Wikipedia talk:Harassment/Archive 8

Archive 5

Archive 6

Archive 7

Archive 8

Archive 9

Archive 10

→

Archive 15

Proposed OUTING revision

I'd like to propose the following revision to the "Posting of personal information" section of this policy page. (No change to the hatnotes or shortcuts.)

Here is a tl;dr of the changes: The proposed new version has 6 paragraphs. The substantive changes are in paragraph 3, "If you have accidentally posted...", where there is guidance about not getting outed, and in paragraph 5, "Posting links to or directions for...", about what was discussed in the archived discussion. In the latter, I really did pay attention to editors' concerns, and I think you'll actually find it an improvement in terms of COI investigations. For the other paragraphs, I rearranged material to make the progression more logical, and did some heavy copyediting. (The archived discussion really ends up being only a small part of what is here. I came to realize that the policy has become kind of jumbled, and really needs some polishing throughout.)

Existing policy: Posting another editor's personal information is harassment, unless that person had voluntarily posted his or her own information, or links to such information, on Wikipedia. Personal information includes legal name, date of birth, identification numbers, home or workplace address, job title and work organisation, telephone number, email address, other contact information, or photograph, whether any such information is accurate or not. Posting such information about another editor is an unjustifiable and uninvited invasion of privacy and may place that editor at risk of harm outside of their activities on Wikipedia. This applies to the personal information of both editors and non-editors. Any edit that "outs" someone must be reverted promptly, followed by a request for oversight to delete that edit from Wikipedia permanently. Any administrator may redact it pending oversight, even when the administrator has a conflict of interest. If an editor has previously posted their own personal information but later redacted it, it should not be repeated on Wikipedia, although references to still-existing, self-disclosed information is not considered outing. If the previously posted information has been removed by oversight, then repeating it on Wikipedia is considered outing. Posting links to other accounts on other websites is allowable on a case-by-case basis.[under discussion as of January 2016] The fact that an editor either has posted personal information or edits under their own name, making them easily identifiable through online searches, is not an excuse for posting the results of "opposition research". Dredging up their off-site opinions to be used to repeatedly challenge their edits can be a form of harassment, just as doing so regarding their past edits on other Wikipedia articles may be. However, if individuals have identified themselves without redacting or having it oversighted, such information can be used for discussions of conflict of interest (COI) in appropriate forums. If redacted or oversighted personally identifying material is important to the COI discussion, then it should be emailed privately to an administrator or arbitrator – but not repeated on Wikipedia: it will be sufficient to say that the editor in question has a COI and the information has been emailed to the appropriate administrative authority. Issues involving private personal information (of anyone) could also be referred by email to a member of the functionaries team. If you see an editor post personal information about another person, do not confirm or deny the accuracy of the information. Doing so would give the person posting the information and anyone else who saw the page feedback on the accuracy of the material. Do not treat incorrect attempts at outing any differently from correct attempts for the same reason. When reporting an attempted outing take care not to comment on the accuracy of the information. Outing should usually be described as "an attempted outing" or similar, to make it clear that the information may or may not be true, and it should be made clear to the users blocked for outing that the block log and notice does not confirm the information. Unless unintentional and non-malicious (for example, where Wikipedians know each other off-site and may inadvertently post personal information, such as using the other person's real name in discussions), attempted outing is grounds for an immediate block. Threats to out an editor will be treated as a personal attack and dealt with accordingly. Nothing in this policy prohibits the emailing of personal information about editors to individual administrators, functionaries, or arbitrators, or to the Wikimedia Foundation, when doing so is necessary to report violations of confidentiality-sensitive policies (such as conflict-of-interest or paid editing, harassment, or violations of the child-protection policy). Only the minimum information necessary should be conveyed and the minimum number of people contacted. Editors are warned, however, that the community has rejected the idea that editors should "investigate" each other. Posting such information on Wikipedia violates this policy.

Proposed revision: Posting another editor's personal information is harassment, unless that person has voluntarily posted his or her own information, or links to such information, on Wikipedia. Personal information includes legal name, date of birth, identification numbers, home or workplace address, job title and work organisation, telephone number, email address, other contact information, or photograph, whether any such information is accurate or not. Posting such information about another editor is an unjustifiable and uninvited invasion of privacy and may place that editor at risk of harm outside of their activities on Wikipedia. Unless unintentional and non-malicious (for example, where Wikipedians know each other off-site and may inadvertently post personal information, such as using the other person's real name in discussions), attempted outing is grounds for an immediate block. Any edit that "outs" someone must be reverted promptly, followed by a request for oversight to delete that edit from Wikipedia permanently. Any administrator may redact it pending oversight, even when the administrator has a conflict of interest. If the previously posted information has been removed by oversight, then repeating it on Wikipedia is considered outing. If you see an editor post personal information about another person, do not confirm or deny the accuracy of the information. Doing so would give the person posting the information and anyone else who saw the page feedback on the accuracy of the material. Do not treat incorrect attempts at outing any differently from correct attempts for the same reason. When reporting an attempted outing take care not to comment on the accuracy of the information. Outing should usually be described as "an attempted outing" or similar, to make it clear that the information may or may not be true, and it should be made clear that the block log and notice do not confirm the information. If you have accidentally posted anything that might lead to your being outed (including but not limited to inadvertently editing while logged out, which reveals your IP address, and thus, your approximate location), it is important that you act promptly to have the edit(s) oversighted. Do not otherwise draw attention to the information. If an editor has previously posted their own personal information but later redacted it, it should not be repeated on Wikipedia. References to still-existing, self-disclosed information is not considered outing, and so the failure of an editor to have the information redacted in a timely manner removes it from protection by this policy. Further information about protecting private information is at Personal security practices, On privacy, and How to not get outed on Wikipedia. The community has rejected the idea that editors should "investigate" each other. The fact that an editor has either posted personal information or edits under their own name, making them easily identifiable through online searches, is not an excuse for posting the results of "opposition research": the dredging up of editors' off-site opinions to be used repeatedly to challenge their edits in a content dispute or in a personal dispute. This can be a form of harassment, just as doing so regarding their past edits on other Wikipedia articles may be. Threats to out an editor will be treated as a personal attack and can also result in a block. Posting links to or directions for finding other websites where private information about an editor is displayed without that editor's consent, such as at doxing websites, is generally also a violation of the outing policy when information found there would have been considered outing on Wikipedia. It is never acceptable to use Wikipedia in any namespace for the revelation of any person's non-public personal information without consent, whether editors or anyone else. Posting private information of non-editors may also be a violation of the biographies of living persons policy. (However, simply linking to or referring to websites where doxing has occurred is not prohibited if doing so does not reveal private information.) Likewise, personal information posted at social networking websites may not be intended for general public dissemination. The prohibition against outing takes precedence over the guideline on conflict of interest (COI). However, on a case-by-case basis, posting links to information that is freely available to the public and which a person has, at any time, voluntarily made public is often permissible when doing so is necessary for investigating COI or undisclosed paid editing, particularly when the information has been disclosed in a commercial context and is not being used here to discredit another editor in a content or personal dispute. Nothing in this policy prohibits the private emailing of personal information about editors to individual administrators, functionaries, or arbitrators, or to the Wikimedia Foundation, when doing so is necessary to report violations of confidentiality-sensitive policies (such as conflict-of-interest or paid editing, harassment, or violations of the child-protection policy). Issues involving private personal information (of anyone) can also be referred by email to a member of the functionaries team. Only the minimum information necessary should be conveyed and the minimum number of people contacted. If redacted or oversighted personally identifying material is important to a COI discussion, then it should be emailed privately to an administrator or arbitrator, but not repeated on Wikipedia – it will be sufficient to say that the editor in question has a COI and that the information has been emailed to the appropriate administrative authority. However, if individuals have identified themselves without redacting it or having it oversighted, such information can be used for discussions of COI in appropriate forums.

What do editors think? (@Jbhunley: you asked me to ping you when I posted this.) --Tryptofish (talk) 23:49, 15 March 2016 (UTC)

I largely Support this version. I think it takes in a lot of the spirit from the previous RFC's. My only caveat is that I didn't see anything about threats to out being considered a personal attack. While I think spirit and intent still cover that area, sometimes it's worthwhile to state it explicitly. --Kyohyi (talk) 16:01, 22 March 2016 (UTC)

Thanks. There is a sentence about that at the end of the 4th proposed paragraph (paragraph beginning "The community has rejected..."). What I did was change the existing language so that it didn't say the wishy-washy "and dealt with accordingly". --Tryptofish (talk) 19:30, 22 March 2016 (UTC)

You are right, there it is. So no caveats at all on my end. --Kyohyi (talk) 20:26, 22 March 2016 (UTC)

I'd welcome hearing from any editors here whether you think it would be a good idea for me to open an RfC about whether or not to adopt the proposed change. I also would welcome any feedback about correcting or revising the proposal, before opening it up to an RfC. --Tryptofish (talk) 21:57, 23 March 2016 (UTC)

"The fact that an editor has either posted personal information or edits under their own name, making them easily identifiable through online searches, is not an excuse for posting the results of "opposition research". What is "opposition research"? Where is it defined, and what discussion has there been about it? --Ronz (talk) 22:29, 23 March 2016 (UTC)

That's old language that I did not change from the status quo. (It's the first sentence of the third paragraph of the existing policy, and the second sentence of the fourth paragraph in what I propose.) I've seen instances where two editors are involved in a content dispute, and one of them had looked around online to find out things about the other editor, and then posted in talk that the other editor had expressed opinions about the page subject off-site, and argued that this evidence should count towards giving less credence to that editor's opinions. That's what I would regard as "opposition research". As for the discussions that led to that sentence being put into the policy, I don't know the answer. --Tryptofish (talk) 22:37, 23 March 2016 (UTC)

I did some looking through the archives of this talk page, and the sentence that you ask about seems to have been first talked about at Wikipedia talk:Harassment/Archive 1#Proposal for "Posting_of_personal_information" section, in 2011. --Tryptofish (talk) 22:46, 23 March 2016 (UTC)

Thanks. Is it just me, or is the phrase confusing? --Ronz (talk) 01:22, 24 March 2016 (UTC)

I agree with you that the existing language suffers from being confusing and/or inconsistent. In the same edit that I am making in this reply to you, I'm also modifying the proposed new language in this regard. Does this change improve it? --Tryptofish (talk) 21:54, 24 March 2016 (UTC)

Yes, but the phrase remains. Unless it's used elsewhere, I don't know why it belongs here. I can't recall if anyone had dug up ArbCom cases that addressed harassment in detail, but I'd hope any such cases would help us to decide wording and emphasis. --Ronz (talk) 16:43, 25 March 2016 (UTC)

Here is another revision, in which I reorganize the sentences in order to present a definition for the term. I'm quite sure that this has been the intended meaning. How about this? --Tryptofish (talk) 22:08, 25 March 2016 (UTC)

Nice. That makes sense, a specific type of WP:BATTLE mentality and behavior. --Ronz (talk) 01:43, 26 March 2016 (UTC)

Tryptofish, what was the language of the "here is another revision"? I don't see any language here in the dialog. Please ping me. I'm interested, as i have been the subject of personal attacks based on opposition research that did not cease when i demanded them to, and then were justified on the basis of "fighting a battleground mentality" essentially by the other party, which i wholly dispute and know was very very wrong. It was used to poison the well repeatedly and to slander me with falsities that were not even correct in disputes about controversial content. So i have skin in the game on this issue and i don't want to see it written wrong, and i recognize the usernames of some people advocating for a position here as being people who were problematic in the past in such sorts of disputes, and for them alone to be influencing this meta debate would be unfortunate. Thanks for help if you can provide it. To be clear, my position is that dialog about content must be about the content and not about the editor, and that opposition research and repeated outing should be bannable offenses. SageRad (talk) 12:38, 11 May 2016 (UTC)

The only person who even came close to OUTING you (and didn't) is Guy, who has not commented on this proposal, at all. Jytdog (talk) 23:28, 11 May 2016 (UTC)

@Jytdog: That was really unnecessary, and unhelpful. @SageRad: Hi, Sage. Here is the diff of the edit that I made just before where you placed your question to me, and you can see the language I was referring to there: [1]. There were several consecutive edits, which you can see by going forward and back from that diff in the talk page history. And, of course, you can see the most current language just above. --Tryptofish (talk) 00:36, 12 May 2016 (UTC)

@Tryptofish: Thanks, Trypto. That language generally does make sense to me. I hope it's enforced more often than it seems to be, so that people really do focus on the actual content questions without going ad hominem and McCarthyist. Thanks. SageRad (talk) 12:08, 12 May 2016 (UTC)

I'm trying to decide whether or not I should just go ahead and make this change to the page. So far, the comments here have been very supportive. Do any other editors have concerns, objections, or corrections? Would anyone like me to open an RfC about whether or not to make the change? --Tryptofish (talk) 21:30, 28 March 2016 (UTC)

I remain opposed to the changes, particulary the removal of "Posting links to other accounts on other websites is allowable on a case-by-case basis," which has consensus. SarahSV ^(talk) 05:00, 29 March 2016 (UTC)

It sounds like you are opposing things that I proposed in the archived discussions, but that are not proposed here. The sentence you quote has not been removed. It has been reworded as "However, on a case-by-case basis, posting links to information that is freely available to the public and which a person has, at any time, voluntarily made public is often permissible when investigating COI or paid editing, and may not necessarily be regarded as opposition research in this context." How is that change contrary to existing consensus, and how is it inferior? --Tryptofish (talk) 19:03, 29 March 2016 (UTC)

I oppose the changes because the paragraph beginning with "Posting links to..." has changed substantively, which I don't agree with. I will give a recent example. In this diff on the Arb Noticeboard, I mentioned that certain discussions took place on Wikipediocracy, which I'm not explicitly linking to. Would such an edit be prohibited, because I'm giving "directions" as to where to find the information? Kingsindian ♝ ♚ 00:22, 2 April 2016 (UTC)

@Kingsindian: I've been following the discussion that you linked to, and the example that you give in your diff caught my attention in just this regard, because it goes to 1/2 of exactly why that paragraph or something like it is desperately needed. Although I haven't followed those threads at Wikipediocracy, I'm pretty sure that they don't out TDA or Cla, and that's probably why your edit was not a problem. The proposal is not meant to say that no one can mention websites like that, and if that is unclear it should be fixed. The intended meaning is that it would have been outing if the discussions you referred to contained outing information about another editor. But there is another 1/2 that you may not be aware of. I made an edit on an ArbCom page, with thinking very similar to yours. I said in a general way that discussions were at other websites, and I was thinking of Wikipediocracy, but I did not even mention Wikipediocracy by name, so I actually posted less than you did in that regard. But in my case, I got blocked for it by ArbCom. We have a problem with inconsistency and lack of clarity. --Tryptofish (talk) 20:28, 2 April 2016 (UTC)

Also, I'm curious: would you support this proposal with that one paragraph deleted? --Tryptofish (talk) 20:38, 2 April 2016 (UTC)

@Tryptofish: I am all for clarity and consistency: but the rule should be least intrusive so that neither your edit nor mine should be prohibited. I don't support any amendment which prohibits both my edit and yours, in the name of consistency. By they way, since the name of the victim in question was linked in the Wikipediocracy thread, and was oversighted from Wikipedia due to BLP concerns, I think a case could be made that me linking that thread would result in a BLP violation. As to your latter question, I haven't read the rest of the proposal carefully, but my impression is that it is ok. Kingsindian ♝ ♚ 21:13, 2 April 2016 (UTC)

That's certainly reasonable, but I guess consistency feels differently for the user who was blocked, and for the user who wasn't. --Tryptofish (talk) 21:31, 2 April 2016 (UTC)

Where do we stand now? Are there still objections? --Tryptofish (talk) 23:08, 30 March 2016 (UTC)

This discussion has been going since 28 November 2015 and I had to give up following it. It is likely that others have also been driven away. Why is there a need to change this extremely delicate policy? It is very hard to predict how such a major adjustment to the text would work. The change might have a phrase which could result in the block of a good editor trying to repel an SPA with a major COI who is using Wikipedia to promote something. Or, another phrase might allow undue harassment of a good editor who has mentioned their identity once or twice in the past, and who now is being pursued by misguided opponents. It would only be worth studying the proposed change if there is a clear reason that a change is required. Johnuniq (talk) 23:42, 30 March 2016 (UTC)

In other words, some of the editors who watchlist this page have decided that they oppose because I am the editor who proposed it, and they have not even read it. Fine. In another day or so, I will make it an RfC, and get fresh eyes. --Tryptofish (talk) 23:54, 30 March 2016 (UTC)

I think that really stinks, and I do indeed intend to open a proper RfC in the near future. But I think that it is useful to demonstrate that there really are multiple things that are sub-optimal about the current wording. To say that the current wording is so delicate that any change will be a step backwards is downright silly. So here is a list of some of the gems in the current wording that the proposed change would fix:

1. "This applies to the personal information of both editors and non-editors." Added ages ago by Jimbo himself, and can anyone actually point to an outing of someone who never edited Wikipedia (as opposed to BLP violations)?
2. "Posting links to other accounts on other websites is allowable on a case-by-case basis.[under discussion as of January 2016]" Oh, yeah, Wikipedia's future depends upon keeping that as vague as possible!
3. "Threats to out an editor will be treated as a personal attack and dealt with accordingly." Yes indeed, we would not want to deal with them inappropriately.

To say that these examples are sacrosanct is ridiculous. To reject changes without attempting to read the changes is, well, WP:OWN. I have no problem with editors feeling like they are not interested and do not want to go to the effort of examining a proposal. That's fine. But it's not a reason to actively oppose. If you don't want to spend time thinking about something, then don't – but don't claim that revisions are forbidden. I really listened to what editors said in earlier discussions, and I am not simply proposing the same thing again and again. If you can bother to actually read it, the proposal is very responsive to what editors said were their concerns. And it's a lot better written. And what exactly is wrong with including advice to inexperienced editors about how not to get outed? --Tryptofish (talk) 00:54, 31 March 2016 (UTC)

Hmmm... This does take care of the potential problems dealing with COI/UPE the other proposal made but it looks to swing very far the other way. It does comply with the Foundation's Privacy Policy as described by WMF legal (@Tryptofish: can you please add the diff of legal's response to your earlier example for new readers? I can not locate it.) but I think it goes way beyond what the community would accept. A couple of the thing where I see trouble are: "editors' off-site opinions to be used repeatedly to challenge their edits - loophole with repeatedly; " websites where private information about an editor is displayed without that editor's consent" opens up social networking sites which have been previously off limits; "However, on a case-by-case basis, posting links to information that is freely available to the public and which a person has, at any time, voluntarily made public is often permissible when investigating COI or paid editing, and may not necessarily be regarded as opposition research in this context" does a good job of limiting the case-by-case to COI/UPE, which I like, but with with the other two exceptions is not really needed and contradicts "The prohibition against outing takes precedence over the policy on conflict of interest (COI)".

While I think undisclosed paid editing is a huge problem here and we need tools and policies which make it easier to handle, it is mostly a problem in "commercial" articles (Companies, bands, autobiographies and articles which are, in general subject to PR/SEO manipulation.). The other side of the coin are editors who edit in controversial topic areas where this opens both the advocates and their opponents up to outing and harassment based on the loopholes where this allows links private information that is displayed on other sites with the editors consent - for instance all editors in Category:Wikipedian amateur radio operators are required by law to have their names and residential addresses in publicly accessible national databases. This would make every one of those editors who post their call sign or use it as their username unOUTable because mapping call sign to name and address is trivial and public. The same for many business owners and professionals who are members of professional societies or licencing bodies. Even normal people who have blogs or use social media are open to this loophole.

No matter what I think changes to this policy requires a widely published RfC. The balance between the strong community values against COI/UPE and for privacy/anonymity is core to the character of the project. COI/UPE by definition compromise NPOV and thereby the value of Wikipedia to its readers so, on the one hand rules that make investigating it are good. On the other, the ability of editors to remain anonymous protects them from outside threats and from their ideological opponents who may carry over their harassment into RL. Strong rules against OUTING allows people to edit in high controversy articles, allows wider opinions and this increases the value of Wikipedia to its readers. Right now there is a, mostly, functional balance between these competing values in the OUTING policy. The existing case-by-case exception combined with the existing strong rules against OUTING gives the flexibility to maintain the balance, albeit with a loss of consistency of enforcement, that a policy which tries to lay out all possible options and permutations. Jbh^Talk 14:14, 31 March 2016 (UTC)

@Jbhunley: Thank you very much for this careful and thoughtful analysis. I fully agree with you that an RfC is going to be necessary, so I'm first going to work on it here, in what I am thinking of as a feedback-and-revise phase, and then I'll make it an appropriately advertised community RfC. Also, with respect to the various issues you raise about whether some things go too far towards supporting COI investigation, an RfC will be the ideal way to make the eventual outcome consistent with present day community opinion. (And in a purely personal opinion, I want to gloat over the fact that, after the WP:OWNers of this policy page got snippy with me over their certainty that I was merely proposing what I had proposed previously, you actually read what is here and concluded that perhaps I altered my original opinion too much in trying to accommodate their concerns.) Now, about your specific points:

1. Here is the diff of the response from WMF legal: [2]. I think the proposal is quite consistent with that comment.
2. About "repeatedly": That's actually the existing language on the page. I did not change that.
3. Social networking sites: Yes, I'll make a revision to fix that.
4. About COI etc.: I'll give this some careful thought, but I'm leaning towards presenting it to the community and letting the community decide how it should ultimately be calibrated. I don't have a problem with having some logical (not literal) redundancy in the interest of clarity. As for inconsistency about how OUT takes precedence over COI, true, but there is less inconsistently after this revision than with the completely unexplained language about "case-by-case" in the current language. I'm keeping the phrase "case-by-case", but trying to locate it in a clearer context. Put another way, acceptable "cases" would rarely if ever include linking to doxing (or social networking) websites. The page now leaves that completely unclear.

Thanks again. --Tryptofish (talk) 23:07, 31 March 2016 (UTC)

• Here's a quibble. This says "the policy on conflict of interest (COI)" but it's a guideline, not a policy... Jytdog (talk) 15:58, 31 March 2016 (UTC)

Thanks! That's a very good catch. After I finish facepalming myself for making that dumb mistake, I'll fix it. --Tryptofish (talk) 22:41, 31 March 2016 (UTC)

In this edit, I made revisions according to the feedback from the two editors above. Is this OK? --Tryptofish (talk) 19:44, 1 April 2016 (UTC)

And added a bit more. --Tryptofish (talk) 19:55, 1 April 2016 (UTC)

• I think this is good to go, yes. At some point I would like to try to elaborate on the relationship between OUTING and COI work but would prefer to do that in a more targeted way; this is a big chunk of change and I don't want to derail the months that have gone into this. This, is good. Jytdog (talk) 23:14, 1 April 2016 (UTC)
• This is a good version to put up for RfC. One posibility to consider, at a later date, is when investigating UPE have the OUTING policy make the Foundation's Privacy Policy is the controling policy rather than OUTING. This would be justified by UPE being a Terms of Use violation which is governed by WMF rules rather than simple COI which is an en.wp matter. This is just a blue sky suggestion, I have not thought about it in depth but I wanted to mention it before I forgot about it. Jbh^Talk 00:30, 2 April 2016 (UTC)

Jbh interesting point but my understanding is that unless a community explicitly adopts a WMF policy it isn't enforceable by the community DR mechanisms but rather only by WMF. Arbcom has stated that about the paid editing piece of the ToU for example. Jytdog (talk) 00:37, 2 April 2016 (UTC)

Good point. I re-worded it a bit above. What I was thinking is when we discuss in more depth the relation between COI/UPE and OUTING it may be worthwhile to consider changing the en.wp policy to make the WMF policy controling in the case of UPE not that there is an argument it is controlling now -- I just wrote something that failed to say what I meant. Thinking about it a bit more having two tiers of privacy would lead to more drama and problems than it resolves. Jbh^Talk 01:08, 2 April 2016 (UTC)

Thanks both of you for the feedback. I'm still interested in considering further input from other editors before proposing something in an RfC. About WMF terms, it seems to me (and this was sort of confirmed in the reply from WMF legal mentioned above) that those terms allow somewhat more divulging of private information than what the community here has typically been comfortable with. I think that the outing policy is actually more restrictive than the minimum that the WMF requires. But in any case, I would not want any of that to be part of the discussion of this proposal. --Tryptofish (talk) 20:36, 2 April 2016 (UTC)

Based on the feedback from Kingsindian, and the general sense that I have gotten throughout all of the present and past discussions, I am wondering about whether any RfC to the community should treat the sentence Posting links to or directions for finding other websites where private information about an editor is displayed without that editor's consent, such as at doxing websites, is generally also a violation of the outing policy; likewise, personal information posted at social networking websites may not be intended for general public dissemination differently, that it should either be revised further before the RfC, or should be presented separately from the rest of the paragraph that it starts in the RfC – by which I mean that the RfC would offer 3 choices: the status quo, the proposal with the sentence, and the proposal without the sentence. Or: should we propose that what is in that sentence is not a violation? What do editors think? --Tryptofish (talk) 21:22, 2 April 2016 (UTC)

In the same edit as this comment, I made some adjustments to that paragraph. Does this work? I am very interested in any feedback. --Tryptofish (talk) 21:16, 3 April 2016 (UTC)

I am a bit concerned about the modification of This applies to the personal information of both editors and non-editors to "Revealing private information of non-editors may be a violation of the biographies of living persons policy". I think the part about non-editors should stay there if only to close a loophole where someone may post the information of a user registered on a different website. For example, User:A might post the personal information of User:B, (a registered user on another website) who is not a Wikipedia editor and neither has a Wikipedia article about him/herself. --Lemongirl942 (talk) 03:31, 11 April 2016 (UTC)

Applying OUTING to non-editors is unacceptable, because it will make it impossible to determine whether BLP subjects are notable, whether certain BLPs are hoaxes or not, whether many sources are reliable...in other words this makes a vast swathe of content permanently unverifiable. It's a terrible idea, and well outside of what the scope of the Harassment policy was intended to apply to (conduct between Wikipedia editors). The protection of non-editors already falls under the BLP policies and should be discussed there, instead of forking it across two different policy pages. Geogene (talk) 20:12, 11 April 2016 (UTC)

Again, thanks to you both for the comments. It seems to me that anything about a non-editor that an editor puts into an article would be automatically covered by BLP, and it is strange to describe that as "outing". I guess the issue would be if, on a talk page or noticeboard instead of in an article, an editor posts something identifying someone who is unconnected to Wikipedia, who is not written about in page content, but who is a participant at some other website. It's a strange thing to do, and it seems improbable enough that it might be instruction creep to carve out a policy component to anticipate it. Let's say that some other website like a social networking site has something resembling an anti-outing policy of its own, and a Wikipedia editor makes a post on a talk page saying that "screen name" at "web site" is really "real name" for someone who has nothing to do with Wikipedia and is not notable. Is that something that happens? Is it something where Wikipedia has a responsibility to that other website? Can't it be covered simply by contacting the oversighters? --Tryptofish (talk) 23:59, 11 April 2016 (UTC)

Hi, thank you for your responses. Let me clarify that my point was not particularly about "Outing". I am just concerned about the posting of non-public personal information of any individual (which is not allowed according to this policy in its present form). It is entirely plausible that someone could post another person's non-public personal information (in any namespace), and it could be used for doxxing. For example, User:A posts the phone number of his neighbour (who may not even be using the internet) with the intention of harassing him. While WP:OVERSIGHT can remove the posted information, the current policy seems to be the only one which prohibits posting such information. (I may be wrong about the last sentence however, and I would be glad if someone could point me if another policy addresses this issue). --Lemongirl942 (talk) 03:11, 12 April 2016 (UTC)

Based on what you just said, I got an idea, and in the same edit as my comment here, I added a sentence about doxing. Does this address your concerns? --Tryptofish (talk) 19:37, 12 April 2016 (UTC)

@Tryptofish: Oops, sorry for the late reply. Would it be better to write it as It is never acceptable to use Wikipedia for the posting of someone's non-public personal information. I guess the term "non-public personal information" is a better term and follows from WP:OS and RFO. Thank you. --Lemongirl942 (talk) 18:56, 28 April 2016 (UTC)

That's a very good suggestion, thanks. Done, with a few further modifications. And no problem with some time having passed – after all, I've been distracted by other things from the RfC plan below, myself. --Tryptofish (talk) 23:41, 28 April 2016 (UTC)

Idea about 2 RfCs

I've been thinking very hard about the points I was asking about directly above, about the "Posting links to..." paragraph, and I got an idea that I think answers my own question. I think that an RfC about whether or not to adopt the proposed version will get stuck on that paragraph, no matter how much the rest of the proposal is an improvement. I think the solution is to establish a consensus on the policy substance first, and only seek consensus on wording after. So I am going to temporarily put the proposal above on hold (although continued discussion is welcome, of course). First, I want the community to determine what that paragraph will eventually say.

I'm planning to start an RfC (appropriately advertised to the community) first, in which I will ask about each of the "situations" that are potentially covered in that paragraph. It won't deal with any specific wording for the policy page. Instead, I will have a numbered list of things that potentially should be considered either outing or permissible. And I will ask that editors indicate which ones they think should usually be considered violations, and which ones they think should usually be considered OK. I'll ask that they indicate their own opinion instead of trying to guess what the community wants.

As a very rough draft, the list might be something like:

1. An editor investigating COI or UPE links to a doxing website where the user being investigated is outed.
2. An editor investigating COI or UPE says that there is a doxing website where the user being investigated is outed, but does not post a link to it.
3. An editor mentions, in any context, websites like Wikipediocracy, but not in a way that outs users here.
4. An editor investigating COI or UPE links to a social networking website where the user being investigated is identified.
5. An editor investigating COI or UPE links to a blog where the user being investigated is identified.
6. An editor investigating COI or UPE links to a company website where the user being investigated has self-identified.
7. An editor investigating COI or UPE links to a company website where the user being investigated previously self-identified, but subsequently took down the information.
8. An editor who has a content dispute or a personal dispute with another user links to a company website where that user has self-identified.

Something like that. And I would welcome suggested additions to the list. After the RfC, we would have a present-day consensus that some of these things are usually violations of the outing policy, and some should usually be permissible (and perhaps no consensus on a few things). Only then, I would revise the proposal above, so that it would reflect the consensus of that RfC. And that would render objections to the paragraph, based on the contention that it fails to reflect consensus, invalid.

@Jytdog: you recently told me at my talk page that you were thinking about some additional possible changes to the policy. I think that this could potentially be quite appropriate to ask about in my suggested "first" RfC. So if you, or anyone else, would like to add them to the list, please propose that here. --Tryptofish (talk) 23:24, 4 April 2016 (UTC)

So sorry for the delay! here are my thoughts.

First, all of those are inappropriate, and some of them might be appropriate, in the right context, which is addressing a concern with another user's behavior per WP:DR - so first, in a civil conversation on the user's talk page, and then at WP:COIN or WP:ANI, in the context of addressing the user's behavior. Deploying any of this in an article Talk page content dispute or in some side conversations is all inappropriate.

Second, it has to be based on a provable-with-diffs pattern of editing behavior - it has to start with content. If there is no valid "hook" in content any of this should be sanctionable.

Third, in order to bring in any external evidence, there needs to be an on-wiki "hook" as well. It could be the user's username (clearly their RW name or an abbreviation, like "Donna Summer" or "Dsummer1234", or Bascomcomm (realworld, recent example, which was an account from Bascom Communications that was in the news). Or an explicit disclosure in an edit note or Talk page note, like "I work for this company and I should know", or more rarely, "Please see this (with a link) blog post I wrote." There is ~interesting~ and hard stuff here, like an account at Upwork claiming to have edited X article in WP for pay, at around the same time that WikiuserY (different name) worked that article over.

(!) In my view, putting a frame on it expressing that stuff will increase helpful feedback and decrease noise. So something like: "In the context of a civil effort to address another editor's potential COI at the user's talk page, COIN, or ANI, and based on a prove-able pattern of POV editing, and based on there being some on-Wiki "hook" that points to an off-wiki source, which of the following....." would be helpful. Also adding something like - "some of these are pretty obvious violations out of OUTING in any context; the purpose of this RfC is to get a sense of where the community draws the line." would also reduce noise, I think. :)

What do you think? TMI maybe? Jytdog (talk)

I agree with everything that you said, and the feedback is very helpful, thanks. I'll give some careful thought to making those various distinctions clear without making the RfC wording too convoluted, and I'm pretty confident that I can do that. The list above is only preliminary, anyway. I asked you in particular because you had previously told me that you were interested in raising some new issues after the previous kind of RfC that I had planned (the RfC about the proposed revision). So I wanted to know if there would be any other kinds of situation that you wanted me to add now, along those lines. Or maybe what you said about "hooks" etc. was it? If there are any other kinds of situations that you would like me to add, please let me know. Otherwise, I'll just go ahead. Thanks again. --Tryptofish (talk) 18:01, 10 April 2016 (UTC)

The only other interesting situations you might want to list are:

• An editor links to pages at a freelancing marketplace site relevant to the activity of a Wikipedia user (e.g. someone has offered a paid job to edit X and wikipediauserY does that)
• An editor links to and cites domain registration information for a website cited by a Wikipedia editor multiple times

other than that, good to go I think! Jytdog (talk) 02:22, 11 April 2016 (UTC)

I agree. Good, thanks. --Tryptofish (talk) 00:04, 12 April 2016 (UTC)

I've stumbled across this conversation and believe the discussion around the RfC will be useful in moving the ball forward regarding building a community of integrity and mutual respect. Looking forward to the wider discussion. Just a brief note, some people will probably not be favorable toward any policy change that appears to lessen anonymity, some people aggressively favor expanded investigations, and the majority are in the middle. Addressing that majority, there should be language that clearly establishes this is not a Get Out of Jail Free card for bad actors; everything has to be grounded on GF reasons to bring forth off-wiki evidence, and there should be consequences for not doing so. - Brianhe (talk) 02:44, 11 April 2016 (UTC)

Thanks so much for those comments, and I'm glad that you see value in working on this. I agree with you. Let me ask: does the proposed revision above need to have better language about the need for good faith reasons? --Tryptofish (talk) 00:04, 12 April 2016 (UTC)

The most salient passage in the proposal seems to be "...posting links to information ... when investigating COI or paid editing" which could be tightened up but I think it's OK to start discussion with what you've already developed. If I were to rewrite that passage it might be something like "...posting links to information ... in a bona fide investigation of COI or paid editing". It's all going to be subject to interpretation of GF when future investigations are underway, so there's not much to be gained by being over-prescriptive now. - Brianhe (talk) 02:35, 12 April 2016 (UTC)

In the same edit as my comment here, I added a bit of language to that effect, about it having to be necessary for COI or PE investigation. I hope that helps tighten it up. --Tryptofish (talk) 19:42, 12 April 2016 (UTC)

My personal concern is that the current version which contains:

• However, if individuals have identified themselves without redacting or having it oversighted, such information can be used for discussions of conflict of interest (COI) in appropriate forums.

Is not rectified by this version:

• However, if individuals have identified themselves without redacting it or having it oversighted, such information can be used for discussions of COI in appropriate forums.

The net-result is that editors who are new to Wikipedia, and don't realize that self-disclosures can (and will) be against them "in a court of law" (i.e. on Wikipedia in COI discussions), places experienced users at an advantage through their ability to hide their own identity, while using the self-disclosures against new editors. DroitInternationale (talk) 06:19, 4 May 2016 (UTC)

I agree with you that this is the case, but I do not see a good way to legislate against it (beyond WP:BITE). There are many other ways in which experience with editing confers insight into how to edit safely, and where new editors are inevitably at a disadvantage. It's not something specific to that sentence. I think that the proposed revision does a good service by adding a paragraph of advice about how not to get outed, something that is currently lacking on the page. But I also think that investigating COI is increasingly becoming something of great importance, and I do not think that an editor investigating it should be prevented from citing information that is voluntarily posted on-Wiki. But an RfC will be useful to determine what the community as a whole thinks about it. --Tryptofish (talk) 17:36, 4 May 2016 (UTC)

User:Ebonelm

I have noticed that Ebonelm tends to follow newer editors in an attempt to discourage them from editing on Wikipedia. Is there any course of action I could take? Antamajnoon (talk) 10:30, 30 May 2016 (UTC)

@Antamajnoon:, it would be great if you could provide some evidence for your accussations. That we have had content disputes on two pages hardly ammounts to harassment and in both instances I have been demonstrated to be on the side of the consensus and your edits have been rejected. Ebonelm (talk) 12:18, 30 May 2016 (UTC)

This is the wrong venue to discuss behavioral issues with specific users. Please open a thread at WP:ANI Jbh^Talk 12:41, 30 May 2016 (UTC)

Ongoing discussion at Meta

See meta:Grants talk:IdeaLab/Propose Wikimedia Code of Conduct (adapted from open source Contributor Covenant) Jbh^Talk 13:47, 7 June 2016 (UTC)

Definition of "private information"

We need to work on getting a common understanding of what we mean when we say "private information".

1. Some appear to be using the term to mean "all information about someone that has not previously been published on WP regardless of whether or not they are a WPian". It is unclear if this covers both self published and information published by others about the person.
2. The WMF uses the term to mean information provided to the WMF that is not public

Do other have other definitions they wish to see used? I do not see public job postings to edit WP as "private information". This appears to be in conflict with the first definition but not the second. Doc James (talk · contribs · email) 06:47, 4 July 2016 (UTC)

From the WMF's standpoint, 'private' means 'private data'. So, the IP address you use as a registered user, anything as a result of checkuser and so on. Information that is private and not publically accessible. The general wikipedian usage is 'anything that has not been disclosed on EN-Wikipedia by the user, regardless of its non-private nature elsewhere' - this courtesy is ONLY extended to wikipedians, if someone isnt a wikipedian, its fair game to shove as much info as they can slip past the BLP policy. Only in death does duty end (talk) 11:02, 4 July 2016 (UTC)

2- Private data as defined by WMF.

1- Personal information or off-site links about any editor, except yourself(oh-ugh/last add/edit again), unless editor has self-identified on or off-Wikipedia. (if an editor is also the subject of an article, BLP applies (edit/add)-and normal in-article/talk page edits are allowed when dealing with article content,(another edit/add)- but NOT editor identity unless editor has self-identified?)--(not sure how it applies to editors who are also the subject of an article, but maybe something about); Unless the editor has acknowledged on Wikipedia, that they are also the subject of an article, we do not "out" them explicitly, but we can ask to confirm or deny a COI. TeeVeeed (talk) 15:34, 4 July 2016 (UTC) edit a few minutes later.TeeVeeed (talk) 15:44, 4 July 2016 (UTC) OK last edit hereTeeVeeed (talk) 15:46, 4 July 2016 (UTC)--last edit again.OK final sat edit here I promiseTeeVeeed (talk) 16:02, 4 July 2016 (UTC)

Here, I think the distinction is clear: If an editor does not self-identify on-wiki, DO NOT OUT THEM. If you think someone is a paid editor or otherwise breaching policy, you can inquire if they have a "connection to article foo," but not "are you person foo?" that's what private email to ArbCom, CU or oversight is for. I think that's an important distinction. Montanabw^(talk) 21:13, 4 July 2016 (UTC)

It seems to me that it comes down to personal information about an editor, that the editor has not voluntarily disclosed on Wikipedia, and where there is no indication that the editor wants the information to be public. Also, if one looks near the middle of #Posting of personal information, above, Fluffernutter presents a rough draft of a flow chart that I think could be developed into a useful tool for distinguishing what kinds of information are, and are not, considered private by this policy. --Tryptofish (talk) 21:16, 4 July 2016 (UTC)

The question is not specifically about "editors", it is about other accounts on other websites. Of course a undisclosed paid editing company would not want links to them posted on Wikipedia as that will harm their business. They want the information "public" in that they have placed it on Elance or another site. Doc James (talk · contribs · email) 21:33, 4 July 2016 (UTC)

I consider advertisements, such as at Elance, to be public information. There is a difference between being private, and just not wanting to get caught. But even if you do not intend the question to be about editors, as opposed to about information, I believe that a correct answer needs to be formulated in terms of editor intent. --Tryptofish (talk) 21:40, 4 July 2016 (UTC)

To answer [[User:Doc James, OH. well I think that is a different thing? Posting this information on Wikipedia? That would have to be done without outing any editors, and I don't think it really should be included in the policy about personal/private and identifying info. But as a different question, It could make it appear as-if linking off-site with specific "accounts", is "OK"? It could probably be done very carefully to avoid adding Original Research directly to the project, yet obviously encouraging OR, (which is normally encouraged). And if it is allowed with no outing, no identification of editors explicitly or even any hint of implication, I don't think it is a huge technical break of no outing, just somewhat unsavory and tasteless, and free advertising for Elance. I think that I would defer to other editors opinions on this, but still wanting to see it done or not done with some integrity here.TeeVeeed (talk) 22:28, 4 July 2016 (UTC)

Posting of personal information

I made a slight modification to the following sentence:

Posting links to other accounts on other websites is allowable on a case-by-case basis.

And have changed it to:

Posting links to other accounts on other websites is not permitted, unless under exceptional circumstances.

If there's no clear consensus for removing the entire sentence, please let the second option stay - It's less ambiguous and doesn't leave room for creative interpretation of "case-by-case basis".

RoseL2P (talk) 21:04, 1 July 2016 (UTC)

That's pretty much saying the same thing, and changing it in the middle of an RfC is only complicating matters.- MrX 21:26, 1 July 2016 (UTC)

And whatever else may be the case, it really is best to discuss this stuff here in talk, and not for anyone to edit war on the policy page (that's directed at everyone, and at no one in particular). --Tryptofish (talk) 21:31, 1 July 2016 (UTC)

Also "under exceptional circumstances" is just as meaningless as "case-by-case". --Tryptofish (talk) 21:39, 1 July 2016 (UTC)

Sorry about that, I just felt it necessary to make it clear that any posting of personal information (unless already published on WP) is usually a form of harassment. I didn't realize my edit (a very minor one) would be controversial but if someone feels strongly against these changes, please feel free to restore the original wording. RoseL2P (talk) 21:51, 1 July 2016 (UTC)

I think RoseL2P's wording is better--it clarifies that this is rare and that the presumption is that it is not permitted. However, it still is vague, and doesn't really solve the problem. DGG ( talk ) 22:26, 1 July 2016 (UTC)

So now that pubmed contains comments by accounts are we now no longer allowed to link to it? For example this paper contains "personal information" in the way some are interpreting it http://www.ncbi.nlm.nih.gov/pubmed/24778001/#comments

If this is the case we at WikiProject Medicine need to look at forking as we would no longer be welcome here. Doc James (talk · contribs · email) 05:40, 2 July 2016 (UTC)

With respect, I think most editors would see this example as a straw man. -- Euryalus (talk) 07:29, 2 July 2016 (UTC)

Doc James, your argument is fallacious because it refutes a position that none of us will support. Many people (rightfully) disagree with the posting of personal information in the following manner:

According to my opposition research, you have previously commented on PubMed [insert link] under the following name: [insert name].

This should not be allowed because it will result in someone being "outed". If the link was cited without reference to anyone in particular, no one will be "outed" by it. Since nobody will oppose the citing of such links for genuine encyclopedic purposes (i.e. without attempting to harass or intimidate), you are esentially refuting an unsupported position.

That is why your argument is fallacious.

RoseL2P (talk) 09:49, 2 July 2016 (UTC)

While the question you post is not the one asked in this RfC. The question asked here is "Can other site accounts ever be linked to" and right now we have a fairly large group of functionaries saying no other account can ever be linked to. From my reading zero exceptions are to be given. This would means that no links to the NYT as comments at the bottom contain "other site accounts" and pubmed may not be linked to either for the same reason.

• User:Thryduulf writes ""Posting another editor's personal information is harassment, unless that person had voluntarily posted his or her own information, or links to such information, on Wikipedia.". That includes (links to) accounts on other websites that are not disclosed on Wikipedia." So if we see a few hundred comments on a NYTs article and one of those comments is by an editor here than that link appears to be unsafe to use. In our example for this RfC the comment can even occur after the link is used on WP.
• User:A fluffernutter is a sandwich! writes "There's nothing good that comes with leading people to believe they can post links that contain private/personal information about other people, because we know based on multiple other policies (and based on the oversight logs) that that's a lie - those things will be suppressed, almost invariably - and the people who relied on the misleading policy are going to be really confused when they do get in trouble for it." In this statement the definition of "private information / personal information" appears to be all information about a person that exists in some place other than Wikipedia.
• User:Keegan makes it clear that by "private information" the functionaries mean all information not disclosed on WP "No. One never needs to nor has to post a link to someone's external information to prove a point or win a conflict."
• User:Courcelles is even clearer when they write ""Posting links to other accounts on other websites may be allowable on a case-by-case basis." sentence needs to be removed, because the number of cases this should ever happen are zero, no exceptions." It does not even appear that accounts need to be belong to a Wikipedian or any attempt to connect between a Wikipedia account and that account is required for it to be disallowed.

If these users do not mean what they have written and have a more nuanced stance of when other accounts can and cannot be linked to it would be useful to hear them. Doc James (talk · contribs · email) 12:54, 2 July 2016 (UTC)

Articles with comments at the bottom are not user accounts. This is a straw man.- MrX 13:07, 2 July 2016 (UTC)

If you look here http://www.ncbi.nlm.nih.gov/pubmed/24778001/#comments you will notice personal details including full name, city, employment, etc so yes it is a type of "other account" on another website. These accounts require a fair bit of work to set up before you can comment with them and the account owner can if they wish connect that account to a Wikipedian account in their comments. Not that different from Elance really, where people can state a Wikipedia account they claim to use. I have seen some claim to be admins here on Elance. If someone was claiming to be me on Elance I would appreciate it if someone told me. If I did not have email enabled some peoples interpretation of the WP:OUTING policy would make this very dangerous to do on Wikipedia. Doc James (talk · contribs · email) 13:23, 2 July 2016 (UTC)

If you are referring to the person whose last name is a fruit, he has already disclosed his identity on Wikipedia. An exception already exists for that case in the first sentence of of WP:OUTING. It's possible I'm missing your point though. Could give a better example demonstrating why connecting an undisclosed identity of a Wikipedia user account to a PubMed paper would be necessary.- MrX 13:39, 2 July 2016 (UTC)

You have (deliberately?) misunderstood or misrepresented my comment and your characterisation of it raises a straw man. You may use an article that has been commented on by one or more editors as a reference in an article or as part of a talk page discussion about an article, or for any other encyclopaedic purpose, as long as you do not (attempt to) identify one or more commenters there as editors here (or vice versa) unless the editor(s) in question have made that link themselves. Not that I can imagine any circumstance in which there would be any need to do so. Thryduulf (talk) 14:03, 2 July 2016 (UTC)

User:Thryduulf you would do well to clarify your comment than. The question was can any link to any account outside of Wikipedia ever be added to Wikipedia (without being in breach of outing). A bunch of functionaries including you have said no these links can never be added and there are no exceptions. If you were not answering this request for comment than I am not sure what to say. Doc James (talk · contribs · email) 14:52, 2 July 2016 (UTC)

The moment you connect an account to a real person, BLP applies and so you need a RS to back up your claim. Ideally you have a RS where the subject self identifies as the owner of an account, similar to how we handle religious claims.--TMCk (talk) 15:02, 2 July 2016 (UTC)

This RFC is not about connecting a real account to a real person per say but about connecting an account on Elance to a future Wikipedia article. One does not know if information on Elance is real or made up. The article being paid for though is typically real. Doc James (talk · contribs · email) 15:11, 2 July 2016 (UTC)

I don't believe Doc James is suggesting that "connecting an undisclosed identity of a Wikipedia user account to a PubMed paper would be necessary". The proposal under discussion is to remove the case by case exceptions, which would make it a policy violation to post a link to a website which contains identifiable information. I don't see that the proposed policy talks about a link to an outside website which is also connected to a Wikipedia user.--S Philbrick(Talk) 14:15, 2 July 2016 (UTC)

Words matter. The prohibitions on posting personal information, including links to other websites, is in the context that such links are used as a connection to an existing editor. Yet, that is not explicitly stated. I'm not sure it is even implicit. You have to know the back-story to understand why we even have such a prohibition. However, if an editor is charged with violating the policy, they will be found guilty if the violate the actual words of the policy, not an "well, we all knew what it meant, even if it didn't say it." Doc James is making the point that many, many links used as references do contain personal information. We need to work on the words to distinguish that perfectly allowable practice from that which we wish to prohibit.--S Philbrick(Talk) 14:31, 2 July 2016 (UTC)

The RfC is about removing the entire sentence "Posting links to other accounts on other websites is allowable on a case-by-case basis.", a version of which was added by Doc James over a year ago. When that sentence goes away, everything falls into place and there's no more dilemma because the policy is sufficiently clear in the first two sentences of WP:OUTING. - MrX 14:33, 2 July 2016 (UTC)

I get that you think we're handcuffing you with barbed wire, and are therefore both annoyed at the restraints and panicked about potential for pain, Doc James, but please pause for a moment and remember that we are reasonable, experienced Wikipedians who are dedicated to the good of the project and its users, and consider how likely that makes it that we're advocating the banning of external links on Wikipedia. Do you really think that's what any of us are saying? Or even that it's a reasonable extension of what we're saying? Because what we're saying is "Don't put something on Wikipedia identifying user:blah as John Smith", and what you're saying we're saying is "don't put anything on Wikipedia, ever." The operative point here is making a link between a Wikipedia user and a purported offsite identity or personal details when the user has not already made that link for themselves. Not "making a link to somewhere on the internet where a human's name is mentioned".

Look at it like a flow chart:

1. Step 1: Are you making an edit that provides or purports to bring in off-wiki information about/account(s) of a specific user or users of Wikipedia?

   If yes -> move to Step 2

   If no -> you're not outing another editor, don't worry

2. Ste 2: Does your edit link to or contain information about a specific user or users on Wikipedia that they have not voluntarily released on Wikipedia?

   If yes -> move to step 3

   If no -> you're not outing another editor, don't worry

3. Step 3: Does your edit link to or contain information that is personally identifying or private about the specific user or users (for instance: real name, location, employer)?

   If yes -> Stop, do not pass Go, do not collect COI bounty. You are outing this person.

   If no -> you're not outing someone if this is truly "no", but it's your responsibility to make sure that the answer is "no" before you save that edit

So to take your PubMed link as an example, if I were to be Dr. Sandwich, MD in real life, and I comment on PubMed with that identity and related information about myself but Wikipedia user:Fluffernutter has, nowhere in her Wikipedia history, identified herself as "Dr. Sandwich, M.D. from Johns Hopkins, currently employed at Princeton", then for you to decide whether posting that PubMed link is going to be outing me and/or get you in trouble, what you need to think about is "Am I making this external link in a context of 'Here's who user:Fluffernutter really is, Wikipedia!'?" If the answer is yes, then yes, you're about to out my secret identity as a mega-doctor. If the answer is no, you're just citing that PubMed page as a source and my comment is peripheral and your link draws no particular connection between my comment and my Wikipedia account, then you're not outing me and you have nothing to worry about, at least in the context of outing and harassment policy. A fluffernutter is a sandwich! (talk) 14:36, 2 July 2016 (UTC)

I totally agree, but your own point is that the problem occurs when making this external link in a context of 'Here's who user:Fluffernutter really is, Wikipedia!. I agree, but that's not what the poicy says. It says you cannot include that PubMed link. I think we can all agree we don't want it to prohibit that publication, but we need to use words like "posting a link to an external site and connecting it to an editor.".

I think many of our policies are loosely worded, but I don't get too excited when the stakes aren't high. In this case, indef blocks can be handed out. Let's tighten the wording.--S Philbrick(Talk) 14:46, 2 July 2016 (UTC)

(edit conflict) Yeah, many of our policies have self-contradictory parts due to years of people building them (and arguing about them) piecemeal. I personally think this one was abundantly clear before the addition of the "external accounts are sometimes ok" line, but even if we assume that even without that sentence, the status of off-wiki accounts is not made clear in "Posting another editor's personal information is harassment, unless that person had voluntarily posted his or her own information, or links to such information, on Wikipedia" (the first sentence of the WP:OUTING section), the solution would seem to be adding "[...] or links to personal information [...]" into that first sentence, not throwing out the baby with the bathwater by declaring that off-wiki accounts have to be fair game because they're not explicitly named. We could even add the word "deliberately", to differentiate the case of "I was adding this paper as a source, I didn't realize it had a comment under Fluffernutter's real name" from "Aha! This link shows Fluffernutter's real name!" There's no case where just saying "You might be able to do this...maybe...or not...you might also get indeffed, who knows" improves clarity of the policy.

As a side point, this brings up another, partially immaterial-to-this-debate, interaction point of oversight policy and outing policy: whether you knew that PubMed link contained my personal information and you intended for it to be noticed has bearing on whether you've violated outing (and resultantly, whether you're likely to get yelled at or blocked for doing it), but less bearing on whether that content may still be suppressible; if you didn't realize it was there, but I notice and am not ok with WP linking to my personal stuff even accidentally, there's a pretty decent change the oversight team would either work with me to help me figure out how to get my PubMed comment deleted, or would suppress the link and try to find a way to link to only the article with no comments instead, but the OS team wouldn't assume that it being there meant that the person who added it intended me ill. Not all suppressible edits are the result of malice, outing, or harassment, but some are suppressible nonetheless. A fluffernutter is a sandwich! (talk) 15:10, 2 July 2016 (UTC)

You gave an example explaining why reddit accounts cannot be linked. Our current wording does not distinguish between those and pubmed accounts. The wording I see proposed by a number of functionaries is no accounts ever. If that is not the intent than we need to define what is allowed. The question in this RfC is is this [3] allowed yes or no. The fact that a large number of functionaries have weighted in under no raises concerns for me. Doc James (talk · contribs · email) 15:01, 2 July 2016 (UTC)

I'm not sure how much clearer I can make it that I and the rest of the oversighters who have opposed are talking about an entirely different situation than you're trying to argue with us about. If you link to reddit user AdorableBunnies, because hey that accounts posts adorable bunny photos and you wanted to share it with me, that's cool. If you link to reddit user AdorableBunnies so that people know AdorableBunnies is me and can begin to see through my evil bunny agenda, that's not cool. Neither of those cases has anything to do with your PubMed example, which links to an article, not an account. If you genuinely think the wording of policy is unclear and we need to find a way to specify that it's drawing a connection between wiki account and off-wiki account/identity that's problematic, I would be willing to work on polishing up the wording to make it clearer, but I'm not willing to go from "it's not entirely clear" to "therefore we should just declare open season on personal information, that'll make it clearer!" A fluffernutter is a sandwich! (talk) 15:17, 2 July 2016 (UTC)

Okay so do you mean that you and other functionaries are answering a question that this RFC does not ask? The question is is this allowed [4]. And if not why not. Doc James (talk · contribs · email) 15:28, 2 July 2016 (UTC)

If that's what you were trying to ask about, I would suggest (with a wince, because I know that's a lot of work down the drain) deleting this RfC and starting a new, clearer one. Because the question I thought I was answering (and I imagine the same goes for almost everyone else who's weighed in, based on the arguments both pro and con sides are making) was the one you named the RfC with: "Can other site accounts ever be linked to" (and then backed up in the details by saying you were asking about the line in policy that said "Posting links to other accounts [...]", which is within the outing policy section about outing other users). And my answer to that is no, you may not link a Wikipedia account to an off-wiki account in pretty much any case I can think of.

If what you actually intended to ask was "Is it ok to post links to Elance jobs?" then the answer has very little, if anything, to do with outing, and I'm not sure why we're having this discussion on this page (fwiw my answer would be "I guess? Probably? Unless anyone has an explanation of why that would be bad?"). If what you intended to ask was "Is it ok to post links to lists of Elance users, while not connecting any of them to Wikipedia users?" then the issue is much more debatable, particularly as you say people then watch these Elance users with the intention of connecting them with Wikipedia users afterward. My answer to that would be "posting a list doesn't strike me as particularly damaging, but linking those Elance accounts back to eventual Wikipedia accounts, on Wikipedia, brings us back around to outing." A fluffernutter is a sandwich! (talk) 15:45, 2 July 2016 (UTC)

Yes so that was the question being asked. We link the job offer. When the job gets accepted and the article created than we have a link on COIN that now links the WPian who created the new article for pay to the data in their Elance account they used to accept the job. This is one reason why we currently have the wording that it is occasionally allowable to link to other accounts on other sites. In this example you in the beginning have a link to the buyers account. Latter you have a link to both the buyer and the creators account. I have no idea how to make this more clear than it is currently. Doc James (talk · contribs · email) 15:59, 2 July 2016 (UTC)

That is, indeed, not the question you asked in the RFC. It's also not a clear example of connecting an editor with an account on another website. After all, the article could have been created by some innocent person merely as a coincidence. WhatamIdoing (talk) 00:21, 3 July 2016 (UTC)

Sure there may be a one in 10,000 to one in a million chance that it was created by an innocent person merely as a coincidence. Doc James (talk · contribs · email) 05:45, 3 July 2016 (UTC)

@Doc James: Thanks for your reply. There are three scenarios you ought to consider:

1. Linking to an article (with a comments section) - If the link is used for a genuine encyclopedic purpose rather than for harassment, it wouldn't be a problem even if the comments section might inadvertently include personal details. (Example: nytimes.com/article.X)
2. Linking to the comments section of an article - Same as the above scenario, except I can't see any good reason to link to the comment section rather than the article itself. (Example: nytimes.com/article.X/Comments)
3. Linking to an external user account - There is no genuine encyclopedic benefit that could be derived from the posting of such links. As someone who bore the brunt of an extended harassment campaign (including fabrication of fake personal details and false COI accusations), I am well aware of the harm that such links could do: It contributes profoundly to battleground conduct and could very easily be used to harass or intimidate, and that is why a lot of people are opposed to it. (Example: nytimes.com/account/John.Doe)

These situations are all very different from each and should not be lumped together, so please clarify whether your RfC pertains to scenarios 1, 2 or 3.

RoseL2P (talk) 14:54, 2 July 2016 (UTC)

The first two contain links to the third. This RfC pertains to this type of edit [5] Doc James (talk · contribs · email) 15:04, 2 July 2016 (UTC)

At the core of all of this mess is the fact that the existing policy, with language like "case-by-case", is simply not meeting the need for a bright-line policy that all good-faith editors can understand. When a troll attempts to out someone, of course it's obvious what to do, and that, I understand, is where most outing violations take place. But we have to contend with situations where good-faith editors just don't understand the boundary between OK and not-OK in close cases. I think that Fluffernutter's flow chart is an excellent start to having something that we could put on the policy page, that would make things clearer. I can well imagine a revised and formatted version of that becoming part of the policy revisions. --Tryptofish (talk) 19:37, 2 July 2016 (UTC)

This needs to be simplified to no exceptions. "What happens on Wikipedia stays on Wikipedia"-(thank-you Las Vegas), It needs to be seen as a Chinese wallCone of Silence, or, in-universe vs out-of universe, -type thing that boils it down to a clear guideline that is not open to misinterpretation. No linking to sources, "evidence"-outside-of Wikipedia. Yes sometimes COI editors do not reply, and continue editing COI, but that is what blocks are for right? All we can do is ask an editor if they have a COI, and if they do not respond, or do not respond and continue COI editing, they can be blocked for WP:IDONTHEARYOU-right?TeeVeeed (talk) 19:42, 2 July 2016 (UTC)

a rule such as "What happens on Wikipedia stays on Wikipedia" is appropriate if this were intended as a game. WP is not: It is an encyclopedia, written about the real world, for the purpose of being of use to real people. It is not something we do to advance our own interests; it is something w do to express whatever perhaps small part of ourselves has the intention of contributing something significant, for no recompense but the satisfaction of having done it. It is not our interest that are at sake here, but the interests of the encyclopedia--and the interests of our subjects. When someone comes here to contribute, we assume they share our purposes, and we have rules to deal with those who may not share or understand them. No person who contributes to a site like this can truly expect a guarantee of anonymity, especially if they choose deliberately to act in opposition to the basic principles. Thus we have terms of use that those who are contributing with a conflict of interest must say so, even though saying so may indirectly expose their identity, and those contributing directly for money may not be anonymous, but must actually use their true name and the name of their client and any intermediaries.

the problem is enforcing it. We can only do this by blocking the contributor and removing the improper content. We can either block those who may not be following the rules on the basis of suspicion, or of knowledge. If we block by suspicion, we will inevitably hurt the innocent--there have been a considerable number of false positives in blocks for coi and socking. The alternative is to block on the basis of actual conflict of interest and demonstrable monetary influences, and to have a way of obtaining the knowledge to do it accurately and fairly. DGG ( talk ) 04:22, 3 July 2016 (UTC)

"those contributing directly for money may not be anonymous, but must actually use their true name and the name of their client and any intermediaries. "----Really? Can you please point me to that, I didn't know about that. Even-so, when we are talking about outing/harassment, it really is not the job here of editors to police other editors behavior unless we need to follow clear policy for a disruption. It is our job to collaborate and AGF. We can achieve the project's goals of having NPOV non-promotional articles by dealing with the content of articles, not the behavior of editors. Hunting for COI off-site, and outside of the project promotes vigilante-behavior, and it is not necessary to throw the full-weight of WP policies behind the cause. It should be short and sweet, yes a bright-line, unambiguous, not open for interpretation whatsoever. As productive contributing editors, we search for sources off-site, we have to make it clear that when dealing-with other editors behavior, that we don't. The fact that paid-editing, and COI editing have different rules, and that those rules are complicated, or unknown, should not affect "no outing". Don't out, don't link offsite to expose editors, keep it on Wikipedia. The "other accounts"-line in the current statement just makes it confusing and it should be a very short statement.TeeVeeed (talk) 12:24, 3 July 2016 (UTC)

edit to add, we also need to apply WP:BLP to other editors info., and that alone is guidance about outing and harassment of other editors off-Wikipedia personal information.TeeVeeed (talk) 12:38, 3 July 2016 (UTC)

I would say the strict requirement of the ToU to declare employer , client and affiliation overrides the cultural value of being able to be anonymous. It does not require a paid editor disclose their name but, as the community noted in the CorporateM situation, if following the requirements of the ToU incidentally "outs" you so be it.

As to allowing editors to use our community value to remain anonymous to violate the ToU and, since there is no legitimate reason to do so, degrade the value of the encyclopedia - we simply should not allow it.

We can come up with a simple process of sending off-site evidence to ARBCOM or the Functionaries list and posting a hash of the sent email to avoid NPA/Harassment claims on-wiki. The members of that list, of course, must evaluate that material and block the editor if they are violating the ToU. I would also, strongly, suggest that the hash be posted using a new template, say {{COI-hash link}}, so we as a community can track the editors being reported, the articles/subjects at issue, and, most importantly, whether and how the Functionaries resolve the issue. If we find that the Functionaries are not resolving/responding - over time - to the issues the data collected by {{COI-hash link}} can serve as documentation to provide to the WMF that the community can not address the problem of UPE, and an office/board action may be needed. Jbh^Talk 17:38, 3 July 2016 (UTC)

The weird thing is that a lot of COI editors DO use a "real name"-which does tend to suggest the COI, if they really are who they say they are-or not. TY for clarifying that point, unless using their real name really IS requested?TeeVeeed (talk) 02:16, 4 July 2016 (UTC)

• Also, it needs to be remembered that "other websites" doesn't include other Wikimedia projects. If you're blocked at Commons for spamming, and then you come here and spam EN Wikipedia, it's allowable to note this fact, and it would be absurd not to. "Outing" was never intended to cover this, and efforts to re-word this policy should avoid any unintentional muddling of this fact. Geogene (talk) 21:00, 5 July 2016 (UTC)

In defence of paid editing

Up until now, I have remained mostly neutral on the potential pitfalls of paid editing and have not voiced an opinion on it, yet. However, in light of the issues raised by the above RfC, I've taken some time to delve deeper into the subject and have come to a tentative conclusion that a) paid editing is not necessarily bad, and that it could b) deliver a potential benefit to our project.

These are some of my reasons:

1. Paid editors are competent and highly skilled: Not all of them, of course, but based on the profiles linked by Doc James my first impression of these people is that they are highly proficient in the English language and the majority are multilingual. Many of them claim to have years of experience in content writing and are apparently well-versed in a wide range of technical skills from general programming (C++, Python, Ruby, etc) to digital illustration (Adobe Illustrator and vector graphics), web design (PHP, HTML, etc), systems administration, and statistical or data analysis (R and MATLAB). Based on what I've seen so far, I don't think the average paid editor is any less competent than the average non-paid, volunteer editor.

2. The client has an incentive to see high quality results: If you're paying for a product out of your own pocket, you will generally want to ensure that the end product is of high quality, regardless of whether this might be an electronic device, a flight ticket or even a Wikipedia article. There seems to be a great amount of concern that paid articles could be overly promotional or even dangerous, but I've been here for almost half a decade and have never heard of someone being harmed by paid editing - which is why I think these concerns might have resulted from unintentional fear mongering. Ultimately, the reader expects an article written from a neutral point of view and if the overall text seems overly promotional, they will not trust what is being written (Remember: The average reader isn't dumb). This means the ideal paid article should be informative and also conform to an acceptable degree of NPOV such that its intended audience will trust what is being written. So, why shouldn't we allow more paid articles to be published?

3. The fight against paid editing is fostering battleground conduct: If every paid editor is viewed with suspicion and treated as if they had some ulterior motive, we'll end up driving away a lot of well-intentioned people from our project.

For these and other reasons, I am of the opinion that paid editing should not be viewed as a necessarily bad form of editing, and that people should be judged by the quality of their edits rather than their actual motives.

Finally, I want to make the following appeal to those who are supporting this RfC: Instead of harassing paid editors and taking away their privacy, why can't we welcome these individuals with open arms and shower them with much love and appreciation, just like we usually do to other editors in good standing?

RoseL2P (talk) 15:58, 3 July 2016 (UTC)

I agree with much of this. I have worked for clients on Elance for several years, but I have never worked on WP as a paid editor. However, it strikes me that on occasions, I might be tempted to. For example, there is a device for more humanely killing lobsters called the Crustastun. Animal welfare is a subject of interest for me and if I had seen an Elance advert for writing about this device, I think I might have been tempted to apply. However, with my WP experience, I would have been fully aware that neutrality is absolutely essential. It would have been in my own interests to make the article neutral so that it remained on-line and was not deleted/salted. I feel that WP should rely on its own internal PAG's rather than trying to hound (witch hunt?) editors on Elance who may be adding perfectly acceptable content. DrChrissy ^(talk) 16:24, 3 July 2016 (UTC)

The problem is that the overwhelming majority of paid editors do NOT deliver good quality of articles. An inherent bias leads to use of promotional language and a deviation from NPOV. This leads to volunteer editors spending their time trying to fix these problems, diverting the attention of editors to certain specific topics and skewing the number of editing hours in favour of paid topics. If you look at it, this is a kind of systemic bias, sometimes referred to as WP:BOGOF editing. I'm thus cautions of welcoming paid editors with "open arms". Personally, I prefer the current system of asking paid editors to propose changes on the talk page which can then be reviewed by other editors in good standing. --Lemongirl942 (talk) 16:30, 3 July 2016 (UTC)

What we end up is more problems like this [6] from undisclosed paid promotional editing. Some types of paid editing are fine such as disclosed non promotional paid editing as we see with most WiR. Doc James (talk · contribs · email) 17:15, 3 July 2016 (UTC)

So are you saying that so long as I declare that I am being paid by, for example, the manufacturers of CrstaStun on my user page, I am free to edit that article within the usual PAG's? Would I have to declare this was in response to an advert placed on Elance? (sorry - I suspect these are rather basic questions but this is all new to me). DrChrissy ^(talk) 17:31, 3 July 2016 (UTC)

As I understand the ToU and WP:PAID you would need to say you are editing on behalf of CrstaStun and there is some debate about whether you say you got the contract from Elance - definitly if you are doing business as a company or similar and best practice no matter what. Once you do that you can edit as any ther COI editor i.e. you should limit yourself to suggestions on the talk page but there is no requirement in the ToU for you to do so. WP:PAID says "Paid editing is further regulated by a community guideline, Wikipedia:Conflict of interest. This advises those with a conflict of interest, including paid editors, not to edit affected articles directly."^{(emp. mine)} It would be arguable that you could edit articles directly since WP:COI says not editing articles directly is "best practice" but not required. The last time I looked at WP:PAID it said that paid editors must follow best practices for COI editing or something similar so I would be interested in the discussions which lead to the current wording which makes something that was previously clear less so. Jbh^Talk 17:54, 3 July 2016 (UTC)

Thanks for that. On Elance I use my real name as this helps potential clients search for my previous work; I suspect this is the case for most editors on there. If I have to disclose that I have a relationship with Elance and part of this is to post a link about this, if you track down my profile, you will immediately see my real name, a photo of me and the area where I live. I have been forced into self-outing. DrChrissy ^(talk) 18:07, 3 July 2016 (UTC)

Just to jump in here, no, you do not need to post a link to your profile. You only need to acknowledge the client and the relationship. Most clients employing Wikipedia editors don't list the name of their business, so it would be very difficult for someone to work backwards from an Elance/Odesk client to your profile, given that they would have trouble identifying the client to do this with. - Bilby (talk) 02:38, 4 July 2016 (UTC)

There was a rather involved discussion re CorporateM, which I mentioned above, when the new ToU came out where, since he was self-employed/worked for a very small firm, by naming his "employer" he would be outing himself. I am not sure where the discussion took place but you may try checking his talk page and COIN. Anyway the consensus was that the ToU/PAID requirement took precedence over no toting himself.

In the case of Elance and similar sites, as I understand them, they provide services like escrow (do they take a cut?) that make them more than a simple pass through like answering a want-ad would be. Also, there is the issue of paid editors from Elance/etc using multiple accounts/SOCKing which would be prevented by requiring a link back to their Elance account. The point of disclosing client, employer and affiliation is to give some form of accountability. If lots of editors from PRfirmA are violating ToU/PAID then PRfirmA can be contacted, sent a C&D letter or whatever. If an Elance account is doing the same then Elance could be notified of the ToU violations because, as I understand it, their ToU requires that people who work through ther service follow the ToU of sites like Wikipedia. (Note when I am taking about firms being contacted I mean by WMF or their designated representatives not by Joe-random-editor.) Jbh^Talk 19:00, 3 July 2016 (UTC)

Eh, if WP is going to embrace paid edits, I think we should cut-out the middle-people like Elance. Drive the paid-editors and pr firms, and COI-edits to Wikipedia, make paid-editors pick-up the requested edits on Wikipedia, then allow them to either take a fee, donate the fee to Wikipedia, or donate the fee somewhere else. Also make them go through a proficiency test and/or acknowledgement of key policies before paid-editing. And don't allow scraping of all available tasks. We know this is being done, why not make it easier, more controllable, and more transparent, AND revenue-generating for the project?TeeVeeed (talk) 20:33, 3 July 2016 (UTC)

I disagree very strongly with the premise of this section, as it applies to undisclosed paid editing. It's true that good-faith paid editors who comply with the Terms of Use and with our policies can be valuable editors, and they are already allowed under the existing system. But the problem is that "The client has an incentive to see high-quality results" is untrue, as we define high-quality here at Wikipedia. The client has an incentive to see whatever they support presented in a favorable POV and whatever they oppose or compete with presented in an unfavorable POV, particularly when the editor has been hired to edit without disclosure. But I would like to think that Wikipedia can walk and chew gum at the same time. We can take effective measures against the disruptive forms of paid editing, while simultaneously protecting the private information of editors. It's not either-or. --Tryptofish (talk) 22:05, 3 July 2016 (UTC)

A neutral point of view is one of the pillars of the project. Paid editors may be highly skilled, but they are highly skilled at making their client look good.

I agree that the average reader is not dumb. That being said I don't think that those who engage in paid editing subscribe to that philosophy. They make an effort to seem neutral but they do not make neutrality their goal, because their goal is to make their client look good. The goal of any advertiser(and that is what paid editors are) is to make people think your client is wonderful, and that is not always in alignment with reality or our goal of neutrality.

The suggestion that a person whose job it is to promote a company would have an incentive to produce high quality work seems to miss one crucial fact, that they have different goals than us. For them a high quality article is one that presents their client in the best possible light, for us high quality means a neutral presentation. I invite you to watch 5 minutes of commercials on TV to test your theory that advertisers think neutrality is the best way to sway potential customers.

It is not a question of suspicion because there is no doubt. Their goal of making their client look good is in direct conflict with our goal of representing a neutral point of view. Unless we give up being neutral, or they give up trying to make their client look good we will always be at odds. HighInBC ^{Need help? {{ping|HighInBC}}} 22:51, 3 July 2016 (UTC)

Well, they are here anyhow. I also think that they should use a different font when allowed to directly edit articles.TeeVeeed (talk) 02:06, 4 July 2016 (UTC)

RoseL2P, just commenting from my pespective:

• Paid editors are competent and highly skilled: Generally no, the paid editors are not competent editors. Most of the paid editors I see working through Upwork are not very good editors. The best of them create questionably sourced articles which tend not to survive at AfD. I am sure that there are competent editors who are paid, but the bulk of the freelancers don't fall under that category.
• The client has an incentive to see high quality results: no, mostly wnat they want to see is a link to their business from Wikipedia. They are not after a high quality, NPOV article, but a short piece advertising their services or products that gives them a link and higher visibility in Google ranks.
• The fight against paid editing is fostering battleground conduct: that I agree with. The problem is that we have very, very few paid editors who follow policy, and large numbers who don't. I would rather us treat those who do follow the policy better, to make it more of an advantage for them to work that way. The problem is that whether they follow policy or not, most of their work is non-notable, and will ultimately be deleted. By being open they make it easier to spot the non-notable stuff. By breaking policies, it is easier to sneak it under the radar. - Bilby (talk) 02:38, 4 July 2016 (UTC)

All of those are good, but I want to reinforce the first. Most of the paid editors that get caught are caught because they're writing (or in many cases, copy-pasting) God-awful ad copy in Wikipedia's voice. Competency is required, and Wikipedia, as an organization, has to deal with this spam somehow, and the COI hammer is the most efficient way. Geogene (talk) 21:34, 5 July 2016 (UTC)

Just to point out a couple things:

• If you are an officer of a company and you instruct your PR reps (whether employees or contractors) "I want our Wikipedia article to be as accurate and balanced as possible, and please disregard the interests of the company" you're committing malfeasance of your fiduciary duties to the stockholders, and not only will you be fired (of course) but you have committed a clear tort, and arguably an actual crime.
• If you are a PR representative and you put the accuracy and balance of a Wikipedia article about the needs of your client, you are in violation of the ethical canons of the Public Relations Society of American are subject to expulsion (if you are a member) and are probably guilty of fraud, for which people go to jail.

Look, if I ever get famous, I'll want a good PR guy. The role of a PR guy is to try to spin things in my favor. His job is to think "how can I get the story about my client into to the Times, and to present him most favorably?" not "What is best for the Times and its readers?" See the difference? Herostratus (talk) 19:37, 4 July 2016 (UTC)

Also interesting, particularly point 2. And that doesn't even get into SEC issues. Geogene (talk) 21:34, 5 July 2016 (UTC)

Thanks for your comments. I'm just sorry to say that, from my NPP experience, you're in a very small minority. I have occasionally run into good COI editors who are willing to respond and accept changes and are open about who they are, but it's rare. Frankly, most small businesses don't hire an expert: they do it themselves and just copy in their "About Us" text.

• I do think we need to admit that often this isn't intentional. Editors just sign up, copy in a generic "about us" promotional text (surely they must know this is not a normal kind of Wikipedia article text, surely?? Apparently not.), they don't stick around to see if they get any messages requesting changes, they don't grasp that they're not notable yet and creating a page on a non-notable business will give the page a scarlet letter in the deletion log. Often there seems to be a desire to not admit to anything - immediately switching to IP editing when asked if they need to declare a COI. (Frankly, reading a lot of these articles, I tend to get a sense that if they approach the other sides of their business this way, maybe there's a reason why their companies aren't so successful.) Or simply someone's been told to delete the "Criticisms" section off their Wikipedia page.

• I think there are constructive ways forward with this. What I would love to see is that COI editing is dealt with at the signup screen stage - when creating an account you're asked, politely, if you are editing on behalf of anyone and invited to fill out a form disclosing this. (Like when you land at an airport you're asked if you're here for business or pleasure.) You would then be given a link to the COI policy and invited to read a basic guide to business notability before writing an article, and the COI details would be automatically placed on your talk page. They would also be invited to connect an email account to receive messages - important because often COI editors, being non-obsessives, only check their Wikipedia account when they have to.

• The reason why this would be good is that I think paid editors often see admitting a COI as humiliating - like being forced to admit when you're a kid that you took five cookies from the jar when you only were meant to take one. So when they're given a warning they don't stop editing, just create a new account or use an IP, forcing people to open SPIs to deal with them. It would be much easier if people saw this as a formality that you do as part of registering an account.

• It goes both ways and courtesy is always needed on both sides - I try when dealing with COI editors always to imagine some harassed junior employee given instructions from some overbearing boss. Just because you may need to revert promotional edits doesn't mean you have to sound mean when talking to such editors. (For that reason, I've written some of my own messages to contact people with rather than use the default Twinkie warnings, which are the furthest thing from friendly.)

Let me know if you have any thoughts about this - keen to hear what people think. Blythwood (talk) 22:45, 4 July 2016 (UTC)

I think that your idea of tying it into account creation is an excellent idea! That should go on the list of software development requests to the WMF. --Tryptofish (talk) 23:10, 4 July 2016 (UTC)

I ♥ that idea too. A somewhat legit/registered-as group of disclosed paid editors could help them and us and maybe their clients?TeeVeeed (talk) 03:30, 5 July 2016 (UTC)

I knew someone could come up with something that was better than, corporate editor ten, unimportant volunteer editor zero. Of course this is a slippery slope one encounters when dancing with the devil.

Here is a view not considered. Corporate= money, and they do not consider it greed, or improper, or wrong in any way to advance making more money. That is usually the only goal. If I can pay an editor to get an article up, for even a short amount of time, I will pay another company to advance it in the search engines, make some presentations, put it up on a screen at a meeting, and show "we" are doing great. By the time the article has gone through a deletion process I have accomplished what I wanted to do. That is why it does not matter if an article "makes it" or not, and why the creating editor does not have to stick around to "answer questions. Not that it will matter because "if" I was a company making money by getting articles created that are inevitable advertisements I would have editors in at least 20 of my covered cities watching this page. Some would be IP's and some would be logged in editors. Pricewaterhousecoopers reported online advertising for the first quarter of 2016 was nearly 16 BILLION dollars. Written out I believe that would be $16,000,000,000.00, and this was a 21% increase over the same time frame last year. In case anyone missed it; Corporate= money. If Wikipedia is not ever going to get any of that money why provide a free cash cow? If I can get someone to give me free advertising space for a certain amount of time for free, then I can make a lot of money saving money, paying small creative editor fees, and by the time my advertising time has expired by deletion, I will have pocketed the cash and moved on, NEXT. Otr500 (talk) 04:35, 5 July 2016 (UTC)

Section for discussion of a specific case

Wondering peoples thoughts on this specific case and whether or not they see it as acceptable:

Here we have someone who is buying an article on Anthony LaPine. They have already bought an article on HipLink and this sock created it UserJuliecameo3 who is already blocked. Doc James (talk · contribs · email)

Doc James (talk · contribs · email) 06:20, 5 July 2016 (UTC)

Is that the right link? I don't see "LaPine" mentioned.--S Philbrick(Talk) 12:42, 5 July 2016 (UTC)

This WP link is correct [7]. The Elance link is from a year ago and is dead but basically it mentioned someone was offering money for the article on Anthony LaPine and it showed a number of bids from different accounts to write that article. I could show a functional example but it is unclear if that would result in an indefinate block Doc James (talk · contribs · email) 13:29, 5 July 2016 (UTC)

I added "wikipedia" as a search term, https://www.upwork.com/o/jobs/browse/?q=wikipedia . The results are grotesque. Comments like, "the client's Wikipedia page"--and, "The client does have a recently established Wikipedia page"....regardless, I contend that we not out editors, or expose individual editors off-site activities specifically on Wikipedia and it is a separate issue with personal identity policy left intact. This search just gives more reason for Wikipedia to take-on the paid editors and buyers here, and eliminate the middle-persons. If something like this were presented to Wikipedia the buyer would be set straight immediately about who "owns" articles.TeeVeeed (talk) 13:48, 5 July 2016 (UTC)

Not sure what you mean by "This search just gives more reason for Wikipedia to take-on the paid editors and buyers here, and eliminate the middle-persons." User:TeeVeeed Doc James (talk · contribs · email) 15:50, 5 July 2016 (UTC)

I just think that Wikipedia or WMF should take-on paid-editing on-site, and have a marketplace of sorts for requests/buyers to post jobs where paid-editors could accept the jobs transparently, and any fee could be taken by the editor or donated to WMF. Also I like the idea mentioned above where SPA of paid editors are registered at account creation as paid editing accounts. In any event, although I 100% support that undisclosed paid editing is disruptive and destructive, I still think that we need to keep that a completely separate matter from no outing, no posting-of other editor's personal information or links offsite to editor's personal identifying info. that has not been self-disclosed on wikipedia by the editor. (no "other accounts", no "certain cases" mentioned in policy)----to eliminate confusion, not to protect undisclosed paid editors, because they can be handled in other ways and not publicly out anyhow.TeeVeeed (talk) 16:12, 5 July 2016 (UTC)

I looked at the case, and I clicked through to the two links that are at the top of that discussion. One is for Upwork ads, including ones for editing here, and the other is the profiles of people who are often hired to edit Wikipedia. I'm getting pretty close to rethinking my earlier RfC reply after actually seeing this. The ads (someone wanting to create a BLP about a supposed "celebrity" etc.) are quite horrifying. And the personal profiles really do not seem to me to be the kind of private information that is supposed to be protected by this policy. The people there are voluntarily making it public information that they are the corresponding editors here. They are clearly indicating that they intend to violate policy and terms of use, and they are presenting it as public information in such a way that it is extremely unlikely that we would misidentify an innocent editor with a similar user name. --Tryptofish (talk) 22:02, 5 July 2016 (UTC)

Yes that is basically the case. Doc James (talk · contribs · email) 06:09, 7 July 2016 (UTC)

Flowcharts

In #Posting of personal information, above, Fluffernutter used the idea of a flowchart to clarify the issues that editors are disagreeing about in the RfC. I think it's a very good way to focus discussion, so I have taken that a step further here.

For editors Step 1: Will the edit that you are about to make contain information from off-wiki about another editor or editors?  If no → You are not outing another editor, don't worry.  If yes → Go to Step 2. Step 2: Does your edit link to or contain information that is personally identifying or private about the specific user or users (for instance: real name, location, employer), or does it make it easier for someone to find such information?  If no → You are not outing another editor, don't worry.  But it is your responsibility to be certain that the  answer really is no.  If yes → Go to Step 3. Step 3: Has all the information in your edit been provided voluntarily on Wikipedia by the user, and never redacted?  If yes → You are not outing another editor, don't  worry. But it is your responsibility to be certain that  the answer really is yes.  If no → The edit is outing. Do not post it!  Go to Step 4. Step 4: Is the information genuinely useful for investigation of conflict of interest or undisclosed paid editing?  If no → Drop the matter entirely.  If yes → Consider submitting the information  privately, but do not post it.

For administrators Step 1: Does the edit violate the outing policy according to the flowchart at the left?  If no → No action.  If yes → Immediately rev-del the edit and have it  oversighted as soon as possible. Go to Step 2. Step 2: Do you believe at this time that the user who made the edit is likely to continue violating the policy?  If yes → Block and revoke talk page access.  Go to Step 4.  If no → Go to Step 3. Step 3: Do you believe that the editor who made the edit will stop immediately after receiving a warning?  If yes → Issue a final warning, and block if the  warning is ignored.  If no → Block and revoke talk page access.  Go to Step 4. Step 4: After the block, do other editors discuss the incident or the block on-Wiki?  If no → No further action.  If yes → Go to Step 5. Step 5: Monitor each comment in those discussions. Do any comments reiterate information that was oversighted?  If no → No further action.  If yes → Return to Step 1.

As you can see, I added a flowchart for administrators as well. In constructing it, I included two aspects that might not always be current practice: (1) treating blocks as preventative, not punitive, and (2) taking care not to allow Streisand effects to undermine the privacy of the outed user. --Tryptofish (talk) 20:40, 10 July 2016 (UTC)

I still think this flowchart is missing a fundamental concept, something along the lines of the the purpose of the edit is to make a hitherto unknown connection between a Wikipedia editor and a real life person.

Let me illustrate with an extreme example. The biography of Jimmy Wales includes a mention of his mother's name. This clearly qualifies as personal information. Now, perhaps he mentioned that somewhere in Wikipedia, but for the sake of this example, let's assume he did not. I can already hear some naysayers—after all, no one is unaware that Jimmy Wales the real person and Jimbo Wales, the editor are the same person. But I see nothing in the flowchart that excludes this edit from consideration.

• ...contain information from off-wiki about another editor or editors?
• ... information that is personally identifying or private about the specific user or users...
• Has all the information in your edit been provided voluntarily on Wikipedia by the user... NO therefore OUTING!

Obviously, no one is going to impose an indef block on whomever added that information, but in what way is it not a violation of policy, if the flowchart reflects policy?

Lest anyone miss the point, I'm not arguing it is outing, I'm arguing our policy is deficient.

If anyone thinks this simply requires a tweak excluding users who use their real name, note there is nothing in the example that depends on that fact.

We know many subjects of biographies have created an account, if only to make corrections to the article about them. In many cases, the user name they use is unconnected to their real name. Yet if the article about them contains any personally identifying information, as all do, then that edit qualifies as outing per this flowchart. It needs work.--S Philbrick(Talk) 21:04, 10 July 2016 (UTC)

That's a very thoughtful response, thanks. Perhaps I just misunderstand, but I would think that the reference to "real name" in Step 2 of the Editors chart addresses that issue: connecting a Wikipedia user to their real name is equivalent to connecting to their real identity. --Tryptofish (talk) 21:19, 10 July 2016 (UTC)

In the example I gave, the offsite page contains Jimmy's real name. One of us is missing something - because I don't want that edit to count as outing, yet it seems to exactly follow your flowchart and qualify as outing. I am fairly certain you don't intend that edit to be outing, so tell me which of steps 1,2, or 3 that it fails to satisfy. — Preceding unsigned comment added by Sphilbrick (talk • contribs) 22:15, 10 July 2016 (UTC)

Oh, bleep, I misunderstood. Thanks for correcting me. You are correct: I wrote that chart based on the assumption that it would be outing. So it comes down to the extent to which information voluntarily posted would cover a situation where the Wikipedia information is "Jimbo Wales" and the offsite information is "Jimmy Wales" or "James Wales". Well, right here we have demonstrated how untrue it is that the community has a consensus about what is, and what is not, outing. For myself, I tend to take a more flexible attitude about it than the chart would indicate, and here I was trying to reflect what I've been hearing from a lot of Oversighters and Arbs, but it's a tough decision. What happens when a username is JJones, and someone links to a blog by John Jones, to indicate something very improper by the JJones account, but in fact JJones is really Jane Jones, someone completely different? There can certainly be situations where inferring a real name from a username can end up being harmful and disruptive, but I agree that there are also situations where it is just plain common sense in a COI investigation. And I don't think the community has consensus about this. When I try to raise these kinds of points, what usually happens is that a functionary tells me I'm asking an unnecessary question, because it's all obvious and simple. See above in #Question about another scenario. So, @Drmies: are you still interested in answering the questions I asked up there? --Tryptofish (talk) 00:03, 11 July 2016 (UTC)

I was working on a response, but I think the light went on. Yes, that's the point, it may not be outing. I think it's salvageable, I like the concept of the flowchart, but I think some things are missing.--S Philbrick(Talk) 00:30, 11 July 2016 (UTC)

I really like the flow chart and was glad to see this structuring came up. I think that the aspect of when an editor self-identifies on wiki needs some additional clarification in this too, but I'm running short of ideas right now tonight. Overall with respect to COI, what do we considering outing or harassment for off-wiki material when an editor has linked themselves to their real life identity by user name or by directly stating their identity in some fashion?

1. John X is editing an article on a new stock market theory. An Amazon search shows this person has a large number of "how to beat Wall Street" guidebooks they sell on this subject (non-academic).
2. John Y is editing the page of a company. A search of the company's front page, main staff page, etc. show a John Y as a Vice-president without any other private information like home address, etc., but their name isn't on the Wiki page.
3. John Z is editing on specific agriculture chemical company pages. A search of his name reveals a John Z who has a webpage stating he does private PR consulting for these companies out of his own home. The website includes his personal phone number, address, etc.

Basically, I don't consider 1 outing because that is a reasonable case-by-case usage of a very basic search in the relevant topic (i.e., not fishing for irrelevant or very weak evidence more suited for a smear campaign). 2 is meant to begin approaching a gray zone. A company directory is available to the public and would be considered another reasonable relevant case. That much seems fine for taking care of entirely on Wiki (remember the editor has identified their name). The murky water starts with 2 if more personal information in included like phone numbers where I'd be more inclined to request an admin to look it over privately. 3 should be a clear cut case as when you're starting to involve someone's home and more private information, you really need to rely on an admin instead. Regardless of how people classify the actual scenarios, how would we address the different levels of information in the examples in this chart? Where should we make a rough cutoff saying caution is needed?

We maybe could also include a dummy example where finding a blog by John Z saying he is a pancake-hater and calling him a pancake hater on pages having nothing to do with pancakes would fall into harassment related to personal information, but that could also be WP:BEANS Kingofaces43 (talk) 05:28, 12 July 2016 (UTC)