fine tune ideas within those miieus later. Cas Liber (talk · contribs) 05:37, 27 January 2017 (UTC)

I guess I am somewhere between 2 and 3. I added 4 to reflect that option. Doc James (talk · contribs · email) 05:42, 27 January 2017 (UTC)
Cas, because you don't specify that the concern is about off-wiki personally identifying information, the whole thing doesn't make sense. With the strong emotions people have they will take "investigations" broadly. Some people got angry with me just for asking an editor if they had any connection to a subject they edited, which doesn't come even close to violating OUTING. It is was the asking - the investigating - that upset them. Jytdog (talk) 05:51, 27 January 2017 (UTC)
Okay, I will concede that - my aim was to cover this whole area and what we do about it, and I guess it doesn't come across clearly. And the fact that there is a lack of consensus on the outing vs investigating is what we have to fine-tune. @Jytdog: and @GorillaWarfare: who would you both frame it? My alternative take was the scenarios, or otherwise principles. But am open to ideas to try and fine-tune this RfC early. This isn't rocket science and I actually think there is a lot of common ground here. Cas Liber (talk · contribs) 07:30, 27 January 2017 (UTC)
PS: @Jytdog: this is another reason to fine-tune policy as you found out that some folks have objected to anything even being asked and would view even speculation as a violation of privacy. Cas Liber (talk · contribs) 10:21, 27 January 2017 (UTC)
Cas, I think you missed "discuss things privately with editors outside of 'the identified group'". I would be surprised if nobody ever sent an e-mail message to a sympathetic admin about a suspected problem. (In terms of BLOCKEVIDENCE, if a block is warranted, and it can't be passed off as merely a DUCK situation, then all the admin needs to do is to forward said evidence to a suitable private mailing list.) WhatamIdoing (talk) 08:15, 27 January 2017 (UTC)
that becomes a rather vexed issue. We'd need some guidelines but maybe instead of admin it'd be "experienced editor" and then the ideal would be that that editor would only be helping forward the person with the suspicions to arb/WMF/appropriate body etc. and not be involved in investigating themselves. Cas Liber (talk · contribs) 10:12, 27 January 2017 (UTC)

Look, I have wondered whether splitting hairs over what is and what isn't paid editing is making it unnecessarily complicated. If we broadly define/categorise COI editing, where a person clearly has a vested interest in a subject by virtue of money, employment or whatever, we then need to determine how they can be investigated and how their editing can be assessed and how proscriptive sanctions are. Cas Liber (talk · contribs) 10:18, 27 January 2017 (UTC)

Casliber I've reviewed this whole thread and appreciate that you acknowledged here that you were not well aware of what goes on on the ground with regard to COI/PAID investigations and the on-wiki signs that folks who work on these issues are aware of and employ regularly. Nor how that work then ties into off-wiki information. I've provided some links and thoughts about past extensive discussions of these issues and the various individual issues that I think need to be clarified, probably through a series of RfCs. I again ask you to withdraw this one; it doesn't build on the past discussions and is too vague to handle the many issues involved and will not help the community move toward a workable consensus. Jytdog (talk) 20:32, 27 January 2017 (UTC)

Cas Liber Could you possibly provide a link to the "new WMF essay" mentioned in your preamble, so that the masses may discover what you think is in conflict with our policy?: Noyster (talk), 10:38, 27 January 2017 (UTC)

Hi @Noyster:, it's at Wikipedia:Wikimedia Foundation statement on paid editing and outing - and linked from Milieu 2 section above. Cas Liber (talk · contribs) 10:43, 27 January 2017 (UTC)
With all due respect for Casliber, what does "milieu" mean in this context? Are we !voting for specific rules (I don't get that feeling) or for a general set of feelings on how to best proceed?
If so I think M1 establishes that we should not rule out linking to other sites in investigation of paid editing. It's almost like "on a case-by-case basis". I'd say "again" but that's where we still are.
M2 seems to say everything should be transparent, which looks like a stalking horse to me. Everything? Better would have been "as much as possible" In any case there's no consensus (leaning to oppose) so it doesn't help in moving forward.
M3 looks like consensus at first glance, submitting personal information to a (new) body that will consider it in private. My concern here is the word "only" which should be a four letter word in this discussion. Looking at the support votes, most of the start out "if", "contingent" or "I interpret". It looks like part of the solution to me, but not the only one.
M4 briefly "private with some release" leaning to support. Looks like one possible variant to me.
M5 few !votes yet, but to me it looks like a warmed over status quo, with some additional restrictions implied but not clearly stated. Nice try, but no cigar.
So where do you want to take it from here? I'd say let's get specific proposals (i.e. real rules) that are consistent with M1, M3 and a bit of M4
Smallbones(smalltalk) 20:59, 27 January 2017 (UTC)
I used the word "milieu" to mean "setting" - I sorta regret doing that now, but wasn't sure how to frame this in the first place. But am happy some discussion has started. Cas Liber (talk · contribs) 20:26, 27 January 2017 (UTC) (moved from subsection below - also an ED) Smallbones(smalltalk) 21:05, 27 January 2017 (UTC)

Break 1 (this may be a long discussion)

In general I agree that the RfC is malformed. ArbCom should not be campaigning to eliminate enforcement of the ToU, and unfortunately it sometimes looks that way. ArbCom cannot veto policy. They have to support some way that the ToU can be enforced and I urge them to get serious on finding a way to do this.

One extremely simple way would be to just state the obvious. Linking to public job postings for paid editors does not out any of our editors.

Another way would be to support a bright line policy on paid editing, e.g. Commercial editors may not edit articles. They may only edit talk pages and noticeboards. Some might say this would be purely symbolic since few if any commercial editors declare that they are doing paid editing. Symbolic statements can have great meaning however. In this case it would allow us to send a very simple message to the whole world. If you are a commercial editor, or you are thinking of hiring one, then they can't edit the article.

I do think CasLiber is correct that fine distinctions about who is a paid editor often get in the way of a reasonable discussion of our policy choices. We want to target the commercial paid editing companies, and I think that folks who try to enforce the ToU concentrate on these folks, letting the other paid editors off with a discussion of WP:COI and a warning, at worst. Let's define "commercial editor" as part of WP:Paid editing disclosure and put strict limits on them. Any of the usual so-called "border-line editors'" can be listed individually as exceptions.

For example, Commercial editors can be defined as

Commercial editors are
  • those businesses which publicly advertise their paid editing services on Wikipedia articles
  • editors who work for these businesses
  • any editor who enters into a commercial transaction valued at more than $100 with a business to edit Wikipedia articles
  • editors who work for the marketing, advertising, or PR departments of a business, when editing articles about that business, its products, or competitors
then list all the distracting exceptions folks often bring up, e.g. Wikipedians working at a Wikipedia sanctioned editathon to help register new users; college professors discussing their books (COI yes, Paid no), etc.

There are other ways to enforce the ToU. ArbCom, please find one that is acceptable to you!

Smallbones(smalltalk) 12:36, 27 January 2017 (UTC)

There is another way to close down most of the paid editing job posting sites that could be done. This has been discussed in various forms and would lead honest job posting sites to stop posting Wikipedia jobs. It would need an RfC at WP:Paid editing disclosure. It needs 2 complimentary parts; 1 Advertisers must do, or else ... and 2) Paid editors must not do ...

  • Advertisers who post paid editing jobs publicly online must link to WP:Paid editing disclosure and require that the paid editor disclose his paid status and otherwise follow the policy. They must state their company's name and the client's name.
  • No paid editor, disclosed or otherwise, may edit for, or accept payment from, advertisers on the list of non-compliant advertisers.
    • Paid editors who accept employment through public job posting sites must link to the advertisement of each job they accept.

Since honest job boards would require their advertisers to follow these rules, we can easily request the ad be removed if there is no link to our policy, or if there is no disclosure from a paid editor. Smallbones(smalltalk) 15:22, 27 January 2017 (UTC)

I would support this "Linking to public job postings for paid editors does not out any of our editors." Add it as an option User:Smallbones Doc James (talk · contribs · email) 04:57, 28 January 2017 (UTC)
agree w/ Doc James--Ozzie10aaaa (talk) 23:56, 28 January 2017 (UTC)

Break 2 - past RfCs, paths to explore

User:Casliber one thing that the next RfC on the intersection of OUTING and COI/paid editing should take into account are RfCs and discussions at WT:HA and village pump (stemming from WT:HA discussions) over the past year or so. There is a great deal that can be learned from them about the various points of view on the relationship. I have reviewed them and I have found no way to thread the needle in a way that will find broad consensus. What you will find, is some careful thinking, but also lots of strong emotion and things painted with broad brushes that allow for little nuance. You will see how hard it is to "herd the cats" toward consensus. I provide some comments as I go...

Based on these discussions, the recent WMF statement, and Arbcom's statement, I think an RfC, or perhaps a series of RfCs, that addressed concepts of:

  • Clarification that PAID/COI editing is a subset of advocacy
  • clear context that what is being discussed, is editors with a pattern of making advocacy edits (which is actually definable), not just anybody. It always starts with content.
  • Clarification of whether the focus is only about money-related advocacy (promoting a product, company, or a person like an author who is trying to sell books) - in other words, edits where there is likely an obligation per WP:PAID or includes advocacy editing more broadly (a fan of some person, men's rights advocates who attack feminist content persistently; you have all seen this sort of thing).
  • extent of the advocacy editing (kind of in order: big sock networks; individual accounts that edit lots of articles; apparent one-off editors (we get lots of these - create an account to do a job, then abandon the account); passer-by who does a few advocacy edits)
  • Clarification that the focus is off-wiki information, and specifying the kind of off-wiki sites that can be used (see start of a discussion at July 2016 above)
  • Clarification of the kind of information that is relevant to the kind of advocacy editing under discussion
  • what to do with off-wiki information (can any of the specified types be posted on-wiki, should any of it be emailed to someone, and if so who?) and if so, in what are the contexts for each (type of disruption going on, type of off-wiki site, type of information, where in WP, what kind of recipients)
  • If off-wiki information is handled through email; how do we make than an accountable process? Should outcome of evaluation be posted somewhere to allow appeals and allow that information to be built on for future sock or other ToU issues that might arise?
  • and somewhere in this, a discussion of whether (!) and importantly, how, the community would benefit or be harmed if specific kinds of off-wiki information, in specific contexts, could be posted on-wiki. This is probably the most emotionally-laden/divisive aspect of all this.

There is more that should be linked from WT:HA. User:Tryptofish has been very active at WT:HA trying to help facilitate discussions over a long arc to try to help the community find consensus on this issue and i reckon they would have something to say here about paths forward. Jytdog (talk) 19:04, 27 January 2017 (UTC)

  • Two additional issues where WMF legal's input would be extremely helpful, one of which we already have:
How the WMF privacy policy relates to all this, which has been brought up several times in discussions at WT:HA. Legal spoke to this in their statement: The privacy policy applies to data collected by the Wikimedia Foundation, by Foundation partners, by some users in special roles (such as checkusers), or by some third parties who provide data to the Foundation. It does not apply to the publicly visible postings of users on the projects or to information collected from other websites and later posted to the projects. As such, our opinion represents our view of the best way to handle these issues, but does not represent a legal requirement for the projects.
Arbs have stated several times they are at individual legal risk if they address these issues. User:Gorilla Warfare just restated this today at the wikimedia-l here. User:Thryduulf has also articulated this, here and here in a response to a question from me seeking clarification to their first post. Folks who want Arbcom to be involved in this need to keep this in mind. And what applies to individual arbs may also apply to any admin or CU or oversighter. This is really important for everybody involved in these discussions. It would be very helpful for WMF legal to speak to this, so everybody can hear. Pinging User:Jrogers (WMF) and User:Slaporte (WMF) who have each edited the WMF statement, which does not speak to this. Jytdog (talk) 19:30, 27 January 2017 (UTC)
Jytdog that is a good timeline - and helps give ideas of where to go next. At least even if malformed, this RfC has started people talking. Cas Liber (talk · contribs) 20:27, 27 January 2017 (UTC)
But we've been discussing, almost non-stop, at WT:HA, and this RfC doesn't build on those discussions at all. It doesn't address the key divisive issues in a way that will allow us to meaningfully gain consensus on specific points and build on them. Jytdog (talk) 20:35, 27 January 2017 (UTC)
But how many participants have taken part there? You need to get a broader swathe of folks involved. Cas Liber (talk · contribs) 23:50, 27 January 2017 (UTC)
In my view no consensus can emerge from this vaguely worded RfC, and the last thing the community needs is yet another unfocused discussion about paid editing, that will not get us anywhere. There is always way too much hot air and smoke on this issue and not enough care, clarity, and consensus-building. I don't see how involving more people in yet another unfocused discussion is good for anything. I'll stop asking you to withdraw this, as you are obviously sticking to your guns. Jytdog (talk) 00:08, 28 January 2017 (UTC)

It was the community's inability to reach consensus on the tension between managing COI/paid editing and OUTING in these many, extremely long and heated discussions (four of which went on simultaneously following the Banc de binary and Wiki-PR scandals), that led to the proposal to change the ToU by WMF, ratified by the communities in a big RfC. User:Dank did a lot of work trying to facilitate the Nov/Dec 2013 discussions and closing them, and had some ideas about ways forward then. They may have thoughts now, so I pinged them. Jytdog (talk) 02:03, 28 January 2017 (UTC)

One position seems to have more support than the others, but (assuming those numbers hold up) it's probably possible to write something up that gives the nod to that one while simultaneously building in safeguards that respect the main concerns of the other positions. There's a lot of overlap here that can be taken advantage of.
I have expressed concern in the past to one of the WMF people I have a lot of respect for that the WMF seems to spring things on the en.wp community from time to time that have the effect of generating ... not necessarily a lot of drama, let's call it exasperated pushback. It doesn't seem like their best strategy. It generates a lot of "oh geez not this again" voting from people who don't necessarily want to spend the time investigating and voting on the issue, but feel like they have to. - Dank (push to talk) 02:22, 28 January 2017 (UTC)
Thanks for commenting! Jytdog (talk) 03:06, 28 January 2017 (UTC)

Break 3 - finding common ground, do we all agree there is a need for some form of confidential area to discuss COI/Paid editing?

My reading of this page suggests most folks think that there is - so maybe the first thing is to gain consensus that this is necessary (which I think it is). Cas Liber (talk · contribs) 00:32, 29 January 2017 (UTC)

Yes, the majority supports this. The question is can we get consensus on how it will work? Have been considering starting a separate Wiki for this work.
Doc James (talk · contribs · email) 04:03, 29 January 2017 (UTC)
Please see also my reply to you at WT:HA#Break 2. I do not pre-suppose that there is any consensus about this, and I advise other editors not to assume it either (although I personally support it). But I am strongly in favor of having a clear RfC about it, and find out where the community really stands. Consequently, I think that the first RfC question must ask whether or not, in fact, editors support the general idea of having any such confidential area. A second question could ask whether it is necessary. --Tryptofish (talk) 21:37, 29 January 2017 (UTC)
This needs to be done with a set of editors designated to investigate this activity by the Foundation or the Arbitration Committee, with those editors signing the agreement regarding non-public data (to apply to everything on the paid editing wiki). A random editor starting an off-wiki website to investigate and determine the identity of paid editors would be nothing short of a stalker blog. This has to be done the right way. ~ Rob13Talk 03:20, 30 January 2017 (UTC)
You mean no different than Wikipediocracy? Doc James (talk · contribs · email) 03:54, 30 January 2017 (UTC)
Your reference to "non-public data" is kind of interesting,Rob. The recent statement by WMF Legal says The privacy policy applies to data collected by the Wikimedia Foundation, by Foundation partners, by some users in special roles (such as checkusers), or by some third parties who provide data to the Foundation. It does not apply to the publicly visible postings of users on the projects or to information collected from other websites and later posted to the projects. (bolding added)
It is ~just~ our own WP:OUTING policy (a very strong thing! but different from the WMF privacy policy) that applies to publicly available but off-wiki information, not the Privacy Policy.. at least according to WMF legal, who I take as authoritative on this matter. (I do not take what they write as authoritative with respect to the en-WP OUTING policy) Jytdog (talk) 04:56, 30 January 2017 (UTC)
Yes, I'm stating that we should treat it as non-public data and have an agreement to that effect. @Doc James: Much, much worse than Wikipediocracy. For obvious reasons, I will not name individual stalker blogs so as to avoid doing further damage to people's lives. If you want details, email me, and I'll explain in great detail just how big of a problem this is. ~ Rob13Talk 07:24, 30 January 2017 (UTC)
I am aware of some of them. Our outing policy no matter how strong we make it has no effect on their existence.
Creating a list of companies that say they edit Wikipedia and some of the blocked account they use seems like a very different activity. But hay that is just me.
That working to enforce the TOU by documenting UPE is "Much, much worse than Wikipediocracy." simply makes me scratch my head. Doc James (talk · contribs · email) 07:49, 30 January 2017 (UTC)
@Doc James: Under your proposal, what stops a sockmaster who I've blocked from coming back on a new legit-looking account, digging up some edit I made early on that's tangentially related to a college, finding some off-wiki evidence that I worked at that college as a student assistant, and using that as an excuse to launch an "investigation" including posting where I went to school, my name, where I currently work, etc. to make the "case" that one old tiny edit was a paid edit? And more importantly, once my name and where I'm work is posted for some time while people debate whether this investigation is appropriate, what stops someone with a grudge from looking up the address of where my office is and coming there with a gun? I don't know what your personal situation is, but for some of us, it takes very little information to figure out a place and time where we'll be. I'm absolutely panicking because no-one seems willing to consider how an overly-broad and ill-defined proposal is going to affect the lives of the editors who are more vulnerable to outing situations. Under current policy, such information would be removed in minutes because it would be a bright-line. Under an incredibly vague policy, a couple edits which technically meet the definition of a paid edit without any of the harm associated with paid editing becomes cause to keep that information live for unknown periods of time on one of the most visited websites in the world. How the hell do I keep editing under those conditions, let alone blocking socks, taking care of LTA cases, etc.? I average one death threat or serious threat of harm a month due to my administrative work on the project. I'm sure other administrators, especially arbitrators and functionaries, get more. How lucky do you feel that not one of our administrators will be targeted maliciously when you open up an on-wiki process to post personal information? ~ Rob13Talk 08:09, 30 January 2017 (UTC)
Unless you are running a paid Wikipedia editing business. And most of your edits are obviously paid in nature. And your account is indefinitely banned. Then the risk of having it linked to a known paid Wikipedia editing business such as like this[3] is astronomically low. Right around zero in fact.
They can already unfortunately post all those details on one of the many Wikipedia "hate sites". I do not look at them as it is simply too depressing. And they can start hate campaigns on twitter which twitter appears to be fine with. Or do so on their blog. Doc James (talk · contribs · email) 08:27, 30 January 2017 (UTC)
That doesn't have the audience of Wikipedia. The audience of Wikipedia includes every sockmaster I've ever blocked, including every single one who's sent me threats. You're arguing that you wouldn't investigate me for paid editing off of a couple perfectly neutral edits in a big pile. I agree with that. Are you happy to argue that one of the sockmaster's I've blocked wouldn't "investigate" me in that way, though? They'd be perfectly entitled to do so under the incredibly vague proposal above, at least for as long as it takes for some admins to discuss whether it's appropriate. There's no boundaries there. All the "unless" qualifiers you just added aren't in the proposal anywhere. It doesn't even specify this is restricted to LTA cases. There's no detail at all. That's unacceptable. ~ Rob13Talk 08:30, 30 January 2017 (UTC)
Google has the greatest user-ship of any site in the world and these sites are "on" google. I am sure all the sockmasters use google aswell. So by that logic it has a greater audience. They (these sock masters) investigate us already. I have seen the investigations :-( When it comes to this divide no solution will make everyone happy[4]. I will look at putting together a more concrete proposal but the current one has a super majority in support. Doc James (talk · contribs · email) 08:38, 30 January 2017 (UTC)

I have no problem with an appropriate process that targets paid editing and removes paid editors from the site, but I cannot continue editing in any administrative areas if that process is wide open to abuse and misuse. If all we were doing was listing PR company names with names of their socks, that would be a very different story. Right now, though, we have vague proposals about "investigating" and "releasing results on-wiki". There's no restrictions. There's no attempts to adjust this to where potential abuse is limited. There's nothing stopping someone acting maliciously from targeting anyone who's edited anything within their professional or personal interests among a large set of other edits. That's unacceptable. The safety and livelihoods of our editors are potentially on the line. We can't be vague. ~ Rob13Talk 08:14, 30 January 2017 (UTC)

As someone who has had law enforcement try to track them down for over three weeks to hand them a court case I am well aware that editing Wikipedia contains risks. (many thanks to the WMF for the defense by the way)
What we are doing is not some sort of game. It has real world consequences when we allow misleading content to exist on Wikipedia. And sometimes these consequences are felt by thousands.[5]
We have a responsibility to our readers as well as to our fellow editors. We need balanced nuanced rules. It is complicated and not a simple black and white. Doc James (talk · contribs · email) 08:34, 30 January 2017 (UTC)
Yes, and we can fix that without outing, using private investigations from a committee/group of editors designated to handle sensitive information. Or with limited outing that contains severe restrictions. Or with edit filters and machine-learning, two avenues I've been exploring for a couple weeks. It's entirely unnecessary to open things up like this with a broad proposal like this. In the meantime, I can't edit or take action in administrative areas with this spectre looming. ~ Rob13Talk 08:46, 30 January 2017 (UTC)
Have been looking at AI aswell over the last couple of years. This will take time and will take a group of Wikipedians seriously interested in the area to teach the machines and follow up on their work. Hard to accomplish with the controversy that hangs over the area. But I am sure we will muddle on the best we can. Legals has helped lighten the gloom a bit with their offer to help "if a controversial borderline case arises".[6] Doc James (talk · contribs · email) 08:52, 30 January 2017 (UTC)
I've been working locally on a related project, code that can tell the difference between text written by Wikipedians and text written by machines attempting to spoof humans (and those machines are getting smarter by the day). If other people get interested in the general subject of the impact of AI on Wikipedia, I'll be happy to share and collaborate. - Dank (push to talk) 17:34, 30 January 2017 (UTC)
what we've got here is bad faith humans trying to imitate good faith human contributors. Much of it is indeed mechanical, but the most damaging of the UPE work has been as sophisticated as many good faith editors--until you start looking at the actual references instead of assuming they support the article. As a suggestion ,it would be useful to have a program that would look at purported sources and estimate their likelihood of being just press releases. DGG ( talk ) 06:05, 31 January 2017 (UTC)
Agree User:DGG what saves us in the medical topic area from being overrun with UPE is WP:MEDRS. We have clear rules regarding what counts as a reliable source and we vigorously apply them. Press releases and press release like material just does not cut it. We need that in more topic areas. Doc James (talk · contribs · email) 01:35, 1 February 2017 (UTC)
Looking at the opening question of whether we actually do agree about the need for some sort of confidential area, here is one fish's assessment of the RfC responses so far: Milieu 3 is trending towards support, so there does seem to be general interest in some sort of confidential area. As for the rest, Milieu 1 is strongly opposed, emerging consensus against always putting top priority on WP:HA. Milieu 2 has no strong consensus either way, but trends towards opposition towards always posting evidence on-site. Taken together, they also seem to make discussion of confidential reporting worth examining more. Milieu 4 does not have a clear consensus, but trends towards support of some private investigation, and the differences in opinion mean that consensus for a confidential area cannot yet be assumed. Milieu 5 doesn't have much support. And Concrete proposal 1 suggests strong consensus in favor of linking onsite to the most egregious sorts of paid editing websites, in which case there would be less need for a confidential area. --Tryptofish (talk) 00:12, 2 February 2017 (UTC)
With this RFC up to 8 proposals and 34,000 words, I think that giving input based on a full analysis of it has gone above the level of most mere mortals. May I suggest that the result should be distilling a few 2-3 proposals out of this which are safely based on concepts which achieved widespread support in the above? Put them in new RFC, giving them the clear imprimatur that they arose from this process, possibly limiting the proposals to those. It sounds lengthy, but that's better than going nowhere. North8000 (talk) 15:24, 18 February 2017 (UTC).
A few of them here have a supermajority of support and can be close as is IMO. Doc James (talk · contribs · email) 15:28, 18 February 2017 (UTC)
Be careful with statements like that as there are lots of qualified supports/qualified opposes so just looking at the bolded words isn't going to give an accurate reflection of what consensus is. Thryduulf (talk) 16:14, 18 February 2017 (UTC)

(don't know where to place this, move it if you wish) Agree with the "common ground" statement in the title of this sub-section: we have to have a place where we can report evidence of abuse of the TOU when outing considerations (or OTRS confidentiality?) prevent us from doing so at WP:COIN. Justlettersandnumbers (talk) 19:26, 2 March 2017 (UTC)

Hi all,

Thank you to everyone for thoughtfully considering this issue and including us in the process. We want to offer a few thoughts on the discussion. We’d also like to apologize for any confusion that resulted from the original statement, which we’ve amended to address the points raised by the Arbitration Committee.

While we hope that our thoughts on this issue are helpful, we do want to emphasize that our statement is only guidance to assist the community with crafting their own policies.

We recognize the importance of balancing transparency within paid-editing investigations with respect for user privacy, in whatever way the community decides is best.

To be completely clear, posting personal information to harass or intimidate someone is prohibited in our Terms of Use, and runs counter to our movement’s values and vision. Harassment of any kind is not allowed under the Terms of Use and other Wikipedia policies, and we’re committed to ensuring that all of our policies (including restrictions on paid editing) are implemented in a manner that's consistent with a welcoming and educational atmosphere. We appreciate the Arbitration Committee's hard work and dedication to these issues, as well as everyone else who joins and has joined the discussion here to ensure we find a respectful and effective solution to undisclosed paid editing.

We also want to clarify that the self-disclosure requirements in the Terms of Use are limited to information that helps other editors identify the work of paid editors. If someone tried to publicly post extraneous, irrelevant personal information in a paid-editing investigation, we think that would be a clear case of harassment.

Finally, we want to help allay concerns related to editors who are affiliated with GLAM institutions. The paid-editing requirements in our Terms of Use only apply to people who receive payments for specific edits. This does not include experts such as GLAM participants or librarians who contribute to Wikipedia as part of their job, but are not compensated specifically for their contributions.

We apologize for anything in our previous statement that was unclear about our commitment to the safety and privacy of community members. As the community discusses the options in this RFC and other potential community solutions to preventing undisclosed paid editing, we want to reiterate that we are supportive and willing to actively provide help along the possibilities we outlined in our statement.

Thanks again to everyone working on these important issues. -Jrogers (WMF) (talk) 23:46, 3 March 2017 (UTC)

Re: This does not include experts such as GLAM participants or librarians who contribute to Wikipedia as part of their job, but are not compensated specifically for their contributions. -- I am not sure I ken it. A random PR suit may contribute to wikipedia, and it will be part of their job (to promote their business everywhere) and they are not compensated specifically per keystroke, so...? Staszek Lem (talk) 00:44, 17 March 2017 (UTC)
@Jrogers (WMF): Please clarify "who receive payments for specific edits" and "who contribute to Wikipedia as part of their job, but are not compensated specifically for their contributions". Under this wording it suggest that all paid editing is ok if the payment itself doesn't specifically require a specific wording or editing methods or even under some interpretations the payment must specify lay out min by min editing timetable to be against TOS. IMO. I would suggest there is not as fine a line between "paid editors" and "Wikipedians in residence" as implied. Endercase (talk) 18:01, 30 April 2017 (UTC)

Clarification on two particular points

There are two point we would also like to clarify regarding the applicability of the privacy policy and the concerns about legal risk in dealing with undisclosed paid editing issues. First, with regard to questions about the applicability of the Privacy Policy, we did our best to address them in the original statement. To clarify, the Privacy Policy defines personal information as information that the Wikimedia Foundation collects (including information collected with Foundation tools like Checkuser), and therefore would not apply to information posted publicly on the projects from other websites or third-party sources.

For the second clarification about how editors could be subject to liability, first, as I'm sure all of you know, we can't give legal advice on specific cases or specific examples under California law (obligatory legal disclaimer link). In general, though, we can say that investigating paid editing is not different from the perspective of legal issues than other activity on Wikipedia. Making reckless false accusations against anyone, whether it’s about paid editing, harassment, or in determining article content can be a problem under the laws that govern contributors as individuals. However, there's nothing special about the types of investigations around paid editing that make them a different sort of risk than other work editors have long done effectively on Wikipedia related to user behavior and content disputes. We'd also like to remind people that we do have the legal fees assistance program, specifically designed to help people who are subject to legal action because of their work supporting the projects. -Jrogers (WMF) (talk) 23:46, 3 March 2017 (UTC)

That is kind of interesting, User:Jrogers (WMF) but what it ignores is that the work that everyone has done in en-WP has been limited by OUTING which has in many ways limited the legal risks of libel etc. Loosening OUTING in some ways (as the WMF legal statement explicitly advised and as has been proposed above) would create new risks. Saying "there is nothing special" about operating in a context of loosened OUTING restrictions and new risk is not reasonable and I am really surprised to see you gloss over that. Do you see what I mean? Jytdog (talk) 05:26, 18 March 2017 (UTC)
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:Conflict_of_interest/Archive_27&oldid=1149826954"