Wikipedia talk:Harassment/Archive 7

This is an archive of past discussions about Wikipedia:Harassment. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Wikipedia talk:Harassment/Archive 6#Proposed clarification of Outing

Obviously, my suggestion in the multi-break section above does not have support, and that's fine. But I thought of another idea, that might perhaps be better, and I'd like to suggest it here. Following the earlier discussion, the last paragraph of the Outing section, the paragraph beginning "Nothing in this policy prohibits the emailing...", was added as a clarification of the ways in which investigation is acceptable. It occurs to me that the last sentence of that paragraph could be changed, from:

Posting such information on Wikipedia violates this policy.

to:

Posting such information, or indicating how to find such information, on Wikipedia violates this policy.

Is there support for adding this? --Tryptofish (talk) 23:52, 29 December 2015 (UTC)

I'd be worried about adding anything that makes things harder for good-faith editors. I support this policy, but I've seen editors threatened with blocks for nothing because of it, and equally I've watched it not be enforced after blatant violations. One of the reasons for the haphazard enforcement is that it's inconsistent with efforts to reduce paid advocacy. Because of the inconsistency it's hard to see how to fix the writing. Without a careful overhaul, every addition risks making things worse.

This would be a violation, and lots of admins would block for it: "If you want to know who X is, look here." We don't need new words for that.

But so might this, according to the proposed wording, and in my view admins should not block for it: "If you look in X's edit history, he used to edit as Y and is active with that name on Twitter," where something on Twitter leads to X's real name, even if he has added it himself – including if he added it himself after the fact to claim he had been outed on WP.

So the words about "indicating how to find such information," would have to say something about intention. That's tricky, but even if we could fix it, the problem of making life harder for editors opposing paid advocacy would remain. SarahSV ^(talk) 00:13, 30 December 2015 (UTC)

I'll take your example of "If you want to know who X is, look here", and make a small change to it: "If you want to know why I am saying that X is violating the paid editing policy, look online." It's not talking about "who X is", it's not stating which website to look for, and it's doing it for the noble purpose of reducing paid editing. But it is absolutely a violation of this policy, and editors should have no doubt that they must not do it, not if they are posting it on-wiki. It does not matter how well-intentioned the editor saying it may be. It should always be a violation.

It may be helpful to consider the context in which I propose this revision to go, so I'll reproduce it within the entire paragraph:

"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, or indicating how to find such information, on Wikipedia violates this policy."

To me, saying it in this context does not in any way make legitimate policy enforcement more difficult, but it does make clearer that something prohibited should not be done out of misunderstanding. --Tryptofish (talk) 18:40, 30 December 2015 (UTC)

At this point, are there objections to my making this change? --Tryptofish (talk) 18:15, 8 January 2016 (UTC)

I have no objections to that addition. It seems reasonable. If you say "X's name is on Wikipediocracy" that's as much outing as saying the name. My only suggestion would be to specify that pointing to info off Wiki is a problem. Some people out themselves on wiki. EvergreenFir (talk) Please {{re}} 18:25, 8 January 2016 (UTC)

Thanks very much, and that revision is a good idea. I'm thinking: add the word "offsite" after "such information" (as that wording might be a bit more accessible for inexperienced users). --Tryptofish (talk) 19:18, 8 January 2016 (UTC)

As I said above, I object to it, and others objected to a similar proposal at Wikipedia talk:Harassment/Archive 6#Proposed clarification of Outing. As worded it could cover many scenarios, some of which admins already block for, and some of which may need to happen to deal with paid advocacy. The spirit of the policy matters as much as the letter. SarahSV ^(talk) 19:39, 8 January 2016 (UTC)

And that is why I have asked in talk instead of actually making the change. But this is not the same proposal as the previous one, although it is motivated by the same concern of mine. If it covers a situation where we agree that it results in a block under current practice, then it is a good thing to make it clear, and there is no harm in doing so. As for your second scenario, that would be where an editor "indicat[es] how to find such information offsite" by posting that on-Wiki, and you seem to be of the opinion that doing so, as opposed to following what it says in the rest of the paragraph (communicating privately) is both necessary and not a violation, just so long as the stated intent is to report paid editing. Well, I'm sorry, but that is just plain untrue. Saying something like the example that I gave above – "If you want to know why I am saying that X is violating the paid editing policy, look online" - is always an unambiguous violation of this policy, and editors need to understand that. It is never OK simply because the intent is to control paid editing. And it is never impossible instead simply to email it privately to a trusted administrator or arbitrator. There is never a situation where posting the information onsite is the only option. Never. And the very fact that you would consider that it could be a permissible option actually demonstrates the need for this policy to make it clearer. --Tryptofish (talk) 20:05, 8 January 2016 (UTC)

I often see situations where people are discussing who someone has said they are elsewhere; or a website they've linked to that seems to be their own, and is being handled as COI spam, etc. This happens on an almost daily basis. We can't insist that an editor email an admin every time someone is spamming a personal website, in case pointing to it leads to someone else finding the editor's name.

Editors adding words to this policy have to be aware of all these situations to make sure we don't inadvertently make them blockable. The best thing (or the best thing we have given the inconsistencies between policies) is sometimes to rely on the spirit of the policy for difficult cases. SarahSV ^(talk) 20:38, 8 January 2016 (UTC)

It's useful to discuss these kinds of examples, in case the examples point us to a better way of writing what I have been trying to say. If someone is spamming a website, that means that they are posting it on Wikipedia. As such, it is perfectly permissible to discuss that under this policy. Self-identifying information voluntarily posted on-Wiki is not protected by this policy. But if an editor has never stated on-Wiki that they are the same person who posts as so-and-so at another website, and someone else says onsite that this editor is that person, that's outing, and if they say, instead, that if you search online you will find another website where the editor also posts, that's outing too, and we need to make that latter situation clear. I still haven't seen anything that contradicts what I said just before: that it is never acceptable to point other editors, publicly, to where they will find outing information. The fact that the editor being outed is violating a policy and that it would supposedly be an inconvenience to submit the evidence by email does not make it OK. In fact, the talk section directly above this one reaffirmed that the fact that an editor violates a policy does not make it OK to out or harass them. --Tryptofish (talk) 20:49, 8 January 2016 (UTC)

Three scenarios:

1. An editor says of X, someone he's in dispute with: "Hey, if you want to know who X is, just google him." That's blockable, unless X has made clear that he doesn't mind.
2. An editor says of X, someone he suspects is a paid advocate: "I'm struggling to find a way to word this, because I don't want to violate OUTING, but I find myself in a situation where I can't even refer to a simple Google search." It is functionally equivalent to (1), but no one should be blocked for it.
3. An editor says of X, someone who is spamming: "X is posting links to My Lovely Website. Please warn him." This also amounts to (1) if the website leads to a name, even indirectly. But it happens all the time, and no one is blocked for it.

Therefore, the policy can't simply say "or indicating how to find such information," unless we find a way to safeguard examples (2) and (3) and similar. SarahSV ^(talk) 20:59, 8 January 2016 (UTC)

Those are good examples, so thanks for offering them.

We agree about (1), and (1) is most directly related to what I have proposed.

I find (2) quite interesting, because of the way the editor saying it appears to be struggling. It sounds like that editor would have been happy to have had some guidance. Telling them to say, instead, "I believe that X is a paid advocate, and I have emailed my reasons for this to (trusted person with advanced permission)" and then having that trusted user confirm the accusation without providing identifying details, is what this proposed revision would do, and I would think it would be very helpful. But there is a real problem with that editor posting "I can't even refer to a simple Google search". That is referring to a Google search! No one should be misled into thinking that's OK, simply because the editor saying it couches it in terms of feeling badly about it. It's no better than saying, "I wish that I could tell you that X's real name is John Doe, but I can't because of the outing policy."

On the other hand, what I propose here has nothing to do with (3), because in (3), the user is posting those links on-Wiki. Once they post it here, there is nothing objectionable to referring to what they posted, even if what they posted contains personal information. And, alternatively, if that user is spamming inappropriate links, one can quite reasonably just revert all those links and criticize the user for spamming them, without going into what the user's motivations were.

I'd welcome more scenarios, but I'm still not seeing any problems. --Tryptofish (talk) 21:48, 8 January 2016 (UTC)

• I thought this had been discussed and not agreed to several weeks ago. The change, as it stands, can be gamed as was pointed out earlier. Jbh^Talk 00:11, 11 January 2016 (UTC)

I have to agree that this has been discussed and rejected. The real problem is that it is so difficult to describe real problem editors that we need to deal with. Anything that makes this even marginally more difficult is IMHO a problem. Smallbones_(smalltalk) 02:01, 11 January 2016 (UTC)

Well, there was a reason that I proposed this in talk before making the edit. If editors have concerns, I am quite willing to discuss it, but please do not expect me to read minds. There were three editors who participated in this discussion, up to this point: EvergreenFir, SlimVirgin, and me. Two of us supported the proposal, and the last comment in talk was my response to SV's most recent concerns, which went unanswered for two days. There was a discussion about a different proposal previously, but I could reasonably interpret the absence of criticisms of this proposal by editors who objected to the previous one as an absence of objections from them. How did you expect me to know that you also objected to this proposal? Why did you think it was appropriate to respond with silence and then immediately revert once I made the edit?

So, I am now hearing concerns on three points, but stated only in very general terms:

1. The first is that the present proposal, in its current form, "can be gamed". I'm not seeing how one can game the existing, and already agreed to, requirement that private information be handled privately by emailing it to a user who can deal with it, and not posted onsite. I remember from the past discussion a claim that a paid editor could deliberately post personal information so as to be able to engineer a "gotcha" when investigated. That's kind of contrived and improbable, but let's consider it. Given that the policy already says that you cannot out the person, please describe a situation in which it is OK to describe on-Wiki how to find material outing that person.
2. The second is that this proposal leaves some real problems undescribed. The real problem that I am trying to address is that editors might think, incorrectly, that it's OK to point to outing material (without posting the actual outing information, just pointing to it) if the person being outed has been doing something wrong, such as paid editing. Can anyone describe a situation where it actually is OK to point to such information, as opposed to handling it privately? As a rhetorical point, does anyone really want to add to this policy page this sentence: "If however you are investigating a paid editor, it is OK to post information about links containing personal information about that editor on-Wiki."?
3. And the third is that it makes something (presumably dealing with paid editors) "marginally more difficult". This appears to reflect a mistaken belief that enforcement of policies like those prohibiting paid editing trumps policies requiring editor privacy. That is false. The already agreed-to policy says that one cannot post personal information about such users onsite, but should instead email it privately, and I'm not proposing anything that would change that. So I guess that if you cannot post something that does not actually contain the personal information but instead would lead people reading your post to find that information, but you can still email it and have it dealt with privately, there is some marginal inconvenience in emailing it instead. Really? Is that such an inconvenience? Is it better to ignore the privacy concerns about the user?

Please explain these three concerns, preferably with hypothetical examples. (I do not believe that you can, but please prove me wrong!) Please do not just hand-wave, by saying that there must be some sort of situation that we have not considered, because that is nonsense. If there is a problematic scenario that my proposal fails to account for, it should be possible to describe it. And if there is, I am happy to treat that concern seriously. But I do not accept silence as being a rebuttal to what I have said. --Tryptofish (talk) 18:08, 11 January 2016 (UTC)

I've let a couple of days pass, waiting for replies. It would make me happy if their absence means that editors now agree with me, after reading my comments above, but for obvious reasons I do not think that I should assume that. I would be saddened if this means, instead, that some editors think that they can just revert edits to the policy page without responding to good faith discussion. I'm going to allow a bit more time, but then I will make the edit and consider it a matter for dispute resolution if anyone reverts again without discussion.

And I am very sincerely interested in addressing the concerns of editors who have raised objections. I'm happy to discuss any concerns before making any page changes. I've been reading various user essays about combating COI editing to see if I can gain insight into what the concerns with this particular proposed revision really are, but I have not found anything. One possibility that I can think of to offer is to revise the proposed change to: or knowingly indicating how to find such information offsite, by adding the word "knowingly". Would that help? --Tryptofish (talk) 00:51, 14 January 2016 (UTC)

For clarity, here it is in the context of the entire paragraph where it would appear:

"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, or knowingly indicating how to find such information offsite, on Wikipedia violates this policy."

--Tryptofish (talk) 00:54, 14 January 2016 (UTC)

Many concerns have been brought up in your prior attempt to make this change and remain the same - brushing aside concerns of others "...but I'm still not seeing any problems." You have not obtained consensus for this change and I see many editors who are still concerned. Do not take exhaustion as acceptance. I will say again your changes are over broad and are subject to gaming. The rules as they stand provide the same level of protection and still allow enough flexibility to prevent gaming. The onus is on you, who wants to make the change, to support the change and show what problem it is it will address that the current wording does not.

"Knowingly" is a drama magnet - what kind of due diligence must an editor go through and how do they document it if some personal information might be in a secondary link or in a place they did not look? That can be gamed from both sides of the table. Without the term you have a strict liability situation which is also unacceptable. What is to prevent someone from adding information that could be considered "outing" after a site is pointed to - it is not like web pages in general have a "history" tag. Another example is wiki-Bob admits to being web-Bob (both pseudonyms) and, for whatever reason, an editor brings up www.web-bob.com in a dispute. Wiki-Bob then cries OUTING! because whois will return their real name. There are many edge cases like this - more than I can think of - but I am sure someone will and drama will ensue. So first off - what is your evidence that the policy as it stands is a problem requiring the solution you propose? Where has it failed? Is it failing frequently? What drama would this change prevent? Jbh^Talk 01:52, 14 January 2016 (UTC)

Thank you for replying now, and I am happy to discuss those points with you. But first, you need to understand that I am not brushing anybody's concerns aside, and I resent your saying that I do. Just look at how I tried to respond to what I could make of what you and Smallbones had said. And this is not the same proposal as the one before. It's different words, in a different place in the policy page section, and thus in a different context. Exhaustion? Oh, come on! I'm not a troll. So please don't brush aside my concerns. It's not good enough to say that you opposed a previous idea of mine, and then, when I say in good faith that I have a different approach, to take the attitude that any new suggestion can be brushed away because you opposed an earlier one.

OK, that said, let's leave out "knowingly". It was only a stab in the dark on my part, hoping it would help, and I agree that it raises problems.

So, you ask what problem I am attempting to solve, and that's a valid question. Editors who read the policy as now written might think that it prohibits making an edit that contains "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", but if the edit does not contain any such information, it cannot be outing. In other words, if someone posts something roughly like "wiki-Bob is a paid editor, and you can find that out by a simple Google search", they could reasonably think that it isn't outing, because none of those things like "legal name..." etc. were in the comment. But if a Google search would lead to outing information about wiki-Bob, then it is outing. And yes, I have seen, first-hand, blocks being given to editors who posted things quite similar to that, in good faith. It's happening. (One can also see how ArbCom currently places a very high value on preventing harassment at Wikipedia talk:Arbitration Committee/Noticeboard#Break due to lack of recent discussion.) And I have seen editors in these discussions here on this talk page indicate that they think it might be OK to post something like that, if it felt inconvenient to email it privately, so there is some confusion about it. It's a real problem.

Now let's consider the scenarios that you have described. The first is that an editor points to a site that does not contain any personal information, so it's an entirely reasonable thing to do, but then the bad player quickly puts up personal information on that site and cries "outing!". I would think that the well-meaning editor would be able to say that the personal information wasn't there at the time of the edit, and that it was added after the fact. I would hope that would be an acceptable defense. Has there been a problem with this happening?

Your second scenario is where wiki-Bob says, onsite, that he is also web-Bob. Once he has disclosed that, his claim of outing is without merit. It does not matter about whois. Once someone voluntarily posts information here, it's no longer outing to disclose it. So pointing to the website is not a violation. But pointing to the website, and then saying that a whois reveals that wiki-Bob's real name, not the pseudonym, is really John Doe, then that is outing. That should be clear from existing policy. But what about if the website also says that web-Bob's real name is John Doe (no need for whois)? Then pointing to it without actually saying the real name is still outing, and no one should think that it isn't. And this policy needs to make that clear. That third situation really isn't clear now, but my suggested change would make it clear.

I've been thinking on my own about possible scenarios, and here is something. Let's say that there is a sleazy website at "spam-wikipedia-for-money.com" Let's say the website says something like: "One of our associates edits Wikipedia under the screen name "username1", and edits for BigConglomerate and CompanyX. If you would like to hire "username1" for your company, contact us at..." And they go on with listings like that for "username2" and so forth. There's nothing wrong with pointing to that site. But if the site says instead: "One of our associates is John Doe from Anytown, Ohio, who edits Wikipedia under the screen name "username1"..." and so forth. Then editors need to know that they should not link to that website on-Wiki. Instead, they need to email the information privately. Of course, there will be a need to discuss on-Wiki the broad problem of "spam-wikipedia-for-money.com", but one can refer to it instead as something like "the username1 investigation". Now I guess there's a third scenario where the website is the way I described it first, but there is a sub-page that contains personal information. In that case, I'd say that linking to the main page is OK, but I'm receptive to discussing that.

I accept that the onus is on me to indicate what problem I'm trying to solve, and I did. But I also think there is an onus on editors who object to substantiate their objections, so long as I am discussing this in good faith, as I am. So, when is it ever OK to link to a website that reveals outing information? Is it ever acceptable, even under existing policy? Please help me understand what situations my proposal creates problems for.

And let me ask you this, in turn: am I correct that you would not want to add something like "If you are investigating COI or paid editing, it is not OK to out someone, but it's OK to link to a website that outs them." to this policy page? If not, what do we gain by trying to leave it ambiguous? Again, is it ever OK to link to a website that contains outing information? --Tryptofish (talk) 03:04, 14 January 2016 (UTC)

Sorry, I had no intention of implying you were trolling or even close, just that the conversation has boiled down to seemingly intractable positions twice now and this change does not really address the issues from last time. I do understand what you are trying to do here I just do not think the change you suggest adds more value to the policy than it adds potential for drama. The admins who enforce the policy and the good faith editors who read it seem to understand what it means and how to apply it and AFIK there are very few innocent transgressions. With the changes you suggest the chance of innocent or forced transgressions increases quite a bit because it takes away the need for intent. In the case above it is not necessary to point out that whois will tell you who wiki-Bob is IRL only that it will. If fact merely mentioning that wiki-Bob.com exists would be outing. This is a problem because there is a huge population who could get sanctioned under the policy and will have no idea what whois is or if they do simply did not think about it.

Thinking about it a bit more the best language I can come up with that closes the loophole you want and still does not create a strict liability situation is something based on do not post personally identifying information or provide directions for finding the same. This will make matters clearer to good faith editors. We write policies to guide good faith editors not to circumscribe bad faith editors we just ban those.

Personally I think if a person links their account directly to their username as you described above it should not be outing because they are not trying to remain anonymous and there is only some fiction that the names are linked everywhere on the internet just not here. Creating some bubble does no one any good. The only use for that is to make it harder to deal with paid editing - it is advertising and a shield to easy investigation. That said I do see a difference when linking to personal/social media sites because there may be different people of the same name, joe jobs in personal conflicts etc. but for commercial editors, not so much.

A problem with the "send it to an administrator" is most admins have not signed the Foundation's access to confidential information agreements. I would be much more comfortable if there were a private wiki or mailing list where all the members had signed the confidentiality agreements that could handle the paid editing investigations. ArbCom has too much on its plate and most people familiar with COI are not admins. PS - Please ping me. I am pretty busy and have turned on email notification of pings. Cheers! Jbh^Talk 15:22, 14 January 2016 (UTC)

@Jbhunley: thank you very much for everything that you said there. I sincerely appreciate it, and I think that we are making progress. I see your point about sending email to administrators, but the requirement to do so is already on the policy page, and nothing that I am suggesting would change it. I suppose that the editor doing it just has to use some judgment as to which administrator to send it to.

I see now what you mean about situations where an editor points to a website entirely in good faith, only to discover that the existence of something like whois would mean that they unintentionally got themselves into trouble. In fact, the things that I have seen, that got me coming here to raise the issue, have been, approximately, incidents in which something rather like that happened, and the user was blocked. There's a lot of gray area in what I'm talking about there, and I find it difficult to say whether or not these were "innocent transgressions" or "somewhat guilty but basically careless transgressions". I'm concerned about that, and I want to make sure that the policy is clearer about, in effect, warning good faith editors about stuff where they might unintentionally get into trouble.

I like the language that you suggested, and maybe we can use it as a starting point, to get to language that we can agree upon. For starters (only the green font is the proposed change; the rest of the paragraph is already on the policy page):

"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, or providing directions for finding personal information, on Wikipedia violates this policy."

As far as I can see, that language would satisfy what I have been trying to accomplish. Does it still need to be changed in order to avoid unintended harm? --Tryptofish (talk) 20:21, 14 January 2016 (UTC)

I can not see any blatant problem with that wording. Pinging editors from earlier discussion @SlimVirgin, EvergreenFir, Smallbones, and Flyer22 Reborn: @Johnuniq, Nagle, Minor4th, and Fluffernutter: @SageRad and Jehochman: to get their input. Jbh^Talk 20:40, 14 January 2016 (UTC)

Thanks! --Tryptofish (talk) 22:16, 14 January 2016 (UTC)

I object to this change for all the reasons already stated. I can't see any functional difference between the proposals. They will make things harder for editors, particularly those dealing with paid advocacy, and will make bad blocks more likely. SarahSV ^(talk) 22:20, 14 January 2016 (UTC)

Responding to ping (late). I think the policy already covers what you're trying to add, but I don't really have a strong preference. Minor4th 17:54, 19 January 2016 (UTC)

Tryotofish, I reverted the change to the order of posts because I intended to post in the previous section. I didn't want my post to appear to be a reply to something I haven't read. Just wanted to explain why I did that. SarahSV ^(talk) 22:47, 14 January 2016 (UTC)

That's fine, no problem. My apologies for misplacing your post. --Tryptofish (talk) 22:52, 14 January 2016 (UTC)

After plowing through the walls of text above and some of the archives, I'm not seeing much motivation for the change. COI editing is a major problem which requires daily attention on Wikipedia. Paranoia over "outing" makes it harder to deal with COI problems. Is "outing" a problem of comparable size? As a practical matter, real "outing" problems on AN/I have involved Edward Snowden and a transgender activist. Those are legitimate. Other complaints about "outing" involve political issues. Those are often legitimate. Less valid complaints about "outing" involve COI, advertising, sockpuppeting, and related promotional annoyances. Those are usually of the form "Why X is great, by someone who just might be associated with X". There, dealing with a COI issue can involve some degree of "outing", especially where X is an individual or a very small group such as a band. Don't make it harder to deal with that. Thanks. John Nagle (talk) 21:53, 14 January 2016 (UTC)

I'll reply to that, with anticipation (based on my prior experience) that this will be a frequently raised concern. First of all, there have been blocks for outing that have nothing to do with what you describe here. And I think that editors who feel that they have been outed would object very strongly to your calling that "paranoia". I've come to realize that most of the editors who closely watch this policy page are actually inclined to feel negatively about this policy, because they want to avoid any impediments to investigating COI and paid editing. I oppose COI and paid editing too, but that does not mean that it is OK to out users who violate COI and paid editing.

If you are going to object to the proposal on the basis that it "make[s] it harder to deal with", you have to actually explain how it might make it harder. It's not acceptable to just say "no" to any revisions. The page already says that personal information found while investigating COI and paid editing must not be posted onsite. It should instead be submitted privately by email. This proposal does not change that. It only adds, explicitly so no one misunderstands, that the same is true for "directions for finding personal information". That's it!

So if that's your objection, then you have to show how it is more difficult to email "directions for finding personal information" than to post it, and that there is nothing wrong with posting it instead. I don't think that anyone could defend a proposal to add, instead, a sentence like: "It is acceptable to post directions for finding personal information about COI or paid editors." The fact remains that it is never acceptable to post that. --Tryptofish (talk) 22:16, 14 January 2016 (UTC)

Thanks for the ping. I support everything said by Tryptofish in the above discussion. I think it makes great sense. Outing has happened to me and it sucked really badly, and the editor who did it kept pushing it and using the knowledge to poison the well in discussions, and it really did amount in one case to the editor saying something like "If you want to see who that person really is, Google them..." -- because i guess i chose a bad username when i signed up to Wikipedia, being totally unaware of the arsehole gaming of the system that does occur on a regular basis in controversial topics. The basic principle is that we ought to be judged on our Wikipedia edits and nothing more, and outing should be a sanctionable offence -- which it is of course, but not enforced well or strictly enough in my experience, not even by a long shot. So that one happened to me, and it hurt my ability to edit for a long, long time, and might have been the straw that broke the camel's back and led to my topic ban in the ArbCom GMO case, by making editing by principle harder for me, and making it easier for others to game the system by poisoning the well. So any good clarification to strengthen a strict no-outing policy would be good.

Relatedly, an editor above poses the hypothetical situation of someone saying "If you look in X's edit history, he used to edit as Y and is active with that name on Twitter," as being a case to not impose sanctions, but i think this is very much an outing, and should be sanctioned. How would that not be an outing? By the way, off-Wiki i'm an ordinary human, though i am interested in knowing about chemical pollution. I'm not paid for that, and it's a labor of love, as rivers near my birthplace were poisoned in the last two centuries, and i want to make sure that the knowledge about this is never lost, for those who are not exposed to history are destined to repeat it. That is not against any policy, and it's simply a point of view, but i edit according to the guidelines and policies. So anyone trying to out me to poison a well by painting a stereotype of who i am ought to be sanctioned for doing so. Judge a person's edits on their merit. SageRad (talk) 22:33, 14 January 2016 (UTC)

I would like to add to SageRad's comment that, at the time, this policy was interpreted as meaning that what happened to SageRad was OK. I hope that the proposed change would make that less likely to happen again. --Tryptofish (talk) 22:55, 14 January 2016 (UTC)

I have only a vague memory of what happened to Sage, but I'm not aware of any outing. I'll leave it there because I don't know what happened and don't want to step on toes.

But in general this is a good illustration of the problem. This is a made-up example, not about Sage or anyone else. New User:A says something contentious about a living person. Someone points out that it's a BLP violation because User:A is in dispute with the person off-wiki. This is the kind of thing we might need to know. Whether it's an OUTING violation is left to admin discretion, because it will involve taking lots of facts into consideration, including apparent intention, something this policy can't deal with. SarahSV ^(talk) 23:09, 14 January 2016 (UTC)

It seems to me that, in your example, the proper way to deal with it is to say that User:A is violating BLP because User:A's edits are disparaging to the living person. It really does not matter what User:A is doing offsite. But let's say that an editor feels a strong need to comment about the offsite dispute. If referring to the dispute is going to lead to outing User:A, then it should not be permitted, so providing those "directions" ought to be a violation. You don't get to out an editor because they violate BLP. But how hard would it really be to email the information about the dispute to a trusted user, who can then confirm onsite that you do, indeed, have reason to say that User:A has a COI about the BLP? --Tryptofish (talk) 23:19, 14 January 2016 (UTC)

Re: "It really does not matter what User:A is doing offsite." It might matter. See WP:BLPCOI. SarahSV ^(talk) 23:26, 14 January 2016 (UTC)

True, but what I was trying to say was that the BLP violations in this case would already be apparent on the basis of the edits themselves; comment on content and not the contributor, and all that. And it remains the case that one does not get to out User:A just because they violate BLP (see WP:HA#Harassment and disruption, second paragraph). --Tryptofish (talk) 23:43, 14 January 2016 (UTC)

They might not be apparent. We've had several cases of editors adding apparently innocuous text to BLPs, until it's realized that the editor has been pursuing the BLP in some way off-wiki too, and the edit is a continuation of it. There are lots of complex scenarios. SarahSV ^(talk) 23:56, 14 January 2016 (UTC)

Yes, and I guess there could also be a COI problem where the material being added is flattering. But as for the actual proposal here, I feel the need to repeat the question I asked above: But how hard would it really be to email the information about the dispute off-wiki connection to a trusted user, who can then confirm onsite that you do, indeed, have reason to say that User:A has a COI about the BLP? --Tryptofish (talk) 00:04, 15 January 2016 (UTC)

Another example:

User:X edits articles about Black Lives Matter, always from a position of appearing to undermine the campaign. His edit history shows that he used to edit as User:Unusual Name. Someone with that name regularly makes unambiguously racist posts to another website. Concerned editors open an AN/I to point out the connection and request a ban.

1. Given this proposal, would linking to the old user name be OUTING?
2. Would linking to the old user name and the off-wiki posts be OUTING?
3. What if text or links on the other website led to an apparent real name? Should the editors who opened the AN/I be blocked?
4. What if User:X/Unusual Name posted the text or links on the other website after the fact, so that he could claim he had been OUTED?

These are the difficulties this proposal would cause. SarahSV ^(talk) 23:25, 14 January 2016 (UTC)

OK, another good example to examine. One fish's opinions:

1. If the User:Unusual Name account is still onsite, and the edit history shows it's the same editor, then there is no outing in pointing it out. Once someone posts personal information here, it's not outing. This proposal does not change that, at all.
2. Not if the old user name is still posted here.
3. This gets to why Jbhunley's wording is an improvement. If the ANI post indicates that one can find personal information in that way, then it ought to be a violation. If however, the ANI post only points to a website that does not contain the personal information and there is no objective reason to think that the ANI poster intended editors to find the further links to personal information, then "directions" were not provided and there is no problem.
4. This after-the-fact gaming is a tough case. It was asked about above, and I said that I would hope that the good faith editor could say that the personal information wasn't there before, and would be believed. I asked before if this ever actually happens, and I'd still like to know the answer. If this is a significant problem, then I'd like to figure out a solution.

It seems to me that 1 and 2 are not affected by this proposal. --Tryptofish (talk) 23:37, 14 January 2016 (UTC)

But you can see the difficulties and arguments this will cause.

You're saying this would not be a violation: "User:A, who used to be UnusualName, posts off-site as UnusualName. Here are his racist posts." But if the real name is there, lots of people will argue, with your proposal, that the AN/I reporters would have seen it, and therefore they acted recklessly and did indeed "post directions." (This would be argued even without your proposal, but your proposal would strengthen that position.)

Re: the argument that we should email these details, realistically there aren't enough people to email them to, and it reduces transparency. Remember that these kinds of things happen regularly to stop spamming. SarahSV ^(talk)

Unless I'm missing something, these difficulties and arguments are not increased by the proposal. If the real name of UnusualName is right there, then it's outing (even if the person is a racist). And if the real name isn't there, but maybe can be found via further research, then it's not "directions", so the proposed change should not make things more difficult.

And as for there not being enough recipients for the emails, then the proposal here has no effect on that. The policy page already has that paragraph about emailing. My proposal does not create that paragraph. --Tryptofish (talk) 00:09, 15 January 2016 (UTC)

Either the proposal would change something or it wouldn't. If it would, it's a change in the wrong direction, because we shouldn't add anything that would make things harder for people. Jehochman wrote in November when this was first proposed: "What['s] important to remember is that outing is an intentional act done to harass somebody. We cannot codify every permutation." SarahSV ^(talk) 01:44, 15 January 2016 (UTC)

Let me say first that I really mean it when I say that I appreciate your discussing these issues with me. But, that said, I am sorry to have to say that that is sophistry. Of course nobody can codify everything. (And if we were to codify the definition of outing as only occurring when it is intentional, then everyone who violates it would have a get-out-of-jail-free card by just saying that they didn't intend it.) But the fact remains that we can codify some things. Saying that we cannot codify everything is a flawed argument to say that we cannot codify anything. There is nothing magical about the extent of detail on the page now. There is no reason to think that any addition will somehow break it. And this proposal won't break anything. The repeated statement that the proposal will "make things harder for people" is something that I have already, repeatedly, debunked. If editors think that any expectation of handling private information privately, by email, makes protecting against COI so very, very hard, then I suggest that they propose deleting the last paragraph of the outing section, because it already says that. To extend that beyond personal information, to also include directions for finding personal information, is not a burden on anyone.

And it is not a change in the wrong direction. I've already explained what it would improve, and, again, no one has shown that it will cause any harm. Looking over the discussion so far (and I remain happy to continue the discussion), the proposal is supported by EvergreenFir, Jbhunley, SageRad, and me. It is opposed by SarahSV, Smallbones, and John Nagle. It isn't the same proposal as the one on the talk page archive. And it is never OK to out someone, even if they have violated BLP, COI, or anything else.

I've done my best to respond thoughtfully to every concern (even at the price of tl;dr). But it is telling that I have repeatedly asked some questions of the editors who are skeptical, and these questions remain unanswered:

1. I've asked what is so difficult about emailing private information instead of posting it, noting that this proposal only extends that to "directions", because it is already an existing requirement for personal information per se. As far as I can tell, the arguments about it being too difficult are just excuse-making.
2. In contrast, the possibility that the proposal could make a violation out of something that should not be a violation is a serious one. A scenario that I keep hearing that might be like that is where the COI editor waits until there is a posted accusation, and then places private information on the other website as a gotcha. I keep asking if that has ever actually happened, and whether editors believed the good-faith editor who said that the personal information wasn't there before. I've never gotten an answer.
3. In response to a great many other scenarios that have been presented, I've tried to explain how the present proposal does not create problems, and no one has followed up by pointing out things that I have left unanswered, that I can make out. But maybe I'm missing something, so now, are there any other previously-mentioned scenarios where my answers are not good enough? I'm asking that very seriously.
4. Finally, I've asked whether editors would actually like the policy page to say that it's OK to out COI or paid editors, and if not, why not. All I'm hearing back is that it's best to leave things ambiguous, but no one seems able to defend the position that, even though no one really thinks it's OK, we should leave it ambiguous that maybe it might be OK.

I'm still very receptive to discussing anything that editors remain concerned about. But I think that if these questions really cannot be answered, then editors need to recognize that the weight of argument favors the proposal. --Tryptofish (talk) 23:52, 15 January 2016 (UTC)

Re: the gotcha question, I can think of one case where it appeared that something like that had happened, and a second where there was something similar. And I take the point that not everything can be codified, but I feel enough has been laid out already. Or, perhaps more accurately, I can't think of a way to improve the policy without introducing more difficulties than would be solved.

Another problem now is that your proposal – "or providing directions for finding personal information" – could include how to find it on Wikipedia. If someone says: "I recall that User:A said (diff) that she was a surgeon in Paris," and perhaps this was a long time ago and A had forgotten she'd said it, would that be a violation? And in case that sounds far-fetched, I've seen someone suggest that pulling on-wiki information together in that way might be outing too.

I also disagree with some of the examples we've discussed that you said would be violations, including (discussing a paid advocate) "I'm struggling to find a way to word this, because I don't want to violate OUTING, but I find myself in a situation where I can't even refer to a simple Google search." We have to leave room for admin discretion, because the issue is always: was there an intention to out (or was there such carelessness that it may as well have been intentional)? SarahSV ^(talk) 06:02, 17 January 2016 (UTC)

Thank you for the reply, and again, I sincerely appreciate the good faith discussion.

Let me first turn to the first of the questions that I asked. There does not really seem to be a good answer about how the proposal would create something that is actually more difficult (distinguished from creating unintended bad outcomes), so I hope that there is now a consensus that the issue of it being difficult is not a significant factor.

I think you make a good point that we should not make a violation out of pointing to something that is on Wikipedia. I think that's common sense, and I agree. Earlier in this epic discussion, I had the word "offsite" at the end of the proposed phrase (so as to restrict it to material that is not posted on Wikipedia), but that somehow got misplaced when Jbhunley and I were working out the previous wording change. We can easily fix this problem by putting the word back (the only material to be added is what is in green):

"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, or providing directions for finding personal information offsite, on Wikipedia violates this policy."

So there have been at least perhaps two instances of the "gotcha" gaming by posting material after the fact. It's still not clear to me from your answer what I asked, about the editors who got caught by it, whether they were able to explain that the personal material had not been there before, or whether they were unable to defend themselves. If it's the latter, then I'm worried about it, because it might be a case of introducing more problems than we solve. But if it's the former, then it comes down to how that weighs out, relative to the more-than-two instances where posting of "directions" has presented genuine problems. And I suspect that the proposal, in that case, solves far more problems than it might create. Feeling that "enough has been laid out already" is meaningful only if there really are problems being created. Again, as I said, there is nothing magical about the amount of codification at this particular time.

I asked whether there have been any other scenarios that have been discussed, where I said that it would not be understood as a violation, but where editors remain concerned that I have failed to address something. I'm not hearing anything there, so perhaps that has been settled.

But you do bring up a scenario where I said that it ought to be a violation, but you disagree, and would not want to make it a violation. It's the one about "I'm struggling...". We previously discussed it in the opening section, before the first break. Since that time, we've moved from wording about "indicating" how to find personal information, to Jbhunley's better language, about "providing directions". The hypothetical quote that you give probably falls a bit short of actually providing directions, so perhaps this is less of a problem. But my concern is a situation in which a Google search will lead directly to something that is way out of bounds, perhaps a Wikipediocracy thread that doxxes the editor and links to some very embarrassing personal information (I've seen this happen way more than two times). Context matters, but there can be situations in which the editor posting it is trying to draw attention to the offsite material, and is being catty about saying "I can't mention it, but...". But I think what is most important here is something that I said before: that an editor who self-describes as "struggling", if they are acting in good faith, is an editor who would have benefited from clarification. Thus, this question comes down to: should we clarify that it's OK to talk about a search in that way, or should we clarify that editors should find another way of saying it (such as "I'm struggling to find a way to word this, because I don't want to violate OUTING, but I find myself in a situation where I can't even explain how I know that there's a COI.")? But, in either case, this is a good argument for a bit more codification, not less.

And I guess that last point goes to my question about whether editors who oppose the proposal see an argument against making it a violation to out COI or paid editors, or would even go as far as saying explicitly that it is not a violation. So don't just take my opinion on it. This very day, ArbCom is issuing a decision about an instance of linking in this way. So please take a look here and, especially, here. And this was something that was both inadvertent and inaccurate! What I am saying reflects current practice. Leaving it out does not. --Tryptofish (talk) 20:35, 17 January 2016 (UTC)

I've had that very case in mind, not only that example, but the rules-following aspect of the whole thing, which includes reference to a CU block reversal after an SPI clerk said it was okay to undo it, and the blocked were a prof and her students. So no harm done, surely, which is what matters. The philosophy of the Wikipedia I joined was "if something needs to be done, do it and try to avoid collateral damage." Now every i must be dotted, etc. So I oppose instruction creep unless I see a very clear need for it. Back to this proposal – I don't see that need here.

You wrote above that "indicating" how to find personal information was now "providing directions," but it will amount to the same thing if someone wants to block for it. I want to avoid anything that might see good-faith editors landed with a block. The point of the struggling example isn't "let's make clear that he mustn't do it." The point for me is "let's not make it harder still by adding it to policy." SarahSV ^(talk) 21:05, 17 January 2016 (UTC)

I meant to add re: your example: "Context matters, but there can be situations in which the editor posting it is trying to draw attention to the offsite material, and is being catty about saying 'I can't mention it, but...'". Yes, and admins know to block for that. Intention is what matters, and all we can do is try to glean intention from the context. SarahSV ^(talk) 21:08, 17 January 2016 (UTC)

Again, I want to start by saying that, even though we disagree, I sincerely appreciate your discussion with me; I feel that needs to be said.

As I know that you realize, you and I are going around in circles, so I am especially interested in hearing from anyone else who might want to weigh in. I've been trying to focus the discussion on those issues that might still be genuinely unresolved, as per my 4 numbered questions, above. I asked if editors have been blocked or otherwise sanctioned after the crafty COI person belatedly posted personal information offsite, as a gotcha, and I'm not hearing that this has happened. So it sounds like there have been two instances when COI people did it, but the administrator judgment that you cite has been able to see it for what it was. That will remain the same if this proposal is implemented. No admin other than a fool would think that the added words require a block after such a gotcha.

You raise the question of whether there is a meaningful difference between "indicating" and "providing directions". The difference is that an offhand comment can indicate, whereas it must be a deliberate action to provide directions. (There is still, of course, judgment that comes into play at the borderline between the two, but that goes to the administrative judgment that you cite.) It seems very significant to me that the change was first suggested, and then supported in this discussion, by the very editor who reverted my first attempt to make a change. Only one editor has changed "sides" in this discussion, and that was him, so I take it as a measure of the direction in which sentiment has moved as we have tried to get consensus. This is one of those discussions in which consensus isn't going to be determined by counting !votes, so I think we have to look at the relative strengths of arguments (and again, I would welcome anyone else chiming in).

We both see today's desysopping by ArbCom as a decision that neither of us would have made. But that doesn't change the fact that it represents Wikipedia today. We both want to avoid seeing good faith editors getting blocked. But you can't have it both ways. If one worries that making things clear will lead administrators to exercise bad judgment, then one cannot also argue that leaving things less clear will be OK because we can trust their judgment. Either their judgment is good, or it isn't. Assuming some basic administrative common sense, in the context of present-day community sentiment about outing, the proposal will not really change how administrators already do things. They are already doing what the proposal says. ArbCom just did it, by an overwhelming vote. All the proposal does is tell editors to be aware that this is what most admins will do. If there is an editor like the one who was "struggling" in your example, as opposed to being catty, then we do them a favor by providing a bit more clarity.--Tryptofish (talk) 21:00, 18 January 2016 (UTC)

While I proposed the new wording as a possible solution I still see the potential problems the others point out and more or less agree with them. It would be more appropriate to characterize my view as "seeing your point and willing to be convinced" than "switching sides". That said, I re-read the outing policy today and found "Posting links to other accounts on other websites is allowable on a case-by-case basis"^WP:OUTING. I do not know how I missed it, blind or not paying attention before typing I guess :) @Tryptofish: how do you see the proposed change interacting with this affirmative statement that off-site linking is sometimes acceptable? There is a clear implication that the off-site account could out a person but it does not say "Posting... case-by-case basis unless it may out the editor". Jbh^Talk 21:49, 18 January 2016 (UTC)

@Jbhunley: thank you for the reply (and I'm watching, so you don't have to ping me back). Thank you also for correcting me about your position, and of course I stand corrected. To answer your question, I see a difference between "Posting links to other accounts on other websites" and "providing directions for finding personal information offsite", although I certainly agree with you that, now that I too take full notice of that sentence, there is a need for revision and clarification (and that would be needed even if we don't implement my proposal). But here is the key difference: "Other accounts" would seem to mean other screen names (akin to user names), as opposed to personal information. So if Username:X also posts at another website as Pseudonym:Y, linking there is not outing. But if that user posts at another website under their real name, linking to that absolutely is outing, and there is no "case-by-case" about it. If there is an implication that linking in a way that is outing really is OK on a case-by-case basis, I want to know what those cases would be.

Also, now that you are back here, I really would like you to be more specific than "I still see the potential problems the others point out and more or less agree with them", because I would like to see specifically which problems. Please also tell me what you think about the ArbCom decision that I just linked to, and also please tell me any answers you might have to the four numbered questions that I posted higher up in Break 4. Thanks again. --Tryptofish (talk) 22:10, 18 January 2016 (UTC)

I believe that with the changes that would need to be made due to the 'case-by-case' language that already exists we have gone beyond the limited scope of whether the proposed 'minor' addition is 'needed' to a fundamental change and tightening of the outing policy. There are now several issues 1) the simple addition as proposed here 2) harmonizing it with the 'case-by-case' language and 3) clarifying the the 'case-by-case' language. There are several options/combinations possible ranging from do nothing to add proposed language and strike c-b-c and variations between the two. That is something I would want to see much more community input from an RfC considering how this intersects with so many hot button issues. I am sorry I can not really address your questions because the explicit c-b-c language throws a spanner into the works. It was added in Feb 2015 by Doc James [1] maybe he can comment.

As to the ArbCom case - the outing principle looks standard and I do not think the FOF related to that had that much impact on the outcome. (ie it is interchangeable with any other big error in judgement to open the case and enough poor judgement was exhibited once opened that there was really no other outcome possible. Overall I do not think outing as outing played much a role at all.) Jbh^Talk 00:57, 19 January 2016 (UTC)

I was very surprised to see that WP:OUTING contains:

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

How is this not outing? Posting links to another editor's other accounts on other social media platforms? Even if it does not give direct access to the "real" name of the editor, it's still outing in the most serious and core sense. We're supposed to be editing here with this being its own universe in all cases except when it comes to serious threats or sock puppeting, as far as i have known. When was that line added to the WP:OUTING part of the WP:HARASSMENT policy? How is that not outing and not possibly harassment? SageRad (talk) 18:09, 19 January 2016 (UTC)

@SageRad: It was added in Feb 2015 [2] by Doc James. Jbh^Talk 19:14, 19 January 2016 (UTC)

P.S. -- i know this is off-topic and feel free to delete if this annoys you, but what's an easy way to search through page history and find when a certain phrase was added? SageRad (talk) 18:16, 19 January 2016 (UTC)

You can try WikiBlame. Jbh^Talk 19:14, 19 January 2016 (UTC)

There is now a link to this (under "external tools") on the revision history page. I'm not sure if you have to set any special options to get it. --Boson (talk) 19:33, 19 January 2016 (UTC)

Thank you very much. SageRad (talk) 21:46, 19 January 2016 (UTC)

I'm going to agree with both Jbh and SageRad. It sure does seem to me that, whatever the "case" might be, it is outing, so I agree with Sage about that, and the more that I think about it, the more that it concerns me. @Doc James: could you please explain your thinking about the phrase "case-by-case" in that edit? Thanks. And I agree with Jbh that this issue does indeed throw a spanner into what I was proposing, so for now, I'm about to take this discussion in a new direction. --Tryptofish (talk) 20:09, 19 January 2016 (UTC)

The case-by-case sentence was pointed out when this was first discussed in November. [3] E.g. see my post, during that discussion, at 19:12, 1 December 2015. SarahSV ^(talk) 22:14, 19 January 2016 (UTC)

Diff [4] SarahSV ^(talk) 22:18, 19 January 2016 (UTC)

Yes, indeed, you did mention it! It went by rather quickly at the time, amid all the other discussion, and I for one completely forgot about it until it was brought up again here. Looking back, I find it interesting that I replied to you that "It seems to me that "case-by-case" is applied by the current ArbCom differently than what editors are saying here." A third editor replied that the community, not ArbCom, should set policy, and asked why we should make a change here – to which I replied: "I guess the shortest answer to that question would be: to make sure editors know not to do what will get them blocked." --Tryptofish (talk) 22:27, 19 January 2016 (UTC)

I should add the specifics from what you said then. You listed ways in which "case-by-case" ought to be considered: "Was it neceessary to offer the information? Was it done with care or gratuitously? Did it happen within an ongoing discussion about COI, disruption and similar? Was the absolute minimum information offered to resolve the issue, or did the poster go beyond what was needed?" In my admittedly subjective opinion, the desysop that just happened at ArbCom cited an outing violation where the answers to most of those questions were the answers that I would think would be OK, rather than a violation. And yet... --Tryptofish (talk) 22:32, 19 January 2016 (UTC)

What ArbCom case? Are there more like it? --Ronz (talk) 23:46, 19 January 2016 (UTC)

From a short way above, in this talk section: "take a look here and, especially, here." --Tryptofish (talk) 23:56, 19 January 2016 (UTC)

Thanks! I didn't think to search the raw text to see what was being linked. --Ronz (talk) 00:39, 20 January 2016 (UTC)

Just above, in #Break 4, editors took note of this sentence that has been in the Outing section for not quite a year:

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

There are a lot of issues that arise around that sentence, but for the moment, I would like to focus on one specific issue, because it follows logically from the discussion just above. I've been thinking very hard about how to "un-stick" the discussion above, and the "case-by-case" language has given me an idea.

I think that it's very easy for editors to envision what kind of "case" would, in fact, be outing, and therefore not allowable.

But what exactly is a "case" where it is allowable?

I've been very concerned during the discussions up to this point that editors who oppose my suggestion are doing so kind of reflexively, out of a concern about making it too hard to investigate COI and paid editing. I've explained what my proposal would address, and we have dealt with some of the other objections that have come up. But it seems to me that the principal obstacle to consensus is the reasonable concern that we should not make a violation out of something that is not really outing – and that would be exactly what a "case" that is "allowable" would be!

So here is a request I want to make to other editors.

• Please provide real examples of such cases. Please don't just talk about it in the abstract, but actually provide links or diffs to incidents that have actually happened.
• And I'm going to insist that we focus on stuff that is reasonably current, so please limit it to between January 2015 and now – nothing older.
• Obviously, we cannot reproduce links that did violate outing, and such links should already have been oversighted.
• And it's not really interesting to look at incidents where the external links contained zero personal information, because everyone already agrees that it isn't outing.
• But I'm asking specifically for links or diffs from the past year, where an editor investigating violations of COI or paid editing or something else posted an external link as evidence, and the external link did contain reveal some amount of personal information – and that proved to be OK. Maybe there was no concern about an outing violation, or maybe someone raised the issue but it was ultimately agreed that there was no violation.

If you are of the opinion that "providing directions for finding personal information offsite" should not be considered an outing violation, because there are times when it is accepted as helpful in combating COI, then you should be able to provide examples. And such examples should be useful in determining what "case-by-case" really means, and thus, how the section might best be revised. And if no one can find such examples, then the argument that my proposal undermines anything seems to me to become very weak indeed. --Tryptofish (talk) 20:35, 19 January 2016 (UTC)

@SlimVirgin:@Jbhunley:@Smallbones:@Nagle:@Doc James: I'm checking with the editors who have expressed concerns about my proposal, as well as the editor who originally added the sentence being discussed here, to see if you have any examples of the type that I requested. Thanks. --Tryptofish (talk) 22:26, 22 January 2016 (UTC)

@SlimVirgin:@Jbhunley:@Smallbones:@Nagle:@Doc James: Per the comment below, I am re-pinging, with apologies if you get duplicate pings. Thanks again. --Tryptofish (talk) 22:58, 22 January 2016 (UTC)

Apologies Tryptofish, I have been distracted these last few weeks. Doc James (talk · contribs · email) 05:12, 23 January 2016 (UTC)

So to clarify is the proposal to ban the posting or urls to sites such as Elance? Often we post a url to Elance that says "someone is offering money to write an article on X" please watch that spot. Than a few days latter article X is created by an account (usually brand new) and one can see account on Elance that has picked up the job. You therefore have a probable link between an Elance account and a Wikipedia account that was created after the fact. Doc James (talk · contribs · email) 05:38, 23 January 2016 (UTC)

No need for apologies, and thanks for answering! I can readily understand that you have had more pressing matters to deal with of late! To answer your question, the opening part of #What is spurring this?, just below, is probably the best tl;dr that I can give you. But, no, no one is proposing something quite so limiting as that. If what is on Elance is that the Wikipedia username is shown as having taken on the assignment, then that is entirely appropriate to link to. The problem arises when the paid editor uses a pseudonym as their username and never discloses their real name on-Wiki, but the Elance link would reveal the real name. In that case, we are considering whether that would be outing, and I personally would argue that it is. If it is outing, then one could still email the Elance information to a trusted admin, and have that admin confirm that there is an external site that identifies the editor as a paid editor. --Tryptofish (talk) 22:40, 23 January 2016 (UTC)

• WP:Conflict_of_interest/Noticeboard/Archive_87#Sofitel_Luxury_Hotels - a complicated discussion over whether a editor doing promotional edits was affiliated with Oglivy Advertising. An ad for Wikipedia edits on eLance was involved. Note the section "I've concluded that one of their independent contractors in Singapore, name withheld due to WP privacy policy, is probably behind at least some of this editing and the recent sub-sub contract."

These two gives a sense of the relationship between outing and dealing with COI problems. These are just examples; there are more. John Nagle (talk) 07:54, 23 January 2016 (UTC)

Copied to here, from #What is spurring this?, by me. --Tryptofish (talk) 22:49, 23 January 2016 (UTC)

My apologies for being slow to get back to this, but I have looked carefully at this, and here are my thoughts. I'm having trouble seeing what this has to do with the revealing of personal information or outing. What I think I see is a mention of "off-wiki evidence" without any clues about how or where to find that evidence, mentions of a PR firm and a banned editor without anything to identify either of them, and then an involved editor showing up and voluntarily offering to help. Per John Nagle's summary above, I suppose that the editor revealed Singapore as a location, but that is hardly outing anyone, and beyond that, the editor is quite clear about not revealing a person's name, and states that more information was emailed privately to functionaries, which is exactly what this policy already says, and would continue to say, is the right thing to do. I'm left wondering why anyone would have thought that this would have run afoul of policy if any of the proposed changes here were to have been adopted. Am I missing something? --Tryptofish (talk) 00:14, 6 February 2016 (UTC)

• WP:Conflict_of_interest/Noticeboard/Archive_93#S.C3.A9page - Note the section "User Milstan, with apparent close connection to Sépage (see founder name) and FullSIX, is inserting content to articles such as Travel Website and E-Commerce as spam vehicles for mentioning Sépage."

These two gives a sense of the relationship between outing and dealing with COI problems. These are just examples; there are more. John Nagle (talk) 07:54, 23 January 2016 (UTC)

Copied to here, from #What is spurring this?, by me. --Tryptofish (talk) 22:49, 23 January 2016 (UTC)

I've now looked carefully at this, and the most conspicuous problem that I see is that the last comment in the thread ("Don't think you can just flick a booger on someone on your way out the door.") is pretty awfully incivil. The relation to the issues discussed here centers on an editor, Milstan, who registered that user name voluntarily, and who voluntarily added COI-looking content to some company pages. Discussing that does not violate anything. But I think that the key issue is where the editor said: "(see founder name)". I haven't looked up any names myself, but I can make a reasonable guess that the founder name of the company is easily found online, and in some way resembles the username Milstan – thus suggesting that Milstan may be this person in real life, something that I assume Milstan never stated onsite. In my opinion, there was no need to say "(see founder name)", and saying that did indeed cross the line into outing. I'm not seeing any good reason why we should protect the ability of editors to have posted "(see founder name)". But I'm sure that other editors disagree with me about that, and that is exactly what we need to discuss. --Tryptofish (talk) 00:26, 6 February 2016 (UTC)

@GeoffBrigham (WMF): I would be very interested in your thoughts about the phrase "(see founder name)" in this example, as it relates to the WMF polices on Privacy, and on Conflict of Interest, respectively. Thanks! --Tryptofish (talk) 22:23, 8 February 2016 (UTC)

Hi all, Geoff asked me to take a quick look at this on his behalf. First, a statement like "see founder name" doesn't impact the privacy policy. Finding public information somewhere else and letting people know about it on Wikipedia isn't a Privacy Policy violation. It might be a local policy violation against outing, but that's unrelated to the Privacy Policy. With regard to conflict of interest, it could be helpful information, but may also not be enough evidence to definitely prove a connection, depending on the quality of outside evidence and how strong the link is. Jrogers (WMF) (talk) 02:27, 10 February 2016 (UTC)

Thank you very much for that feedback. I think that goes a long way towards delineating the boundary between what is, and what is not, outing. Although I have previously been advocating in favor of clarifying that boundary in terms of making it clear that some kinds of edits would be a problem under the outing policy, I've been having second thoughts about that lately, in light of how other editors point out situations like this one, in which there is a need to avoid hampering investigations of policy violations. This clarification from WMF – that "finding public information somewhere else and letting people know about it on Wikipedia isn't a Privacy Policy violation" – strikes me as a very unambiguous position (and admittedly surprised me a bit). Although it does not prevent local consensus from setting a stricter standard, editor sentiment about COI etc. seems to me to be a good reason not to set a stricter standard here.

I'm thinking hard about the phrase "finding public information somewhere else", particularly in terms of what the word "public" signifies. It seems to me that if a user has willingly posted personal information about themselves elsewhere, it is therefore acceptable to cite it and link to it, even if that user has never indicated on-Wiki that they want that material to be public – and even if they subsequently claim that they had not intended it to be public. I'm thinking we should make it clear that editors should not be sanctioned for pointing to such information when legitimately conducting an investigation of COI or other offenses. On the other hand, there is also such a thing a "private information" that can also be found online (for example at a doxxing website), and I'm thinking that we should make it clear that pointing to that is not acceptable under this policy. Perhaps the delineating factor is whether or not the person has ever made that information public voluntarily.

What do other editors think about that? --Tryptofish (talk) 18:41, 10 February 2016 (UTC)

I think that it is situations like this that require the policy to be general. For instance a person's name, per the statement from WMF, can be private or public information. Private in the case of the average editor yet, as this case illustrates, public in the case of being publicly associated with an organization as "founder". Case-by-case gives the flexibility to address the nuance of each situation. Trying to build one size fits all policy for Wikipedia does not work by design. That is the whole point of IAR - what is best for the encyclopedia in a given situation is the proper thing to do. It also prevents gaming and allows the the community's implementation of policy to change in response to the changing needs of the project and views/values of its members. Jbh^Talk 18:49, 13 February 2016 (UTC)

Thanks very much for the feedback. I get it, that we must not attempt to get too formulaic, too algorithmic, too specific, because it would be counter-productive. But I continue to feel that the existing wording is too vague, too IAR, too cobbled together over time, in such a way that – despite all the claims that administrators always know it when they see it – I'm seeing first-hand that editors are getting blocked or not blocked in ways that are objectionably inconsistent. And the fact that you just said: "Private in the case of the average editor yet, as this case illustrates, public in the case of being publicly associated with an organization as "founder"." indicates that you parsed the difference between public and private the same way that I did. And I think it's the same way the WMF attorney did, and the way that most editors would. I'm going to let this sit for a few days, and see if there are any other comments. Then I'm going to make some suggestions in talk, and I think that these will be seen as an improvement, but not as a change to the status quo. --Tryptofish (talk) 19:29, 13 February 2016 (UTC)

One thing to keep in mind is that while the public/private delineation is, IMO, clear in this case, in others, like the self-employed or the 'volunteer' activist, things can get more fuzzy and even in cases similar to the one presented there are people who would argue it was OUTING. Personally I would argue if there is a clear linkage to a public enterprise (Company, campaign, organization etc) then the information gleaned from the official materials associated with that enterprise would be "public". I do not know that many would agree though. Jbh^Talk 19:47, 13 February 2016 (UTC)

Moved by me from #What is spurring this?. --Tryptofish (talk) 21:23, 27 January 2016 (UTC)

Here's a live example right now from WP:COIN: WP:Conflict_of_interest/Noticeboard#Herbertrafael. This brings the controversy into sharp focus. "Herbertrafael has, in the past, created Wardrobe Trends Fashion, which was speedy deleted twice. He recently created WardrobeTrendsFashion, which I nominated for deletion, as it was sourced to press releases. I later noticed that Herbertrafael added Herbert Sim to Oklahoma City University's list of notable alumni. Doing a web search on Sim reveals that he is the founder of Wardrobe Trends Fashion and is associated with ETHOZ Group and CEOWorld Magazine (for example, this press release). Google searches of LinkedIn show further connections. These two articles were also recently created by Herbertrafael and were sourced to press releases. I prodded them. In addition to those articles, Herbertrafael seems to have attempted to created a biography for Herbert Sim at Wikipedia talk:Articles for creation/Herbert Sim. Herbertrafael has avoided answering whether he has a conflict of interest, but his editing history would seem to indicate that he is either editing on behalf of Herbert Sim or he is Herbert Sim. I notice that on Wardrobe Trends Fashion, Herbert Sim's full name is Herbert Rafael Sim. NinjaRobotPirate (talk) 16:37, 26 January 2016 (UTC)"

The involved editor replies: "Also, whether or not, I am Herbert Sim, it is a personal matter, this is becoming a personal attack and harrassment should you wish to pursue this ground. E.g. If my username was something else, would this question even came up? WP:PERSONAL WP:OUTING" ... Herbertrafael (talk) 08:46, 27 Jan 2016 (UTC+08:00)"

There you have it. A COI editor, one whose editing history consists almost entirely of promotion for one company, insisting on their right not to be outed. Who's in the right here? John Nagle (talk) 04:58, 27 January 2016 (UTC)

• Why does Wikipedia believe that it is okay to treat COI and SPA editors as sub-human? Yeah, this is outing an editor. And it's wrong! --MurderByDeletionism^"bang!" 07:48, 27 January 2016 (UTC)

Matching a username to a person's name in the course of a COI report isn't harassment, nor outing. It's a common approach to sifting through problems with COI/SPA editing.

It is a case that gets close to being acceptable for oversight, so maybe we should be looking closer at WP:OS for guidance. --Ronz (talk) 17:42, 27 January 2016 (UTC)

• Not only is it OUTING, it's harassment. But I do agree with you on one point, harassment is a common approach on Wikipedia! --MurderByDeletionism^"bang!" 18:35, 27 January 2016 (UTC)

Why do you believe so, and do you think there's any consensus for this viewpoint? --Ronz (talk) 19:10, 27 January 2016 (UTC)

• Perhaps that is because I ask different questions. Like, for instance, why does any of this matter as long as the basic tenets of Wikipedia is being followed? And, the only answer to that question that I have been able to ascertain is to harass editors on Wikipedia. Now Wikipedia is probably 80% white males and how do we find enemies among ourselves? With nit picking and labels like COI or SPA. Reminds me of the Dr Seuss story The Sneetches. --MurderByDeletionism^"bang!" 20:48, 27 January 2016 (UTC)

If you cannot answer how, nor place it in the context of our policies and guidelines, then it's almost certainly not going to help change anything. --Ronz (talk) 20:57, 27 January 2016 (UTC)

• Hmmm . . . I read "Why" not "How", so I'm getting the inkling that this reply has more to do with goalpost moving than anything else. Still doesn't change things, outing is outing, no exceptions, and therefore it's wrong. --MurderByDeletionism^"bang!" 04:28, 28 January 2016 (UTC)

This is exactly the kind of example that I was looking for, thanks! (I am going to respond to examples 2 and 3 too, when I have time, but I've recently been side-tracked, but that doesn't mean that I've left the discussion.) And I think that the discussion above demonstrates that editors have sharply conflicting views about how to balance COI and OUT, and I think comments on both "sides" are good-faith, so it is useful to figure out what the community consensus is, since it really isn't clear. I waited a day before commenting, because I wanted to look back at the COIN etc., to see if there had been any sanctions related to outing, and there were not. In my opinion, this sits right at the edge of what is, and what is not, a violation, and it falls just on the side of being OK. The editor voluntarily used the username Herbertrafael (before, apparently, changing the username following the COIN). He edited in a manner that reasonably drew attention to the possibility of a COI, and did not respond helpfully to initial inquiries. He made repeated edits about Herbert Sims. Taken together, there was sufficient evidence voluntarily disclosed onsite to look into a COI connection offsite. Thus, searching offsite for Herbert Sim and finding a connection to that company was not outing. The one thing that comes close to the borderline was reporting publicly that the middle name is Rafael. That is disclosing personal information beyond the minimum necessary to examine the COI, and it makes it quite personal, in that it moves from being either being Sims or editing on Sims' behalf, to being Sims. That's gratuitous, because editing on Sims' behalf would be the same thing, in terms of COI policy. But then again, there was that username. So it comes close to being a violation, but in my opinion does not quite become a violation. --Tryptofish (talk) 23:02, 28 January 2016 (UTC)

In light of Example 3, I now think that there is no privacy problem over the middle name, because the middle name had been made public information voluntarily. --Tryptofish (talk) 18:45, 10 February 2016 (UTC)

That ping may not have worked because it also edited some other text. Is there an actionable proposal on this page? It seems to have been circling for months with one editor wanting to harden wording to move actions that might currently be debatable to a definitely-outing category. That is not helping the encyclopedia. Please find something else to do. Johnuniq (talk) 22:47, 22 January 2016 (UTC)

Is there a proposal? Yes. Do you have an example to present, or are you just going to make personal attacks against me? Have you read the comments here from other editors (not just me) who support the proposal? --Tryptofish (talk) 22:56, 22 January 2016 (UTC)

Whether somone can think of an example that cover all future possibilities is not relevant. I explained on 28 November 2015 that the DUCK principle applies to policies and it is not necessary to define exactly what outing is. There is no personal attack. Johnuniq (talk) 23:37, 22 January 2016 (UTC)

If you had bothered to read anything here, you would know that I did not request "an example that cover all future possibilities". I'm looking for a very specific kind of example, to help evaluate a specific question. I certainly hope that your responses here will not turn out to be representative, but if so, then the inescapable conclusion is that the arguments that have been made in opposition are without substance. --Tryptofish (talk) 23:43, 22 January 2016 (UTC)

This has been going on for two months. If there is no actionable proposal, please give it a rest. The talk page of a major policy is not the place for someone to seek a very specific kind of example for two months. Johnuniq (talk) 00:05, 23 January 2016 (UTC)

And now, you are ignoring the fact that I just told you that there is a proposal, and that other editors beside me are supporting it. It is not the same proposal as in the archived discussion, and as the present discussion has gone along, other issues with the outing section have been identified and are being discussed. Please do not talk to me like I am a troll. Telling me to "give it a rest" is not a reasoned argument on the merits. --Tryptofish (talk) 00:09, 23 January 2016 (UTC)

Your comments might acknowledge that a two-month discussion is excessive. Is your aim to alert others so they avoid the situation that led to this block? Such an aim is not achievable because people who might be helped are not going to study the exact wording currently in this policy before posting a link. It's better for the policy to use general language and rely on the DUCK principle—simple is good. Adding more verbage is unlikely to do anything more than give wikilawyers something to argue over, with the side effect of making it much harder for editors to discuss ongoing problems of off-wiki planning to subvert Wikipedia's procedures. Johnuniq (talk) 00:34, 23 January 2016 (UTC)

I saw this comment earlier today, and I was so offended by it that I decided that it would be best for me to wait for several hours before replying. Did I first become interested in this policy as a result of what happened to me? Yes, of course. But no, I am not merely trying to address that. I have thought very carefully about the policy implications as a whole, and I am trying to make things better. There was a proposal that I made, that is now archived. I accepted that it did not have consensus. Here, I am making a different proposal, and furthermore, the ideas being discussed have changed as other editors have responded. Most of the lengthy discussion just above is just between me and one other editor, SarahSV, in a friendly exchange of ideas and possible scenarios. For you to paint that as a single discussion that has gone on for months is false. Believe it or not, I have been editing Wikipedia in many other ways throughout this time. And I have told you repeatedly that other editors besides me have said here that they agree with my concerns, and some of them have pointed out problems that have nothing to do with what happened to me. In fact, numerically, the support and opposition is roughly balanced in numbers, with most of the opposing editors being quite open to discussion, and giving serious consideration to a more extensive revision, instead of just adding a few words. I get it, that paid editing is a serious issue, and I have no intention of making it more difficult to deal with. I'm listening seriously to other editors, at least so long as they have better things to say than to try to read my mind and ascribe maladaptive behavior to me. You keep arguing that it is DUCK to distinguish between what is, and what is not, outing, yet you also seem to think that what I am discussing is a borderline case. You claim that you don't want to see me "not helping the encyclopedia", yet you offer nothing helpful to improve this policy page, while you seem quite dedicated to telling me that you can read my mind and ascertain my motivations. --Tryptofish (talk) 23:49, 23 January 2016 (UTC)

Maybe someone should summarize the discussion with each break?

First, I'm unaware of strong consensus for such a change. What is spurring this? Anything from ArbCom? Given how much difficulty we're having enforcing the COI/FCOI changes over the past few years, it seems that the proposed changes here could undermine our need to reduce COI violations. --Ronz (talk) 20:23, 19 January 2016 (UTC)

For a start, please just read from the top. I'm not eager to repeat everything that has already been said, and there are already multiple replies to whether it really undermines anything. The purpose of the discussion is to find out whether or not there is consensus; nobody is preemptively assuming that consensus already exists. --Tryptofish (talk) 20:29, 19 January 2016 (UTC)

I notice others are concerned about the huge amount of text with no summaries. I'm with them.

Could you just answer the question? What is spurring this? --Ronz (talk) 20:36, 19 January 2016 (UTC)

I'm trying to be careful about my giving a short reply, and then someone playing gotcha with me over something that wasn't in my short reply, but that was in fact discussed thoroughly already. The policy page makes it clear that posting personal information about another user is outing, but it is ambiguous about posting information that could lead people reading the post to find that personal information off-Wiki. Some editors are concerned that the latter is needed, as something permissible in fighting COI; other editors are concerned that there have been blocks of editors who have done so, and that other times there have not been blocks when there should have been; there are real examples of these things. Just yesterday, ArbCom desysopped someone, making it very clear that it is outing to link to information offsite, even if doing so was both unintentional and inaccurate. Those are some highlights, but they are not complete. --Tryptofish (talk) 20:51, 19 January 2016 (UTC)

Thanks! --Ronz (talk) 21:09, 19 January 2016 (UTC)

I am definitely concerned now, as the rule against outing is important, and i thought it was pretty solid. But having this exception in it for linking to "other accounts" on a "case-by-case" basis makes it much weaker. It opens it up to abuse. Outing is a serious thing. I think the outing policy needs to be strengthened, and here i find out it's actually weaker than i thought because of a recent addition (February 2015). I await more discussion on this, and thank Tryptofish for opening up this discussion that revealed this. SageRad (talk) 21:55, 19 January 2016 (UTC)

Is "outing" undesirable in a commercial context? Paid editors are required to disclose their affiliations. Some fail to do this. I would argue that WP:DISCLOSEPAY should override WP:OUTING. Paid editors have been located from off-Wiki sources, such as ads for Wikipedia editing. The Banc De Binary episode, where BdB offered $10,000 to anyone who could make info about their troubles with the SEC disappear from Wikipedia, is a well known and highly publicized example. This is a reasonably common problem at WP:COIN. Don't take away tools we need to stop paid editing. John Nagle (talk) 23:17, 22 January 2016 (UTC)

Their affiliations, yes. Their COIs, yes. Their real life names or home addresses, no. Just above, I pinged you to ask for a specific example of what you say here is a common problem at COIN, and I hope that you can provide a helpful example. --Tryptofish (talk) 23:25, 22 January 2016 (UTC)

Some examples:

• WP:Conflict_of_interest/Noticeboard/Archive_87#Sofitel_Luxury_Hotels - a complicated discussion over whether a editor doing promotional edits was affiliated with Oglivy Advertising. An ad for Wikipedia edits on eLance was involved. Note the section "I've concluded that one of their independent contractors in Singapore, name withheld due to WP privacy policy, is probably behind at least some of this editing and the recent sub-sub contract."
• WP:Conflict_of_interest/Noticeboard/Archive_93#S.C3.A9page - Note the section "User Milstan, with apparent close connection to Sépage (see founder name) and FullSIX, is inserting content to articles such as Travel Website and E-Commerce as spam vehicles for mentioning Sépage."

These two gives a sense of the relationship between outing and dealing with COI problems. These are just examples; there are more. John Nagle (talk) 07:54, 23 January 2016 (UTC)

Thank you very much, John Nagle, for those helpful examples. This is exactly the kind of information that I am looking for. I am going to copy what you have posted to the section above, as #Example 2 and #Example 3, where we can discuss them further. --Tryptofish (talk) 22:44, 23 January 2016 (UTC)

Coming late to this matter I may not have grasped the whole thing, but on the face of it I would also be unhappy about weakening the response to paid editing (and to disruption generally), and would see tightening up the restrictions here as doing just that.

I'm also not clear about the reason for the proposed change (which, I would agree, looks innocuous in context) Am I right in thinking you simply wish to warn people that despite it not being explicitly discountenanced, in practice editors are being blocked for this very thing? If so, changing the policy as a way of warning people is the wrong way to go about it; such a change would simply write the practice into the rulebook. If people are being blocked unfairly then the place to tackle it is through an appeals process; and a better warning would be to add a footnote here, something like “though not explicitly forbidden, editors should be aware that ….. has been the cause of sanctions being imposed on those doing so”.

I can also see that treating matters on a case-by-case basis may also not be a bad thing, as it suggests a range of factors being taken into consideration, not just the one.

And, like John, I am not convinced that outing a commercial or disruptive interest is a bad thing, or undesirable. Protecting individuals from malcontents is one thing; protecting malcontents from their just deserts is another. Xyl 54 (talk) 00:30, 24 January 2016 (UTC)

Thank you very much for your comments here. And I want to let editors here know that you and I have a parallel conversation going on at my user talk, which anyone can feel free to see. As I said to you at my talk, I'm starting to realize the ways in which what I have been saying raises problems, because editors have understood it as prohibiting things that I don't want to prohibit. I agree with you and John Nagle, that we should deal with commercial and disruptive interests! But let me make the following distinction: what you called "just deserts" ought to mean being separated from Wikipedia, or compelled to edit according to policy. It does not mean public shaming; see the second paragraph of WP:HA#Harassment and disruption.

To some extent, editors are being blocked or otherwise sanctioned for this; see here and, especially, here (note: no appeals!). To some extent, editors here have pointed out a lack of protection from abuses of the same thing; see [6]. There is a need to find a consensus that better draws the line. And personally, I reject the argument that the best thing to do is to make things as vague as we can get away with, and somehow trust that everything will be fair because of good judgment. --Tryptofish (talk) 20:19, 24 January 2016 (UTC)

I'm looking much more critically at this policy now, and I see in the middle of the first paragraph this sentence:

"This applies to the personal information of both editors and non-editors."

("This" is referring to outing.) That doesn't make sense to me. How can someone out a person who never edits Wikipedia? --Tryptofish (talk) 00:37, 20 January 2016 (UTC)

It's very old, from Jimbo [7] later changed to "non-editors" [8]. --Ronz (talk) 00:47, 20 January 2016 (UTC)

Yes, that certainly is very old! I'm wondering whether it's now an anachronism that should be removed. --Tryptofish (talk) 00:59, 20 January 2016 (UTC)

It's surprising to have here, but consistent with BLP and NOT. --Ronz (talk) 17:26, 20 January 2016 (UTC)

That's a good point, but strictly speaking it's a BLP violation rather than outing, and it might be helpful to make that more clear than the page does now. --Tryptofish (talk) 18:14, 20 January 2016 (UTC)

• I'm inclined to leave it as is. If someone points to another place that is outing an editor in an attempt to obviously out them, they are going to get blocked for outing now. I understand the desire to clarify policy, but I think the original reason for the change is already covered by policy. I know I would block. As SarahSV state early on, the problem is consistency, not clarification. Dennis Brown - 2¢ 14:59, 7 February 2016 (UTC)

The thing that keeps getting to me is that there is indeed a problem with consistency, so either we have dimwitted administrators (said facetiously), unclear policy, or both. I don't get the logic of saying that there is a problem with consistency, so let's not change anything. --Tryptofish (talk) 19:18, 7 February 2016 (UTC)

What is and isn't outing is sometimes a matter of interpretation, sometimes hints are given unintentionally, and admin aren't mind readers. The obvious outing cases always get blocked quickly, then RevDel'ed (or OS'ed). Sometimes someone might make a reference to an offwiki site and it isn't obvious that they are outing someone, or you aren't sure if the editor hasn't already admitted who they are. If I don't know the outed editor, I have to first go make sure they haven't said who they really are here publicly, and sometimes, editors have and will lie about it. This is not an easy thing to do and I might not have the time. In that case, I'm not going to block anyone. The problem isn't really the policy as much as it is simply very difficult to prove they haven't disclosed, and the other editor's intention was to out them. Both of those must be in place to block, which is the problem: verification. Admin are already pretty good about construing the policy broadly if it is a clear cut case of outing someone who has never disclosed who they are, regardless of how it it pointed out. In my eyes, it doesn't make any difference where the material is hosted, outing is outing, and I'm pretty sure that most admin feel the same way. Accuracy of the information isn't a factor either, we really can't verify anyway. We just don't want to make a bad block because a block log entry for OUTing is a pretty serious thing. Dennis Brown - 2¢ 01:21, 8 February 2016 (UTC)

Let me throw a hypothetical at you: User:ABCD goes on Reddit and says "Yes, I"m ABCD on Wikipedia, but my real name is John Williams", and User:XYZ says "On Reddit (link), you said you were John Williams, is that true?" How do you handle that? I think you have to wait for ABCD to complain, then RevDel, then find out what is really going on. Really, he outed himself on another site, so I'm not likely to jump to the block button until I have a full grip of the situation and understand if it was really malicious outing, a Joe Job, or something else. Another admin might draw a bright line and just block because he linked to page that put a name to the user name here. I would disagree with that, but it would still be a good faith act. That is why it is inconsistent. The proposed change has nothing to do with this kind of problem, and dozens more I can think of. You will never get perfect consistency with this complex of an issue. Dennis Brown - 2¢ 01:50, 8 February 2016 (UTC)

I guess the first thing that I need to say is that my previously-proposed change is no longer a change that I am seriously contemplating, for a variety of reasons, one of which how another editor pointed out somewhere above that another sentence, one that refers to case-by-case examination, throws a "spanner" into what I had been proposing. For the hypothetical that you gave, I think I would agree with you, albeit with the caveat that it could have been a false flag at that other site. But, as you just said, another admin might have blocked for that. Perhaps for you, the fact that the other admin did it in good faith is good enough, but for the blocked editor, that might be a very valid reason to be offended. I agree with you about perfection being impossible, but I don't think that is a reason to rule out improvement. Obviously, I'm not going to attempt to force any change in wording on anyone without consensus, but I am sincerely interested in figuring out how to make the outing section better written. For now, I'm primarily interested in the talk sections above that are about "examples". If you feel like it, I would welcome your thoughts on any of those examples. --Tryptofish (talk) 22:16, 8 February 2016 (UTC)

And I trust your faith in doing so, you obviously really want to improve things, but I would just remind you that the written policy is nothing more than a mirror image of previously established practice and consensus. It's easy to change words, it is harder to change practice. For what it is worth, I would start with private or talk page discussions with admin that work a lot of outing cases, (that wouldn't be me) and get their perspectives, what they consider to be the problem with policy. It is one thing (and a good thing) to hear from the community at large, but the people that have to actually enforce it might have different ideas. Dennis Brown - 2¢ 00:34, 9 February 2016 (UTC)

If nothing else, a critical read of the section (portions of which haven't changed that much since none other than Jimmy Wales wrote the first version, with subsequent additions being pasted in later without really making it work together) shows that a good copyediting would not hurt. --Tryptofish (talk) 01:17, 9 February 2016 (UTC)

Those interested, please see Wikipedia:Village_pump_(policy)#WP:ARBPOL.2C_Harassment.2C_and_Private_hearings. Evangeliman (talk) 16:58, 29 March 2016 (UTC)

I had a curious question about the Outing Policy. What happens if a user defends a source on Wikipedaa for an article, but does so because it is a book or article that they themselves have written. Can another editor bring up the fact that the user is source's author, thus confirming their "real world" identity? This would seem like outing but also might be necessary to keep NPOV. -207.245.177.1 (talk) 17:49, 29 March 2016 (UTC)

I believe This would fall in line with Conflict of Interest investigations. There are policy discussions about this at the moment (has been going on for a while) so my recommendation would be to see if you can take any personally identifying comments off-wiki. I would start with a post at COIN, probably with the source, but not identifying which editor you think is the author. When someone else gets involved, you could send them your ideas via email, thusly keeping you within bounds of OUTING. Though it would be best if you had a user account with an attached email. --Kyohyi (talk) 17:58, 29 March 2016 (UTC)

I work at COIN a bunch.

• If someone is editing under their real name of some obvious abbreviation of it (or if they actually write something like "I wrote this, it is great", on-wiki) it is completely reasonable to first go to their talk page and write something like, "It appears based on your username (or what they wrote, with a dif) that you are the author of the work you are trying to add. Please do see WP:SELFCITE which is part of Wikipedia's COI guideline. Instead of directly adding the source to the article, please bring it to the talk page for discussion." (as with any behavioral dispute or issue, it is always best to try to talk to the person at their Talk page to resolve it, per the relevant part of DR.
• is someone's username is not obvious or they haven't OUTED themselves, but they have some pattern of editing (like they are spamming the reference into a bunch of articles or some other pretty clear pattern showing a connection), you do the same thing but just step back a step, and go to their talk page and ask something like. "Based on your pattern of editing you seem to have a connection to X, who is the author of that publication. I want to be sure you are aware of WP:SELFCITE - please read that and the rest of the COI guideline. You probably should bring that source to the Talk page for discussion, and would you please disclose any COI you have if there is one?" Something like that. Simple, not accusing, but direct and polite.

• if either thing fails, then going to COIN makes sense. Please be careful not to mix up content disputes with discussions of COI - that can get very messy, very fast. Jytdog (talk) 23:26, 29 March 2016 (UTC)

Hi Mike V, I reverted your removal of "Posting links to other accounts on other websites is allowable on a case-by-case basis." This was added [9][10] after an RfC. [11] Whenever it has been discussed since then, it has had clear consensus (at least so far as I've seen). As a matter of practice, it happens regularly and is often unavoidable. SarahSV ^(talk) 03:06, 30 April 2016 (UTC)

I agree that it's appropriate to keep it on the page. There has been and will continue to be discussion about revising it, but until such time as a consensus to change it might occur, it seems reasonable to leave it on the policy page. --Tryptofish (talk) 21:43, 30 April 2016 (UTC)

@Od Mishehu: This very issue is being discussed in an Arbcom case [12]. There is no concensus for it there so let's please wait until it is hashed out there before changing policy. Jbh^Talk 20:58, 10 May 2016 (UTC)

My first reaction to seeing this edit and revert was to feel that I should wait on the planned RfC (above on this talk) until the ArbCom case is resolved. No point in examining changes to policy in possible uncoordination with the pending decision there. And I probably will wait for that.

But on the possible merits of the edit, it seems to me to consist of two separable parts: "if noticed by an administrator, it may be [[WP:REVDEL|redacted]] while waiting for the oversighters getting to it," is the first, and "and this may even be done when there is a [[WP:COI|conflict of interests]]" is the second. The first part is entirely noncontroversial insofar as I can tell. It seems to be current practice, and I think it might be useful to point it out explicitly in this policy. After looking at the ArbCom case, I see that the second part is the issue here, in that "COI" in this case relates to WP:INVOLVED. I agree with Jbhunley that we ought not to specify in this policy that INVOLVED does not apply here unless ArbCom yields some reason to think otherwise, or if subsequently the community reaches a clear consensus to make such a change. --Tryptofish (talk) 23:37, 10 May 2016 (UTC)

• While the discussion in the ArbCom case (linked above) does cover involved revdel, it does not (to my knowledge) cover "posting of personal information" - which is the section discussed above as being edited by Od Mishehu & reverted by Jbhunley. I consider that there may be (an as yet unestablished) concensus that involved revdel of personal information is acceptable, but that involved revdel more generally is not. In either case, I concur that a consensus should be established, and that it would be prudent to wait for the ArbCom case to close. Hope this helps. - Ryk72 ^'c.s.n.s.' 23:51, 10 May 2016 (UTC)
  • @Jbhunley:If it were the case that Gamliel thought that the edit in question was oversightable, and immediately contacted oversight in the matter, then your point would be accurate; however, this is probably not the case. Please also see Wikipedia:Revision deletion#Hiding oversightable material prior to Oversight, which explicity says "Even if the material is ultimately found not to be suppressible, administrators are allowed to err on the side of caution, even in cases with an apparent conflict of interest, provided it is in good faith and they quickly seek oversighter review" (emphasis in original), which explicitly agrees with my point. עוד מישהו Od Mishehu 02:58, 11 May 2016 (UTC)

• I can't see a problem with the edit: "if noticed by an administrator, it may be redacted while waiting for the oversighters getting to it, and this may even be done when there is a conflict of interests." WP:BLP says something similar: "Administrators may enforce the removal of clear BLP violations with page protection or by blocking the violator(s), even if they have been editing the article themselves or are in some other way involved." The point with these issues is to act quickly. SarahSV ^(talk) 03:50, 11 May 2016 (UTC)

The only objection I have is the topic seems to be the subject of discussion in a high profile case and it looked to be preempting that discussion. I do not have an issue Oversightable material being revdelled by whoever first sees it and the "while waiting for an oversighter to get to it" addresses the matter of unreviewed COI/INVOLVED action, ie COI revdel of non-oversigntable material without contacting Oversight. Jbh^Talk 07:00, 11 May 2016 (UTC)

PS if I misread the applicability to the ongoing discussion at the case, and it seems I have, I remove my objection. Jbh^Talk 07:03, 11 May 2016 (UTC)