This is an archive of past discussions on 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.
The first sentence of WP:OUTING wording forbids the use of any and all offwiki sources. It says,
Posting another editor's personal information is harassment, unless that person had voluntarily posted his or her own information, or links to such information, on Wikipedia.
As written, this is too broad and not in line with community practice. The article on Johann Hari, for example, discusses his pseudonymous and abusive Wikipedia editing, because it was discussed in high-quality reliable sources. It provides links to the press articles that outed Hari. Similarly, Wiki-PR editing of Wikipedia discloses the legal names and job titles of Wiki-PR's principals, again because this was a matter discussed in reliable sources. And there have been other cases. Community practice has been that once an editor has been outed in the press, the information is public and no longer protected. Consensus seems clear on this point.
In other scenarios, consensus is not clear.
There is presently no consensus on what to do when paid PR writers voluntarily and publicly disclose their involvement in Wikipedia editing offwiki, for example in professional profiles, press releases or public bids for paid editing contracts, as discussed in the sections above.
There is similarly no clear consensus on what to do when academics voluntarily disclose their Wikipedia activities in their academic writing, providing information that allows others to identify their Wikipedia account. Should it still be considered "harassment" if someone speaks their name on Wikipedia? To my mind, if someone publishes an academic article saying they did a particular thing on Wikipedia, then that means that they are happy for their peers and the world at large to know that they, as a named person, did that, and are in no way concerned about remaining anonymous.
Another situation is when pseudonymous Wikipedia editors voluntarily speak about their Wikipedia editing to the press, and press articles quote them under their legal names, describing them as, say, a GLAM project coordinator, providing enough information to allow their Wikipedia account to be identified. Should onwiki mentions of their legal name be considered harassment? Again, to my mind, if an editor provides details about their Wikipedia activity to the press, under their legal name, then that indicates that they are not concerned about remaining anonymous, whether or not they have made a prior disclosure of their identity somewhere on Wikipedia.
So a way forward would be to gauge community consensus on which sources of information, other than self-disclosure on Wikipedia, should be sufficient to establish that a pseudonymous editor's personal information is no longer private, and to add those to the "unless" clause. For example:
Posting another editor's personal information is harassment, unless the personal information in question
has been voluntarily posted or linked to by the editor themselves on Wikipedia;
is contained in WP:BLP-compliant reliable sources discussing the individual's Wikipedia activity;
has been disclosed by a PR writer in public statements about their paid-editing activity (for example in press releases or publicly viewable bids for paid-editing contracts);
has been disclosed by the editor in the context of their academic writing;
has been disclosed by the editor in a press interview.
In the follow-up RfC, we can have people voting on which of the clauses numbered 2 to 5 (and any others people can think of) should be added to the policy wording. --AndreasJN46609:09, 24 August 2016 (UTC)
I broadly agree with you that we need some sort of follow-up RfC along these lines, and I also think that we need to put a lot of careful thought into composing that list, before we actually go ahead with the RfC.
It seems to me that #3 on your list is the important question, and we would do well to expand it into multiple scenarios, with differing degrees of promotional intention and of personal revelation. I also think that #2, 4, and 5 are relatively less important; previous discussion suggests that most editors here are less interested in those issues than you are.
I also think that we need to be very clear about the difference between an academic publication or press interview in which the editor actually volunteered their account username, versus such sources in which the editor said under their real name how they edit, but did not actually release their username, and the username is identified by connecting the dots. The latter may well be outing even if the former is not.
I'd like to add to the list a case where the personal information is at some sort of social networking site, such as Facebook. And also a case where it is at a personal blog.
As for #3, we could differentiate between someone who is, and someone who is not, being paid and not disclosing it here. We probably also have to parse some aspects of what is, or is not, a "public statement". Does it include a website for soliciting editors for paid assignments, a personal commercial website, a corporate PR department website, a personal blog, a social networking site? Does it matter if it was posted publicly but later taken down? Or if it was taken down right after a COI issue was raised on-site? How much personal information can be linked: is it OK to link to things like social security number, phone number, or home address in such contexts? What if the linked name is not an exact match of the account name here?
Although this discussion has gone a bit quiet lately, I've been thinking about it a lot, and I really do think that we will need a new RfC about this, about defining the borderline cases. But I also think that we will get the most mileage out of such an RfC if, first, we pin down some community sentiments on some underlying issues. The recent RfC about WP:COILIST did some of that: [1]. But I increasingly think that there is another point that we should examine before the RfC: the issue of how blocking should be used in cases of outing. Here's why: in the RfC planned here, the community might want to say that a particular example is something that should be discouraged, but that it does not rise to the level of being block-worthy.
Thinking about some of the other recent discussions on this talk page, one of the divisions in opinion has been roughly along the lines of functionaries and some editors feeling like outing is so bad that there should never be exceptions in COI cases, and some other editors feeling that some COI situations should not be protected by the outing policy. In my personal opinion, some of what I heard from functionaries sounded somewhat like: outing and harassment are so bad, and such big problems, that we really must put the priority on shutting it down. And there have been some high-profile outing blocks where there was controversy about whether or not the blocks were really preventative.
And the policy as it is written now says in part: "attempted outing is grounds for an immediate block." That makes it sound like the block is automatic, without stopping to consider whether there is a realistic risk of continued outing that needs to be prevented. I really don't know whether that actually reflects present-day community consensus. And if not, we obviously should revise it.
So, I'm starting this discussion informally now, but in the very near future I want to start an RfC explicitly asking the community whether there should be an exception carved out of WP:NOTPUNITIVE for outing blocks – in other words, whether punitive outing blocks should be expressly permitted. Personally, my view is that the answer should be "no", that outing blocks should not be different from other blocks in this regard. And I think it would be useful if a community consensus can be established that this is the case (or the opposite, if that's what happens). I'd prefer to pin that down before we have an RfC about borderline cases. --Tryptofish (talk) 23:08, 13 September 2016 (UTC)
I just received an automated note via the Wikimedia message system: "Ласкаво просимо до україномовної Вікіпедії, Nagle!" This is an automated greeting apparently sent because I merely browsed the Ukrainian Wikipedia while logged in Wikimedia-wide. I did not edit there. That's a bit creepy, and would be much creepier to someone who had connections to Ukraine. I was just chasing down a reference someone had made indicating better info on a subject on that wiki.
I agree with you that this is a problem, and I find it rather scary. I don't think that it quite fits the definition of outing that we use here, because it did not make your real-life information public. But it's a serious problem nevertheless, because (aside from gathering anonymous group statistics in order to better understand readership), the Wikimedia software should never be identifying and keeping track of who reads what, simply as a reader and not as an editor. (I've looked up some fairly oddball things here on the English WP, just out of curiosity, and I would not want to think that there's a record of that somewhere.) It seems to me that this issue should be raised with the WMF. --Tryptofish (talk) 18:04, 30 September 2016 (UTC)
It is not a problem. If your SUL account does not have an attached local account at a project that you visit, one is created and attached. This creates an entry in the user creation log at that project. The automatic welcome message is triggered by that log entry. — JJMC89 (T·C) 20:02, 30 September 2016 (UTC)
Oh, thanks for clearing that up. (Obviously, I don't read Ukranian!) If it's just a welcome message that is unconnected to the particular page(s) read, then there is indeed no problem, and my previous comment does not apply. --Tryptofish (talk) 20:10, 30 September 2016 (UTC)
That explains what happened, but does not excuse it. Creating an attached local account merely because of viewing content is not good. John Nagle (talk) 21:42, 30 September 2016 (UTC)
It may or may not be good, but it definitely is not outing or harassment, so it probably is no longer an issue for this talk page. --Tryptofish (talk) 21:55, 30 September 2016 (UTC)
This has been in effect for years. My Arabic Wikipedia page was created in 2010, and I have no edits to ar.Wikipedia (in fact I don't think I had even been to ar.wikipedia at the time). --kelapstick(bainuu) 22:01, 30 September 2016 (UTC)
I want to thank everyone who participated in this discussion. I'm making this closing statement, but if anyone objects to my doing so, please feel free to revert and get an outside closer.
There was an overwhelming consensus that this RfC was badly constructed, and asked the wrong question. Based on what responding editors said, the issue really has less to do with WP:NOTPUNITIVE, than with Wikipedia:Administrators#Expectations of adminship. Also, members of the community continue to disagree about some key issues. Therefore, the RfC should not be over-interpreted.
Overall, the community affirms that NOTPUNITIVE applies to all blocking decisions, including blocks for outing violations. However, editors also note that blocks that have some punitive aspects nonetheless can have valuable preventative effects, and there is considerable resistance to blaming administrators for possible punitive aspects of outing blocks, and sympathy for administrators resorting to blocks when the administrator truly cannot ascertain intent. The community also considers outing, when deliberate and malicious, to be a particularly odious kind of bad conduct, and supports treating it harshly.
It is very significant (especially when taking the discussion sections into consideration) that the majority of editors who supported punitive blocks under some circumstance and the majority of editors who objected to describing outing blocks as punitive nonetheless specified that not all conduct that might fall under the outing policy should be treated the same. Overall, the community rejects the idea that outing blocks should be automatic or procedural. Rather, there is an expectation that administrators and functionaries must evaluate the specific circumstances before determining whether a block should or should not be made. There is clear support for blocking when the outing is deliberate and malicious, but not when it was unintentional or where there is substantial evidence that the editor was acting in good faith. Ultimately, blocking administrators are responsible for evaluating whether or not the action was truly intentional and malicious, and also evaluating how best to avoid drawing attention to private information, before making a block, and the blocking administrator is responsible for justifying the block without simply saying that the block was automatically required. --Tryptofish (talk) 23:21, 19 October 2016 (UTC)
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Yes, purely punitive blocks for outing should be allowed.
Yes Wikipedia should block punitively and often does, claiming it's preventative. Tryptofish's thought experiment below reveals that I can get a freebie for outing someone so long as admins are convinced I won't do it again. Is that what we want? DGG's response to Tryptofish admits that en-wp blocks editors as a punishment all the time out of a feeling of necessary retribution. I, for one, welcome punitive blocks. Editors should not believe that bad behavior will be tolerated so long as it's not repeated. Chris Troutman (talk)14:12, 20 September 2016 (UTC)
These punitive blocks are against policy, and therefore bad by definition, and should be identified and individually reviewed. The responsible admins can then be desysoped if needed. Geogene (talk) 00:57, 21 September 2016 (UTC)
I don't know if "punitive" is the right word, but in general clear outing should often, but not always, result in an immediate block even if we don't think another outing case is likely. I think WP:OUTING already makes that clear ("...attempted outing is grounds for an immediate block."). Hobit (talk) 00:20, 21 September 2016 (UTC)
I'd like to note the presence of "grounds for" means its optional--not containing an obligation to block--and "immediate" rules out waiting a week after the oversight and then blocking, "because we always block for outing". In other words, I'm not sure why Hobit is !voting "yes" here when the policy they just quoted doesn't support it. The policy they cited seems to be against. Geogene (talk) 00:47, 21 September 2016 (UTC)
Humm, I'm not sure I follow you. This section is "Should be allowed" not "Should be required". I'm saying that if user A outs user B, an admin should be able to block user A without worrying if the block is preventative or punitive. There are certainly cases, spelled out in WP:OUTING, where blocking would be inappropriate (a clear mistake in addressing someone by their real name when you are friendly with them IRL for example). But on the whole, if anyone is outing someone else as an attempt to harm, that should be blockable without the punitive/preventative argument coming into play. Hobit (talk) 16:55, 21 September 2016 (UTC)
I would agree with you that an admin *should block* if the attempted outing had just happened or if the outing looks to be malicious. And, my reading of the policy you quoted supports that. The question is whether every outing must result in a block, including days after the material has been oversighted, the user that outed is communicative, the outing was perhaps accidental, and the matter appears to be cold. "We always block for outing" has been used as a justification in such instances. I think the question is whether "we always block for outing". Geogene (talk) 20:50, 21 September 2016 (UTC)
Other than clearly accidental outing I'd generally favor blocking, even if cold. But in any case, that's not what the question asks. It asks if punitive blocks for outing should be allowed. And my answer is "yes". Hobit (talk) 03:56, 22 September 2016 (UTC)
Well, in this case we are talking about exemplary blocks. That's the only reason for anything punative here. We have no interest (I hope and assume) in revenge or retribution purely for its own sake, and little interest in justice the sake of the appearance of justice being done, which are the the only other reasons (I think) for plain old punishment. Rather, a punative block is to set and show an an example for the next guy: "Jeez, if they did that to him, what might they do to me? I'd better hew to the straight and narrow". This is a reasonable justification for a punishment BTW. And if we are ever to have exemplary punishment, I would think it would be for "I dug around and found that User:SuchAndSuch is actually Ralph Spoilsport of 123 Meadow Lane in Sandusky Ohio, and BTW here's his SSN and the route his kids to take school" or whatever. Herostratus (talk) 22:20, 30 September 2016 (UTC)
I see I need to clarify my position. There is no clear distinction between punitive and preventative blocks in any case at all. All blocks are preventative, in the sense thet they prevent repetion of the same conduct. They also have the effect of serving as a "punishment" for people who think that certain actions ought to be punished. In a sense and punitive effect is also inherently preventative, because they serve the purpose of preventing similar problems by warning others. But sometimes a block is worded so the punitive effect is predominant, and this should not be done. In my opinion the real agenda here is that some people consider some of the recent blocks for outing as excessive, and in particular by being excessively punitive. I agree with this opinion in some, but not all recent cases. I'll say more about this when I run for re-election to arb com. DGG ( talk ) 18:53, 7 October 2016 (UTC)
Well of course some of the recent blocks are excessive. The question was about outing in general. It's a terminology problem: if I understand correctly, the term "outing" applies to both these situations:
"There's User:InItForTheBenjamins here, and a user with the same name on LetsCorruptWikipedia.com, and both accounts avow that they are the same person, and that the account here is for corrupting the Wikipedia on purpose for money; and the person hasn't listed personal information here, but I can't guarantee that the person, at his account on LetsCorruptWikipedia.com, hasn't, of his own free will, listed personal information -- or might in the future."
"I'm in a dispute with User:SomeName, and I employed a private detective and found out his name and address and other personal information, and here it is."
Well its silly to use the same term for both these situations. They're very different! However, a lot of people don't think the two cases are different. They think they're the same. And so they think the same term should apply to both cases. I think the first case should be called "protecting and defending the Wikipedia" and the second case should be called "outing". Different things, different terms! It makes the subject easier to discuss. So let's agree to use these terms.
So, you didn't ask if there should be punitive blocks for protecting and defending the Wikipedia (if you had, I would have said "of course not"). But for outing -- well, maybe. If it's to set an example rather than for its own sake. Herostratus (talk) 18:11, 10 October 2016 (UTC)
A block for an outing does not need to consider whether it is a punitive block. Unscintillating (talk) 16:22, 15 October 2016 (UTC), but as usual on Wikipedia, as long as enforcement is a political process, it isn't going to happen no matter that is subject to . It is clear that the VIP class thinks, and the evidence supports that thinking, that they are above the rules. The problem here is who is going to do the block? Unscintillating (talk) 16:22, 15 October 2016 (UTC)
No, blocks for outing should follow the same policies as all other blocks.
No. WP:OUTING says in part that "attempted outing is grounds for an immediate block." That makes it sound like, perhaps, such blocks should be automatic. Recently, there have been some high-profile blocks where the preventative nature of the block was not obvious. And there certainly is an argument that outing is sufficiently objectionable that there should be an approach of "block first and ask questions later". But I think that sometimes oversight blocks are made of experienced editors who would quickly agree that it was an error and would not do it again, and the block ends up needlessly creating a Streisand effect that paradoxically draws attention to what the oversighted edit said. I decided to start this RfC because I think that WP:NOTPUNITIVE (for all its conceptual limitations) ought to remain an important community norm, and I think it would be useful to determine what, in fact, the community currently thinks about this issue. --Tryptofish (talk) 22:45, 19 September 2016 (UTC)
I'd like to clarify what I meant about some recent "high-profile blocks", without actually naming anybody. There have been instances when long-time good-faith editors posted something that qualified as outing, but did so unintentionally and non-maliciously. They then redacted it when they realized what they had done, and apologized in a manner that made it very clear that there was zero likelihood of repetition. Then, a couple of days went by, quietly. And only then, they were blocked on the grounds that there always has to be a block. And a ruckus ensued, in which subsequent comments by concerned editors drew so much attention to what had happened, that it became obvious to anyone looking on what the oversighted edit had said. --Tryptofish (talk) 00:48, 20 September 2016 (UTC)
WP:OUTING currently covers this case pretty clearly.
"Unless unintentional and non-malicious (for example, where Wikipedians know each other off-site and may inadvertently post personal information, such as using the other person's real name in discussions), attempted outing is grounds for an immediate block."
I agree with you that it covers this, and you'll see that I quoted the same passage in my initial comment. But I disagree with you that it is covered "pretty clearly". Should the block always be "immediate", or are there times when determining intent justifies some time? A couple of days? When you say that it is covered, how do you reconcile the specifics of what I described, with "unintentional and non-malicious"? I described the situation as being one where "unintentional and non-malicious" was the case, and demonstrably so. That's why things are really pretty unclear, and we need to determine what the community wants. And please note that this RfC is occurring in the context of multiple editors wanting to revise the outing policy, so any existing language is potentially open to revision. --Tryptofish (talk) 17:29, 21 September 2016 (UTC)
Fair enough set of questions. But on the whole, if it isn't clearly an accident, I'd suggest blocking first and sorting it out later. Mostly as a preventative I'll admit. Hobit (talk) 17:39, 22 September 2016 (UTC)
No, for two main reasons. Blocking a contributor for outing draws so much attention to the outing that it is often impossible to put/keep the cat in the bag. Secondly, if assurances are quickly made that the offending information will not be reposted, there is no reason to ignore WP:NOTPUNITIVE. Tazerdadog (talk) 23:00, 19 September 2016 (UTC)
I'd be very surprised if anyone presumed that a second outing would be anything other than damaging. Of course some users are just disruptive, and likely to repeat the outing conduct, and of course they should be blocked, and such a block is indeed preventative. --Tryptofish (talk) 23:56, 19 September 2016 (UTC)
No. Administrators (including members of ArbCom) may not block anyone for OUTING unless they have reasonable cause to believe that another OUTING attempt is imminent. An admin may, at their discretion, assume that other outing attempts are likely if an outing attempt has just been made. However, if a number of hours have passed in the interim, and no further attempts have been made, and/or if the user seems apologetic, then the administrator may not block except under aggravating circumstances. "Because we always block for that" is not a legitimate reason when the matter is cold. Geogene (talk) 23:04, 19 September 2016 (UTC)
No, per comments above. Administrators should have the discretion to determine what the best approach to prevent further outing would be, but a default presumption of blocking is not acceptable and may draw further attention to the outing material. Regards, James (talk/contribs) 23:26, 19 September 2016 (UTC)
No As for other problems, for someone who is a first offender, the procedure is normally a warning and an explanation . (In some exceptional cases the nature of the outing is such that it can be reasonably assumed it will continue unless there is a block, and then a block is appropriate-- and necessary-- even for a first offense.) If a first offender should continue in defiance of the rules, then of course a block is necessary. If they continue through an apparent inability to understand the rule, a block is still necessary to prevent further damage. DGG ( talk ) 07:43, 20 September 2016 (UTC)
The question is poorly framed, as has been observed, but I think that is because the OP believes that current or recent enforcement of the policy has been uneven. "Outing" (a term I dislike in this context, incidentally, but that is a detail) covers a multitude of sins ranging from the minor to the banworthy. I can think of lots of circumstances in which inadvertent revelation of non-public information happens innocently, and blocking for it would be highly unwarranted (to say nothing of any potential Steisand Effect issue). The canonical example is inadvertently referring to a Wikipedian whom one knows in real life, say by seeing him or her at meet-ups, by his or her RL first name rather than his or her screen-name. This should of course be strictly avoided, but if it happens in an isolated instance, the remedy is to quietly delete or suppress the edit and remind the editor to do better, not impose a block. As another hypothetical but not far-fetched example, if John X. Jones, the well known authority on topic Z, is anonymously editing on Wikipedia as User:JXJ, another editor with expertise on Topic Z may not realize or may forget that on-wiki his colleague is JXJ, not "John" or "Mr. Jones." I don't go so far as to say that "outing" is the same as any other violation of policy, obviously, but an inflexible policy of "we always block for outing first and ask questions later—sometimes weeks later" would be unreasonable. Newyorkbrad (talk) 19:06, 20 September 2016 (UTC)
No. Otherwise, there will be more exceptions, resulting in a question on what amount might be appropriate for which fault, leading to a virtual legal system within Wikipedia, none of which any user has the authority to.--*thing goes (talk) 16:02, 26 September 2016 (UTC)
No. Let's imagine a situation with a punitive block. Say User A posts User B's phone number and address on wiki. User A realizes their mistake and asks an admin to redact it. Admin C does redact it, then gives User A a block for outing, even though User A was simply outraged at User B or another thing at the time and had cooled down within the time they got blocked. So, how would user A go about appealing their block? They can't, because they've been punished. It's not our responsibility to hand out punishments though (that can be decided legally), but to prevent more such outings.Or, let's imagine a situation where a cop shoots a suspect as punishment for resisting arrest, even though the suspect has already calmed down. That's not very fair, is it? Well, neither is punishing a user after the fact. epicgenius (talk) 18:46, 26 September 2016 (UTC)
What's key is the intention of the user who attempted outing. Sometimes it's not clear whether it occurred inadvertently or maliciously, especially if it occurs in the heat of a contentious discussion, and sometimes it is clear that it occurred inadvertently, for example by referring to another user by their username on another website because they are friends with the other user in real life. In the former case, a block may be justified so that potential harassment can be prevented. In the latter case, a block would not be justified. In all cases, judgment – not a blind rule – is needed. Mz7 (talk) 13:56, 7 October 2016 (UTC)
No Summoned by bot Firstly 3 examples of 'outing' 1) An editor notices a similarity between an editor name and article subject or content and leaves an 'editor may have close connection' tag on the article, has technically 'outed' another editor. … 2) Early in my days here, an editor used a RL first name on another editor's talk page. The first name was unusual and characteristic of a particular nationality and effectively 'unlocked' the editor name. The editor had thus, inadvertently outed the RL identity of someone with a (fairly minor) off-wiki involvement in the area they tended to edit in. … 3) An editor made a similar connection to that described in example 2 (a crypic editor name which could be 'decoded' by someone with knowledge of a particular language/names), the editor used his suspicion as part of battleground behaviour on an unconnected article (are you that same XXX-journalist who writes for YYY-magazine, linked to WP article on the journalist). These three behaviours are clearly different, the last to me warrants at least a final warning (but was actually ignored at an ANI, in which this was one of many, many examples of bad behaviour), the other two are clearly innocent, the first even admirable. Blocks, like everything else here, need judgement. Nothing about present wording, to me, suggests that blocks should be either automatic or punitive. Depending on the seriousness of the case, an admin may be entitled to 'err on the side of caution', but preventive intention on the admin's part should be demonstrable (s/he had reason to believe the behaviour would not stop, or was SO serious as to warrant a block in itself), Pincrete (talk) 15:42, 16 October 2016 (UTC)
Every block could be considered punitive. But blocks are simply a form of determining whether an editor can have the continued privilege (and it is a privilege, not a right) to edit this site. In order to edit, one implicitly agrees to follow the policies and guidelines of the community, and we assume they will. If there is sufficient reason to believe they will do it again, we block them until they can assure us otherwise. I don't see the difference between vandalism and outing in this regard; both are violations of our policies. It's just that we block thousands of vandal accounts each week and move on, while outing does also have real-world repercussions, and is more rare and downright scary, which is why it gets more press. The severity of the offense can certainly be a factor in this determination; we treat test edits differently from Grawp. --Rschen775400:15, 20 September 2016 (UTC)
What if a seemingly non-malicious outing occurred, the material was redacted, and the user apologized? Should they still be blocked a week later, if some admin comes along and says "Oops, we forgot to block them"? Geogene (talk) 00:21, 20 September 2016 (UTC)
It depends. Sometimes when you are working in the outreach realm, are on private mailing lists, or meet editors in real life, you learn information about them that is not public, and then sometimes you may accidentally use their first name onwiki. I know that quite a few functionaries and administrators would be blocked if that was blockable. But the editor who keeps making these "honest" mistakes and isn't careful? Or the editor who apologizes but doesn't seem to understand the policy? Well, WP:CIR. --Rschen775400:49, 20 September 2016 (UTC)
They would need to be blocked. That would also cover the "snake hook" blocks that Dennis Brown just mentioned, that might be applied right after the outing occurs and are needed to get things under control. The question, as I understand it, is whether *any* outing should be blocked well after the material was oversighted and the situation is under control, "because we always do that", or for similar reasons. Do you support those? Geogene (talk) 00:55, 20 September 2016 (UTC)
I feel like you are asking about a specific incident, and not knowing the specifics, I would have to decline to comment. --Rschen775401:02, 20 September 2016 (UTC)
Yes, in other words, what if there is not "sufficient reason to believe they will do it again"? Or are you saying that there is always sufficient reason? --Tryptofish (talk) 00:30, 20 September 2016 (UTC)
It really depends. Each case is different. Sometimes oversighted material isn't outing. Sometimes the oversighted material was so bad that we really cannot believe that there is any other intent than malicious. Sometimes there are other factors that have to be considered as well (and I am speaking as someone who has seen oversighted content before - I will say that even though ArbCom does the blocking and appeals, oversighters and stewards can see suppressed content and can generally guess why the block was applied). --Rschen775400:25, 22 September 2016 (UTC)
Thanks for that thoughtful answer. I agree with what you said. I think this comes down to recognizing that there needs to be an evaluation of the particular situation, and that it should not be a matter of blocking automatically without considering that each case is different. --Tryptofish (talk) 01:05, 22 September 2016 (UTC)
There are only a finite number of reasons why someone is blocked for outing, none of them punitive:
The blocking admin believes that a block is necessary to prevent further outing, now or in the future.
The blocking admin is unsure whether a block is necessary and is waiting to hear back from the blocked user.
The blocking admin is unsure whether a block is necessary and has passed it to oversighters and/or the arbitration committee to discuss.
After discussion, oversighters and/or the arbitration committee believe that a block is necessary to prevent further outing, now in the future.
It's worth pointing out that an indefinite block is not the same thing as an infinite block - just because someone doesn't understand now why outing is a Bad Thing, doesn't mean that this will always be the case for example. Thryduulf (talk) 17:23, 21 September 2016 (UTC)
@Thryduulf: While you were on ArbCom, there was an ArbCom block where the stated rationale was that "ArbCom always blocks when it happens on case pages". Which of those bullet points does that rationale fit into? --Tryptofish (talk) 17:51, 21 September 2016 (UTC)
I don't recall the incident off the top of my head, but it seems most likely that it would come under the first bullet point. Thryduulf (talk) 19:04, 21 September 2016 (UTC)
Thanks for the reply. I'm very familiar with it, and in fact the blocking Arb and several other Arbs stated that they did not really believe that there was any future risk, just that "ArbCom always blocks when it happens on case pages", and there was a massive Streisand effect. This is why I asked. Users can say that these blocks are never punitive, but saying it does not make it true. Users can say that the blocks are always made on the basis of those bullet-pointed criteria, but there is recent experience to the contrary. I think those who make the blocks can be complacent that the system is working as it should, but they may be missing things that I hope this RfC will bring out. --Tryptofish (talk) 20:00, 21 September 2016 (UTC)
Sure, that's fine, thanks. Please just keep an open mind. There can be a difference between the way things are supposed to be, and how they actually are. I'm not looking to criticize anything from the past, but I think that it is necessary to clarify how things are really supposed to be. --Tryptofish (talk) 21:52, 21 September 2016 (UTC)
I think "never punitive" has to do with the severity of the real-world/off-wiki consequences which outing brings. People can use someones personal information against them in the most consequential of ways. Outing isn't simply itself harassment, but it gives others the incentive to harass/out them as well, so what comes up can never really come down. Outing has been well known to drive away valued and well established contributors, which isn't only a loss for Wikipedia, its the fact that these are actual people whom feel they can no longer do what they once loved and have to deal with the repercussions of having participated. So unlike most other policies and guidelines that get violated and can easily be remedied by following them, the stigma of outing never goes away, because the damage can never truly repair itself. As such, outing is something that should be taken very seriously, more seriously than just about any other violation of policy, because its not only the damage to the encyclopedia we have to think about, but the damage to real people we have to think about too. If we have any hope of drawing the line on outing, it needs to be taken very seriously and should not be tolerated. —Mythdon (talk • contribs) 19:55, 22 September 2016 (UTC)
Maybe not never, but outing can be IRL harmful in a way which typical blockable offenses like writing "poop" or "you suck" or "stop or I'll sue you" on an Internet page is not. It's appropriate for WP to respond with a correspondingly larger degree of caution and less AGF, blocking preemptively and insisting on compelling evidence that the behavior will not be repeated. FourViolas (talk) 22:50, 30 September 2016 (UTC)
They are always in some sense preventative, in that they serve to warn others, but a more realistic question is whether the punitive aspect of some recent block is excessive. I think some are, but there's a much more fundamental underlying question, whether our policies on outing are excessively rigid. I think they are., and in at least 3 respects: a/ we are insufficiently flexible in considering whether the person has already outed themselves b/ we should not apply the policy in cases involving the action of clearly bad faith contributors (though how to identify them is a very tricky matter--I won't pretend I have any solution there.) c/in cases where the outing does no harm, it's hard to justif a penalty on any basis. DGG ( talk ) 19:00, 7 October 2016 (UTC)
As a long-term member of the functionaries team I suppose I know about as much about this sort of thing as anybody, and this is where I land on it. Deliberately outing another user is never acceptable behavior. People who fall so far afoul of our expected behavioral standards need to be blocked to prevent them from further damaging the project. That's not punitive action, it is appropriate defensive action. Beeblebrox (talk) 03:41, 13 October 2016 (UTC)
In general, outing should not be treated in the same way as other offenses. As mentioned below, blocks for legal threats are not considered punitive blocks, but they warrant immediate blocks nonetheless. Outing is, in my opinion, a far worse offense than any legal threat, and deserves to be handled as such. So outing blocks are not punitive, but outing is an immediately blockable offense nonetheless. Pinguinn🐧20:12, 15 October 2016 (UTC)
Discussion about outing and blocks
Outing and No legal threats are very similar. NLT is there to stop people threatening to sue someone to stop them doing something (editing an article). Outing (if not by design then effect) is usually aimed at causing a chilling effect where someone has to seriously consider if continued participation here is going to cause them any real world issues. They should continue to be treated the same way and blocks aren't punitive for NLT then they aren't be for Outing. The phrasing of the RFC however doesn't give an option to vote for no they aren't punitive. Amortias (T)(C) 23:04, 19 September 2016 (UTC)
That's an interesting point, that NLT and Outing should be treated the same way. In practice, when a user backs away from (usually striking through) a legal threat and apologizes, they don't get blocked. Admins are smart enough to know if somebody tries to game that system by coming back with more threats later. But with Outing, even accidental outing, there are some Admins that think they can wait a few days after the user has apologized and the content been redacted Oversighted, and then block somebody. Which is nonsense that this addresses. Geogene (talk) 23:36, 19 September 2016 (UTC)
I agree with Geogene very much about that. I expect that one effect of this discussion is that we will see a cognitive dissonance of sorts in which administrators say that there has never been such a thing as a punitive outing block, because such a punitive block is logically impossible, and react rather strongly to the opening of this RfC because the RfC questions that assumption – and then non-administrators will point out the context that zzuuzz asks for below. I think it will be valuable to establish, one way or the other, where the community as a whole really stands on the issue. If, as Majora says just below, that it is so important to prevent outing that we should assume that all blocks are preventative, is what the community agrees with, then let's find that out. But if the community decides, instead, that making an example of somebody is not the same thing as prevention, then I will expect administrators and functionaries to abide by the consensus of the community. --Tryptofish (talk) 23:48, 19 September 2016 (UTC)
This RfC has been worded horribly. Sorry, it just has. Outing blocks are not punitive by design. They are protective for the person being outed. That isn't going to change and therefore this entire RfC is a moot point. If you out someone, or attempt to out someone, you will be blocked. Period, no questions asked. I would hope that anyone that sees that as a problem take a step back and think about how they would feel if someone posted your personal information on here for everyone to see. Wikipedia is built on the idea that you can remain completely anonymous if you so choose to. Blocking those that try to out, or do out, someone allows that. I can't even !vote in this RfC because the question has been worded in a way as to only result in one answer, the one the OP wants, regardless of the fact that the entire question is a moot point. --Majora (talk) 23:13, 19 September 2016 (UTC)
Thanks for raising this point. The issue of the RfC phrasing had occurred to me, too. But it seems to me that you are all actually saying that outing and NLT blocks are preventative. So there is only a problem if what you describe as the intended chilling effect leads an administrator to block without considering prevention, just because it is so very, very bad. In fact, it occurs to me that a big part of the stated reason for NLT blocks is that it is very difficult for the normal editing process to continue at the same time as a legal action, which goes beyond simply the existence of intimidation, so that's an imperfect comparison to outing. I encourage editors to take a look at Wikipedia talk:Harassment/Archive 8#Flowcharts, the right-hand side. I'm fine with the addition of the third option to the RfC, but I also think that it will be interesting to see whether the community as a whole agrees with administrators about this. --Tryptofish (talk) 23:28, 19 September 2016 (UTC)
To clarify, I feel like commenting on this RfC would be like jumping into the middle of an argument I know nothing about. Why is this RfC here? What problem is it trying to solve? What policy changes, if any, are being proposed? -- zzuuzz(talk)00:12, 20 September 2016 (UTC)
It is all about how you look at it. Editing can continue just fine during legal action, but the suing editor isn't allowed to participate, per the Foundation and because being blocked prevents disruption and future chilling during the suit. If you out someone and I block you for two weeks, I'm preventing you doing it again in the future because you don't want to get blocked again. It isn't just about preventing you from doing it during those two weeks. "Preventative" is a very broad word, meaning more than the literal interpretation. Some would confuse that with punitive, but it isn't about revenge, it is about introducing a disincentive: the block. This is why not having context makes it impossible to take a stand here. Dennis Brown - 2¢00:16, 20 September 2016 (UTC)
Altering someone's behavior by way of any kind of operant conditioning is punitive by definition. No amount of handwaving will change this simple fact. You may be able to rationalize deterrence here, but before you are in a position to apply it WP:NOTPUNITIVE will have to be done away with first. Geogene (talk) 00:29, 20 September 2016 (UTC)
Dennis, I do understand what you are saying, and I recognize that WP:NOTPUNITIVE is inherently problematic, because punishment can itself be preventative. But the issue here is that a considerable number of administrators assume, as I think you just assumed here, that anytime an editor posts something that reveals personal information about someone else, that they are likely to keep on posting it. And that's not been true, in some very prominent incidents in the last year or so. So that, in turn, puts you in the position of blocking someone because, in effect, you are making an example of them. Maybe that has a preventative effect, and probably it does. But I do not accept that it really adheres to what the community expects. --Tryptofish (talk) 00:37, 20 September 2016 (UTC)
Keep in mind that often a block is done to simply control the situation. If I block you for malicious outing, I'm limiting you to your talk page so I can fully assess the situation and determine if it was accidental or intentional. In that case, it is preventative but when we don't really know what the future risk is, so we err on the safe side and investigate. Sometimes it is obviously accidental and no one is blocked, we just revdel. Most of the time when we run across outing, it was reported to us and we have to act using limited context and make an educated guess. NLT is the other area where we might act quicker than we like because of the potential damage it could cause. Other issues, we can work slower. So, I don't block because I assume they will do more, I block because it is a breach of community standards AND I have yet to assess the risk of it happening again. It is something you can be banned for, so blocking when it is potentially malicious is the safest course of action. Dennis Brown - 2¢00:46, 20 September 2016 (UTC)
I hear you on that, but it comes down to where you said the block is for malicious outing. As I understand it, the overwhelming majority of outings are committed by trolls and other bad-faith editors for whom it makes very good sense to block quickly. But there have been some recent cases of blocks where the outing was very obviously accidental, and it was obvious to anyone with a brain that the future risk was zero, and the need for a block to buy time in order to evaluate future risk was non-existent. And the block was made (in some cases days later!) on the grounds that we always block. That's not a block "because it is a breach of community standards AND I have yet to assess the risk of it happening again". It's a block "because it is a breach of community standards", period. And I want to find out in this RfC if the community does or does not want that to happen. --Tryptofish (talk) 00:57, 20 September 2016 (UTC)
Which takes us back to the OP in this thread, that the RFC lacks context. It is a very complicated issue. Like most any block, you will seldom find everyone agreeing with every OUTing block. I don't know any admin that thinks a block is required for outing. It is more likely they suspected malice, or just made a bad block. Dennis Brown - 2¢01:03, 20 September 2016 (UTC)
About context, somewhere in this already-tl;dr discussion, I provided links to prior discussions. And of course there's a catch-22 about context in outing: one cannot link to the actual outing. But please don't think this is just about some random admin making a bad block. In at least two such cases I can think of, these were ArbCom blocks, and there were explicit statements that the blocks were required. --Tryptofish (talk) 01:09, 20 September 2016 (UTC)
I'd like to offer a suggestion to editors who object to the choices in this RfC, and who are inclined to say that outing blocks are always preventative. Please ask yourself: if that is what you believe, then does that not mean that you support the premise that such blocks should be subject to NOTPUNITIVE? If you believe that this is how things are and how things should be, then surely you agree with that position? And if you do not agree, please ask yourself whether, in your heart of hearts, you know that NOTPUNITIVE is often disregarded for outing, and you object to the RfC because you do not want to admit it. --Tryptofish (talk) 01:22, 20 September 2016 (UTC)
That is only mildly insulting seeing as your statement makes it seem like you think people don't understand what they are !voting for. But whatever, I'll still answer your question. If someone attempted to, or did, post my personally identifiable information I would want them blocked. Immediately. I would want them blocked until they can assure the oversight team or arbcom, and yes it should be an {{oversightblock}}, that they are not going to do it again. NOTPUNITIVE is a red herring in this discussion. All outing, purposeful or accidental, is incredibly harmful to the person being outed. This is one of the most serious things that could happen here and can have real world repercussions. People knowing who you are in the real world can lead to stalking, harassment, and threats to yourself or your family. You just have to ask some of the admins here that have disclosed their real names to understand that this happens. All outing blocks are preventative and protective of the person that was outed. I've already stated that. If the block was "forgotten" the oversight team or arbcom still needs to ensure that the action will never happen again. By any means necessary. --Majora (talk) 01:39, 20 September 2016 (UTC)
Policy is not to out people. Period. I want to be assured that people that do are not going to do it again. Once they assure the people the community has elected to handle these matters (ArbCom and by extension the oversight team who are appointed by ArbCom) they are more than welcome to come back and edit again. If you see that as wanting my "pound of flesh" for a policy violation so be it. But don't say "policy notwithstanding" because policy says don't out people. If you can't handle that simple request you don't deserve to have the privilege of editing here. Very simple really. --Majora (talk) 02:49, 20 September 2016 (UTC)
True. But there is more than one policy here. And if there's good reason to think that a user (1) did not out someone maliciously and (2) is unlikely to out again then there is no reason to block them. Period. Also, vengeance tends to lead to injustice. If you can't go on with your own editing career without destroying everyone that ever wronged you, you should not be editing here either. Geogene (talk) 02:55, 20 September 2016 (UTC)
I feel like that reasonably describes what current policy is: if there is no reason to think that disruptive behavior will happen again, then the user should not be blocked, but when a user frequently has lapses in judgment or when such a lapse is egregiously unbecoming of any established editor, that's when we need to consider preventative remedies so that it really doesn't happen again. And Wikipedia is not a place to hold grudges. There is no reason to change any of these principles, is there? If administrators have been too hasty with blocks, there has to be better solutions than changing policy to make it acceptable, which is why this RfC strikes me more as a poignant statement than a serious proposal and should be closed per WP:SNOW. Mz7 (talk) 03:57, 20 September 2016 (UTC)
Mz7, I'm not going to agree to a snow close, because I really do want there to be a full RfC. If the result is lopsided, so be it. I'm not trying to be poignant. And I'm not proposing a change. Personally, I think NOTPUNITIVE is indeed the official status quo. But whatever the community feels about it, let's make it clear.
Majora, there is absolutely nothing wrong with you saying here that you are in favor of punitive blocks. That seems to be what you are arguing, so you might as well take that position formally in the RfC. I'm not trying to be insulting, really. I don't think that you don't understand. Whatever "insult" you perceive, and whatever reason you have for not just going ahead and answering "yes" to the RfC, is in your mind. I agree with you about how bad outing is! I'm not arguing otherwise. But I disagree with you about "whatever means necessary". Maybe my opinion is at odds with the community as a whole, or maybe your opinion is. Let's find out. --Tryptofish (talk) 22:00, 20 September 2016 (UTC)
Outing and blocks, section break
In some now-archived talk from this policy page, I suggested the following flow-chart as a sort of thought experiment:
For administrators
Step 1: Does the edit violate the outing policy?
If no → No action.
If yes → Immediately rev-del the edit and have it oversighted as soon as possible. Go to Step 2.
Step 2: Do you believe at this time that the user who made the edit is likely to continue violating the policy?
If yes → Block and revoke talk page access. Go to Step 4.
If no → Go to Step 3.
Step 3: Do you believe that the editor who made the edit will stop immediately after receiving a warning?
If yes → Issue a final warning, and block if the warning is ignored.
If no → Block and revoke talk page access. Go to Step 4.
Step 4: After the block, do other editors discuss the incident or the block on-Wiki?
If no → No further action.
If yes → Go to Step 5.
Step 5: Monitor each comment in those discussions. Do any comments reiterate information that was oversighted?
If no → No further action.
If yes → Return to Step 1.
I'm reproducing it here as something to think about, and I welcome comments. For me, this illustrates what might be "best practice" for administrators making outing blocks, but I want to see if others disagree about the steps concerning the evaluation of editor intent. --Tryptofish (talk) 01:29, 20 September 2016 (UTC)
Dennis, you referred above to malicious outing -- as distinct from the usual outing that occurs during editing disputes and discussion of COI and sockpuppettry. If it is clealry malicious a block is of course necessary ,as for any other malicious attempt to dsrupt WP. The majority of outing we have seen recently is not of thisnature. DGG ( talk ) 07:46, 20 September 2016 (UTC)
Malicious isn't always obvious at first glance. You can wrap malice in sugar coating and fool some of the people. By malicious, I meant and should have said "obviously malicious", where there is no question of intent. That is the issue with outing, the motivation isn't always obvious. I revdel'ed an outing a few days ago, didn't make a big deal of it because I knew it wasn't malicious. Much of what happens in outing around COI or socking may or may not be malicious, but it is often hard to instantly tell in the heat of debate. Dennis Brown - 2¢15:44, 20 September 2016 (UTC)
I think that most users who are clueful enough to become administrators can determine maliciousness most of the time, and rather quickly. Most of the time, it's not something that requires hours or days of research. And please note how I constructed that flow-chart. I ask if the administrator believes that the user will do certain things, not if the administrator knows. So, if the answer is that the administrator cannot be sure, then the administrator does not believe that the user will refrain from further disruption. And that points to blocking. But surely there are times when the administrator actually believes that a continuation of the outing is extremely unlikely. You just described examples of that. So in those situations, the question becomes: should the block be automatic, or should the administrator take into consideration the likelihood or unlikelihood of repetition? That's an easier question than a lot of editors are making it out to be, and there is no "right" or "wrong" answer. --Tryptofish (talk) 22:13, 20 September 2016 (UTC)
I'm with all of the people above who note the way that the wording of this RFC has poisoned the well. It basically presents two options designed to steer the discussion into a false dichotomy, and use the psychology behind the phrasing of the question. It basically works like this: "Q: Is cutting someone with a sharp blade allowable. Option 1: Yes we should be allowed to murder people with knives Option 2: Are you fucking crazy, we can't allow murder with no repercussions" And then restricting people to those two options. Making all the surgeons be like "What about us?" --Jayron3220:15, 20 September 2016 (UTC)
I'm listening sincerely to everyone who objects to the approach that I have taken. But I think that you are over-dramatizing the situation. One editor has already said that they support punitive blocks, and it's a perfectly reasonable position to take. It's not crazy. And nobody is saying that it's crazy. And if you object to punitive blocks, you are free to say so. Nobody is asking you to say that a particular, specific block that has happened is right or wrong. I think most editors do believe NOTPUNITIVE, but those who are getting upset about the question are misdirecting their criticism. I think it looks a lot less "crazy" to say forthrightly that one supports punitive blocks than it does to say that it is unfair to ask the RfC question. --Tryptofish (talk) 22:00, 20 September 2016 (UTC)
Also, for anyone who may be looking for some kind of ulterior motive on my part, all you have to do is to look at #About blocks to see my motivations for creating this RfC. No deep dark secrets. --Tryptofish (talk) 23:21, 20 September 2016 (UTC)
I don't think your crazy. I think a vague RFC leads to meandering discussion like this, but frankly, debating you is always interesting. Disagreeing is an opportunity to learn something new. I still have no idea where this will go, as far as this RFC is concerned. Dennis Brown - 2¢00:08, 21 September 2016 (UTC)
I can't find any huge issues with Tryptofish's flowchart above. Steps 2 and 3 could probably be combined; if the user isn't blocked for attempted outing, there would almost always be some kind of warning. The relevant question as far as NOTPUNITIVE goes is whether there is an imminent or ongoing risk of continuing disruption. I'm reading the outing policy as it currently stands and the relevant sentence is Unless unintentional and non-malicious (for example, where Wikipedians know each other off-site and may inadvertently post personal information, such as using the other person's real name in discussions), attempted outing is grounds for an immediate block. (emphasis mine) So the policy right now is to first evaluate the editor's intention when deciding to block, as the flowchart shows. I can envision cases where an outing occurs during the heat of an intense, acrimonious discussion. In such a case, it's not clear to administrators whether it's a one-time thing, so I can see how an "immediate" (as in, without discussion with the user) block is appropriate (i.e. "yes" to step 2 or "no" to step 3). Mz7 (talk) 03:00, 21 September 2016 (UTC)
That's one of several perfectly reasonable interpretations of the existing policy. But the problem is that it's not the only possible interpretation. Please note that Hobit has cited the same wording, but with, in effect, the emphasis placed instead on the part about "grounds for an immediate block". So how do we reconcile what you call "evaluat[ing] the editor's intention" with the immediacy of "immediate block", and with what Dennis Brown has described as the need that administrators sometimes have, for some time to actually figure out the intent? I've seen blocks where the blocking administrator (actually ArbCom) took a couple of days to make the evaluation, concluded firmly that the intent was non-malicious, and concluded firmly that there was no risk of repetition – and then blocked anyway, on the grounds that blocking is always required. I think that you would have a hard time reconciling that with what you just said. So there are significant problems with assuming that the existing policy is applied in a consistent way, or even in a reasonable way. That's a big part of why this RfC is important. Let's determine what the present-day consensus of the community really is. And let's expect administrators and functionaries to respect consensus, whatever that consensus may be. --Tryptofish (talk) 17:21, 21 September 2016 (UTC)
Procedural blocks
In my opinion, as Dennis notes, malicious [versus good-faith, accidental, constructive, etc.] isn't always obvious at first glance, especially in, for example, highly contentious longterm disputes where people are doing opposition research against each other. I think in those situations (absent mitigating context), a procedural (automatic) block is certainly within the realm of appropriateness. As others have said, this RfC lacks clear context, and outing is such a high offense that sometimes an automatic block needs to be the response, for the sake of the community and the privacy of its members. As Majora notes, blocks for outing are protective of the person being outed. Other actions on Wikipedia (edit-warring, disruption, vandalism, etc.), have no actual repercussions to other users, but outing certainly does and therefore blocks for outing are to protect the person in question (and serve as a deterrent to others). The algorithm above can be gamed, like any other guideline. The algorithm does not take into account the motive for the outing (e.g. whether it was deliberate opposition research against a longterm opponent, or whether it was COI research in the case of a disruptive new SPA spammer). Each instance of outing or near-outing must be judged on its own merits, not rigidly via an algorithm. Softlavender (talk) 05:09, 21 September 2016 (UTC)
To add, this is often compounded by the outing happening after 100k or more of discussion that no admin has seen before, then an admin walks in and has to digest a huge amount of data fast, while the discussion is getting more and more heated, and they are arguing whether or not it was intentional. Then someone claims "You already outted yourself last year at ANI", and you have to go dumpster diving for diffs, which you usually don't find, so eventually you have to make the call. There is no way we can always get it "right" in these cases, we can only act in good faith, knowing someone is going to complain no matter what you do. And yes, we will often default to the action that gives more protection to the outed person, which I think is the safest course of action. Admin'ing is easier if you don't have the actual bit, after all. Dennis Brown - 2¢10:34, 21 September 2016 (UTC)
I think that the word "procedural" is a very useful one, so thank you for adding it to this discussion. But there is a legitimate question as to where to draw the line between procedural and punitive. I don't think that anyone (certainly not me) argues that when an admin is in a situation where it simply is not possible to determine that the editor acted in good faith, and consequently the administrator cannot be confident that the outing will not continue, there is any reason not to block preemptively. Yes, block when that happens – and it's preventative. But if the administrator judges: "OK, I see what's going on here. The outing was unintentional and non-malicious. I know what's going on well enough to be confident that all I have to do is issue a clear warning to the editor who did it, and I'm sure that they'll obey me." Should the administrator do exactly that, or should there be a block on "procedural" grounds – with all the predictable Streisand effects that most definitely do not protect the outed editor? I agree: do not do it "rigidly via an algorithm". But that is not consistently the current practice. --Tryptofish (talk) 17:45, 21 September 2016 (UTC)
There is no basis for a block if it was unintentional. That already stands in policy, primarily under WP:COMMONSENSE. I think trying to codify common sense is riddled with problems, namely that if any admin lacks common sense, they shouldn't be an admin, making a new policy statement unnecessary. Even SPI blocks aren't "procedural" after they have been confirmed as socks. Each is done on its own merits. Like I said earlier, someone did a minor outing on my talk page this week, or what I believed was one and it was obvious it was innocent. I just revdeled and informed them they need to be careful. Wasn't a big deal at all and no one was blocked. Had I blocked someone for that, I would have expected to get dragged to WP:AN. No specific policy guides me on this, except WP:COMMONSENSE. Each case is different. The solution isn't a new policy, however, it is the liberal use of WP:ADMINACCT at WP:AN. Then consensus does it's work. I'm curious who this admin is (along with diffs) that thinks we MUST block if there is any form of outing. Dennis Brown - 2¢22:08, 21 September 2016 (UTC)
I agree with everything you said there (and thank-you for this discussion). I don't want to codify anything via this RfC, but I do want to have a clear assessment of what the community thinks today about punitive/preventative. And I'm inclined to think that most of the users who have reacted negatively to this RfC are not doing so because they believe that blocks should be punitive, but rather because they agree with the principle that they should be preventative and they just are not aware of the inconsistency of practice. Partly for privacy reasons, and partly because I don't want to reopen past wounds, I'm not going to supply diffs, but it wasn't a random admin; it was an ArbCom block in late 2015, and several Arbs said exactly that they always block. And there ended up being a very big Streisand effect. --Tryptofish (talk) 22:38, 21 September 2016 (UTC)
Regardless of the phrasing used, I am always very uncomfortable with the application of blocks according to an algorithm (procedure, "automatic", "brightline" whatever) rather than based on the individual assessment of every situation. Admins should always exercise due judgement on every admin action; there should not be a situation where blocks are automatic or handed out according to a plan. "Action X = Block Y always" is never good practice. --Jayron3223:29, 21 September 2016 (UTC)
Just seconding this. I think there are times when a block should be nearly automatic (clearly trying to disrupt) and other times would it should be nearly an obvious non-action (a self-outing for example--I've done this twice in one day and had to get a revdel done). But for the most part, it should be a judgement call. Hobit (talk) 17:44, 22 September 2016 (UTC)
That's an interesting distinction. Above, you argued that punitive blocks should be allowed, but not required. And here, you draw attention to the need for judgment. Thus, I think it might be reasonable to conclude that, while you accept the legitimacy of punitive blocks for users who have acted with malice, you are not endorsing automatic or purely procedural punishment. Is that accurate? --Tryptofish (talk) 23:01, 22 September 2016 (UTC)
Of course. I think that's the obvious thing and the framing of this RfC is such that anyone who believes what I do should be supporting punitive blocks as an option. Which is what I'm doing. I would say that the blocks shouldn't be "procedural" but they probably should be the default--if it looks like it might have been malicious it probably was.r Hobit (talk) 19:40, 23 September 2016 (UTC)
Who gives a damn? When OUTING is accepted and unpunished so long as it's done by an admin, or its done by a traceable Wikipedia editor, admin, or supported by the cabal of admins using an off-Wikipedia site (and yes, the obvious candidate site is one example), then what's the point in pretending that OUTING is policed at all effectively when the little people do it? Andy Dingley (talk) 22:34, 15 October 2016 (UTC)
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Planning the next RfC
I'm starting this talk section in order to pick up from where Wikipedia talk:Harassment/Archive 13#Structure of the follow-up RfC left off. (Pinging Jayen466 in that regard.) We were planning to ask the community to consider a variety of borderline situations pertaining to outing, and to ask which situations should be considered blockable violations, and which should be considered permissible.
It would be a good idea to start formulating the list of scenarios now. But first, I'd like to suggest an approach to the format of the RfC, based upon what editors recently said in #RfC about outing and blocks. It seems to me that we should ask responding editors to assign each scenario to one of three categorizations:
A. Permitted behavior, not objectionable.
B. Sub-optimal behavior that should be discouraged but that does not rise to the level of a blockable offense.
C. A potentially blockable violation of WP:OUTING.
I think that this menu of three choices, and specifically with choice B, reflects what many editors said about there being situations where we expect administrators and functionaries to exert judgment on a case-by-case basis.
We elect admins because we trust their judgement. Forcing them to follow a narrow set of rules when it comes to the topic is counter to that trust. If we trust them with the block button, we should trust them to know when to use it. --Majora (talk) 21:54, 29 October 2016 (UTC)
In a way, I agree with you. I don't see this as leading to what you called "a narrow set of rules". But I do see it as leading to changing, one way or the other, the parts of the outing section that are currently tagged "under discussion". We probably will end up either saying that certain types of COI investigations are OK, or that they are not exempted from this policy, and it is very clear that, at this time, the community has not yet come to an identifiable consensus about that. Administrators are indeed trusted to have judgment, but they are not assigned the role of deciding for the community what community norms are in effect. They are expected to apply policy in accordance with community consensus. It's a bad idea to just leave stuff labeled "under discussion", without resolving that discussion one way or the other. --Tryptofish (talk) 22:01, 29 October 2016 (UTC)
You stated that It would be a good idea to start formulating the list of scenarios. It will be literally impossible to come up with scenarios that cover every situation. Creating them and then deciding that this is ok and that is not and everything else doesn't have community consensus is the definition of a narrow set of rules. If we don't trust admins with the block button to decide for themselves what is blockable and what isn't then they should be admins. --Majora (talk) 22:08, 29 October 2016 (UTC)
In no way do I imagine covering every situation! And I do not expect clean up-or-down results for every scenario offered. I suspect that we will end up with some things where there is consensus that they are OK, and some where there is consensus that they should be forbidden, and yet others where we will end up with no consensus. The question, though, is where that boundary region of "no consensus" is located. It seems pretty likely that there are going to be scenarios where some editors will say "this should absolutely be encouraged" and other editors will say "this should absolutely be forbidden", about the same thing. This isn't going to constrain administrators, but it will provide them with guidance about the current views of the community. --Tryptofish (talk) 22:16, 29 October 2016 (UTC)
I hope whatever emerges out of this will be permissive with regard to:
Okay Noting self-disclosures in previous on-WP edits: "Welcome to my user page! I'm SnorkelWeasel. You may recognize me from [name a bunch of forums here also edited as SnorkelWeasel], where I've been active about [topic they are PoV-pushing about at Wikipedia] for years." This very scenario came up recently and was going to be used (or possibly disallowed to be used) as evidence at ArbCom, had not the user in question gotten themselves indeffed before I could open the case.
Okay: Noting fully public organizational connections: User:JMPeake shows up and starts making clearly aggrandizing, promotional edits to the "XYZ123 Inc." article, and a quick look at http://www.xyz123inc.com/aboutus.html shows that the company's public relations manager is "Jennifer M. Peake", this is enough to act on, and is not any kind of privacy violation. A scenario like this also came up recently, in the case of a founder of a new religious movement editing their own article here, using as username the epithet their followers call them off-Wikipedia.
Okay: Noting undeniably obvious cross-site identity: If some ranter called User:WrathOfJehovaInTexas appears and is making anti-muslim edits that are exactly in line with anti-muslim rants posted in various forums by "wrathofjehovaintx", "Wrath_of_Jehova_in_TX", "WrathOfJehova-Texas", and other near-identical usernames, that should also be actionable and not considered outing, especially if it comes up because other users here have encountered this person before, not because they went searching the 'Net for days or weeks trying to find a similar character.
All of these are clearly distinguishable from
Bad: Seeing a username like User:JMB-NZ, noting they edit about southern-hemisphere reptiles a lot, trawling .nz academic institutions, turning up a herpetologist named James Mark Barton at a university there, and trying to draw a connection between the two.
Bad: Seeing an ID like User:David_Brown posting about Linux, and remembering a David Brown from an Ubuntu mailing list last year and claiming they're the same person.
Bad: Just claiming that so-and-so WP editor sounds a lot like such-and-such forum ranter or blogger because of their wording, without any other public evidence of connection.
Bad: Happening to know that user User:SnorkelWeasel's real name is Jimmy-Bob O'Flaherty (e.g. from an e-mail sig) and telling everyone, e.g. because O'Flaherty works for some company and has edited their article in a way you don't like.
As WP becomes more and more a target of organized, long-game, crafty WP:CIVILPOV by people who really are not here to write an encyclopedia but to inject sub rosa advertising or outright propaganda, in response the rise of WP's influence and reach and thus the real-world value of managing to controlling certain bis of [alleged] information it presents, then WP must have the ability to adequately defend itself from these manipulation efforts. Our ability to do this without our hands too tightly tied is also entirely necessary to investigate and stop meat-puppetry; identifying a specialist-forum posting asking people to come vote in an RfC on WP and tying it to a specific WP user with a similar username was necessary in 2014, and raised no eyebrows (it even resulted in a confession by and a wrist-slap to the meat-puppeteer). This is a compressed version of what I was trying to get across last time. I'm mostly taking a break, so I may miss the next round. I ask that these issues be given some consideration when drafting the next version of the proposal. — SMcCandlish ☺☏¢ ≽ʌⱷ҅ᴥⱷʌ≼ 07:59, 28 November 2016 (UTC)
Thanks, those examples are very helpful and are very much the kind of things I've been intending to ask editors to examine. In fact, I expect that there will be some disagreements about OK versus Bad on a few of those examples. I've been slow/distracted about getting on with this, but I do intend to get around to it. --Tryptofish (talk) 22:12, 28 November 2016 (UTC)
Very good examples. I've encountered them all. If I'd add anything, it would be that "clearly aggrandizing, promotional edits" could be supplemented with spamming and the inappropriate use of the company's press as sources. --Ronz (talk) 17:22, 29 November 2016 (UTC)
Note. It occurs to me that this plan is becoming supplanted by the demands of recent events (the WMF statement and its aftermath). The ideas so far do remain useful, and should be drawn upon in crafting the community RfC, but that RfC is now going to have to be something bigger in scope. --Tryptofish (talk) 01:31, 27 January 2017 (UTC)
Your proposal seems to be about a message at registration about the requirement to disclose edits made for pay. That is not the same as a message about COI. This is absolutely critical to get right, because if this is implemented I give it about five minutes before a new user clicks "no", goes on to edit about his workplace/political party/garage band/local museum/favorite coffee shop/stud dog/etc., and gets blocked for "deception" because they denied having a COI. Opabinia regalis (talk) 19:18, 10 November 2016 (UTC)
You are right: I wrote this message at WT:Harassment rather carelessly, but it's just a heads-up. I'm pretty sure that what I composed at meta was carefully worded in terms of undisclosed paid editing, and not COI, because I based it on the terms of use. --Tryptofish (talk) 19:46, 10 November 2016 (UTC)
The proposal has been getting very strong support, from those editors who have voted on it. However, there are other proposals that are getting significantly larger total numbers of votes. The way it works is that the developers are going to focus on the ten proposals that get the largest absolute numbers of support votes. I'm posting this because it seems to me that a lot of editors who have been watching here would like to see this proposal get implemented. And at the current rate, it is unlikely to happen, unless a larger number of editors who have not yet voted there do decide to go there and vote support. --Tryptofish (talk) 01:14, 6 December 2016 (UTC)
I was just on ANI and noticed a broken link to an RFC in June of this year, and searched the archives to find it. It was on page 9 out of 13! There is one archive page covering 2005-2011, one covering 2011-2015, and then eleven more archive pages devoted just to the past 21 months. Most of the discussion seems to be focused on a very minute discussion of the wording of WP:OUT, but am I reading this correctly?
What's the big deal? Admins block accounts for unambiguous and clearly malicious violations, and warnings are issued for accidental and/or apparently good-faith/borderline violations. In cases where off-wiki discussion of Wikipedia (such as canvassing) has occurred, posting links to it is usually taken as acceptable but it is played by ear.
The only grey areas I see that might need to be dealt with in the wording of the policy page are (1) when User A is clearly engaged in disruptive behaviour on- and off-wiki, User B posts links to their off-wiki disruption, User A subsequently gets indeffed or site-banned, and User B's posting material that may have qualified as "outing" gets overlooked, and (2) when User C is very carefully staying within the letter of the law while going as far as they can to OUT User D without technically getting in trouble.
In case (2) I imagine most admins would be happy to issue a block and most other admins would not accept a lawyer-ish unblock request, but isn't engaging in reams upon reams of discussion on this page just giving ammunition to such wikilawyers?
@Tryptofish: Kindly keep your snide remarks to yourself. This talk page has a highly unusual history. It has been the primary talk page for one of our oft-cited conduct policies for more than ten years, and yet the net amount of discussion that has taken place has multiplied six-fold in the space of only two years. That's roughly a 3000% increase in the rate of activity on this talk page over the past two years.
As someone who has suffered off-wiki harassment because of my Wikipedia activity (read: creation of a bogus social media account under my real name, mentioning my Wikipedia username in the profile, and "liking" several hundred adult-entertainment actors and actresses) and countless on-wiki outing attempts (not going to give details) -- or, heck, even just as a Wikipedia editor -- I have a right to know if someone or a lot of people are trying to make it either easier or harder to get away with such behaviour. I asked a reasonable question, and I expected a reasonable answer. I should not have to read through the 1,500,000+ bytes of discussion to that has been taking place here since 2015 as a prerequisite to asking why there has been such a massive explosion in activity here. There are already users on ANI posting material that is clearly in violation of OUT and citing RFCs on this page as justification, so asking what is up is clearly justified.
Either you are just as much in the dark as me (in which case you should have agreed with me that something is weird or kept your mouth shut and mound your own business rather than insulting me for asking the same thing you should have asked) or -- more likely -- you have some awareness of what is going on (in which case by answering me the way you did rather than with a polite attempt to respond to my query you were being deliberately and unnecessarily antagonistic).
I'm not here to be impolite to other users, but I could say the same to you. No, I'm not in the dark, having put a lot of effort into the discussions that you disparage. I'm very sensitive to your issue of having been the victim of harassment, but the solution is not to get angry at editors who are trying to make the policy page better, nor to assume without reading the discussions that they are somehow going to make things worse. --Tryptofish (talk) 00:00, 3 December 2016 (UTC)
(Just noticed this. I stopped following this thread after I got the responses below.) I did not "disparage" anything, and your initial response was very dismissive and didn't make the slightest attempt to answer my questions. Furthermore, as other editors had already answered my question, there was absolutely no need for you to post the above response except to paint my comment as something it wasn't. Hijiri 88 (聖やや) 12:43, 24 January 2017 (UTC)
Attempting a good-faith summary of what I saw discussed here ... apologies if I leave out any essentials. There has been at least one long-term editor indeffed recently for WP:OUTING, which led to a lengthy discussion around the tensions between privacy and neutrality on Wikipedia. Privacy favors not discussing any real-world information on editors, whereas neutrality (some say) requires discussion of off-wiki activities, specifically, advertising for Wikipedia writing jobs on Fiverr, Upwork, and the like. Questions remain on how or when to "legally" post even a link to a single off-wiki job site without falling afoul of policy and getting oneself indeffed. - Brianhe (talk) 23:53, 1 December 2016 (UTC)
Well, I can understand the concern there, but it still doesn't explain the sheer mass of the discussion, and I don't know how recent was "recently" since it appears to have started in early 2015. Anyway, I don't see how altering the policy would affect things. I have seen very blatant outing attempts go unchecked because the user they outed was indeffed shortly thereafter, and I have seen users get blocked for breaches that honestly were relatively minor. I even saw someone get community-banned for actually contacting someone's employer roughly a month after they were blocked for threatening to do so which was in turn about a month after the actual outing occurred. It seems like the case-by-case status quo works pretty well: a minor breach that was made while attempting to improve the encyclopedia and protect it from undeclared COI editors can in theory lead to a block, but if the user appeals the block in a contrite manner or even in a manner that justifies what they did they will likely be unblocked. If all of this hubbub were because of ome long-term contributor briefly slipping over the line then the efforts of those involved would be better directed to requesting the block be overturned. Keeping the policy's wording as broad as possible to allow admins to use their discretion to block first and ask questions later seems like the best course of action, as no admin who blocks indiscriminately and blatantly/unapologetically flouts community consensus will stay an admin for long. Enshrining wiggle-room for people who want to engage in borderline outing as part of the daily workings of the project in the policy is not a good idea. Hijiri 88 (聖やや) 00:09, 2 December 2016 (UTC)
(edit conflict)Could I recommend reading GorillaWarfare's self-described "wall-o-text" which I think provides a good starting point for grasping the issues that were discussed. The "dedicated place for these [ToU violation] concerns to be privately reported" never materialized AFAIK. And I personally don't feel like we have a clear cut policy on when posting any external links is permitted (I think Doc James would agree with me on this, as he brought up many examples in medicine when this is non-hypothetical). - Brianhe (talk) 00:15, 2 December 2016 (UTC)
While posting to ANRFC, I saw that there was an RfC here that was 167 days old, and looking into it I discovered that the thread was archived nearly two months ago. I have closed the RfC as stale. Scolaire (talk) 18:39, 14 December 2016 (UTC)
Basis of harassment
"Harassment of an editor on the basis of race, sex, gender..."
An admin inadvertently outed an editor by connecting different accounts together. After the admin was informed of this the admin ignored my comment. What can be done at this point? QuackGuru (talk) 22:47, 9 February 2017 (UTC)
It very much depends on the specifics, and obviously you should not provide those specifics here, but I would recommend privately contacting WP:Oversight, using the email link there. --Tryptofish (talk) 22:53, 9 February 2017 (UTC)
User's (Redacted) and (Redacted) were blocked as a result of an SPI [(Redacted) here] due to both editing the article MDMA in the midst of an edit war. Suffice to say, using two wikipedia accounts on the same article which is in the midst of a contentious edit war - well you do that at your own risk. And frankly the idea that linking two wikipedia anonymous usernames where the editor is themselves editing (on multiple articles) with both, is 'outing' is so beyond ridiculous that the concept of 'personal information' appears to have completely missed you by QuackGuru. Only in death does duty end (talk) 23:14, 9 February 2017 (UTC)
As it happens, I did not know the specifics when I gave my response above, but I just found through another editor's talk page that what Only in death says is correct. At this point, I can see that the oversighters are already very much aware of it, to the point that my advice to QuackGuru is to just drop it for the time being. --Tryptofish (talk) 23:16, 9 February 2017 (UTC)
I've commented on this at the user's talk page, but I carefully checked those edits at MDMA, and none of them were part of the edit war, and none of them appear to in any way have been collaborative between the two accounts. I'm increasingly skeptical that this was deceptive use of the two accounts. --Tryptofish (talk) 03:15, 10 February 2017 (UTC)
I'm not seeing it either. Perhaps there was a bad-faith violation of the socking policy, or perhaps there was a good-faith error that sort of resembles editing while logged out (my gut feeling is the latter, but I could be wrong). That needs to be sorted out privately, not here. --Tryptofish (talk) 23:30, 9 February 2017 (UTC)
The block itself, as you say, pertains to VALIDALT, but editors here are questioning whether the effects of the block involve outing. --Tryptofish (talk) 03:12, 10 February 2017 (UTC)
I appreciate that, but it remains the case that this issue does not involve the revealing of "personal information" in the terms of this policy. It may be a breach of the right to operate valid alternative accounts, but that right is defined in VALIDALT and not HARASS. -- Euryalus (talk) 20:27, 10 February 2017 (UTC)
Okay so I have some concerns. I know a number of medical editors have alternative accounts. One they use to work on issues that can connect them and another that is more anonymous.
We have rules that allow legitimate alternative accounts and of course people occasionally edit with the incorrect account just as they sometimes edit by mistake well logged out. We have Wikipedia:Oversight#Policy which says "includes hiding the IP data of editors who accidentally logged out and thus inadvertently revealed their own IP addresses"
IMO something similar should apply in this case. One should have evidence of bad faith editing before connecting accounts which if one looks closely is does not appear to be present in this case. Doc James (talk · contribs · email) 02:00, 10 February 2017 (UTC)
The checkuser block should be reversed unless there is evidence of inappropriate editing. Reviewing the history at MDMA shows that one of the two accounts is normally used to edit that article. After a 21-hour break, the wrong account was used to make two innocuous edits. Presumably the person noticed their error and switched to using the normal account 14 minutes later. That was unwise but is a very minor breach which does not warrant an indefinite block. If the person had known they should have declared their alternate account to Arbcom, there would be no problem given that the use of two accounts on the same article is an obvious blunder rather than an attempt to subvert procedures. Johnuniq (talk) 05:18, 10 February 2017 (UTC)
After an independent review I have unblocked both of the accounts that were referenced above. Any discussion of changes to procedures should take place outside the context of that, or any, specific case. Newyorkbrad (talk) 17:25, 10 February 2017 (UTC)
(Redacted)
Proposal" "If the Checkuser is informed there is no evidence with using an alternative account for illegitimate purposes and the Checkuser continues to ignore there is no evidence of sockpuppetry then the Checkuser must be blocked by Arbcom." There is no reason it should be deleted. Admins should not be above the community. QuackGuru (talk) 13:51, 10 February 2017 (UTC)
That's a horribly written addition. Informed by who? Ignore how (not all evidence in these cases can be made public)? Who decides there's no evidence? And the notion that Arbcom "must" do anything is completely presumptuous. --NeilNtalk to me13:57, 10 February 2017 (UTC)
The removal of that addition was entirely correct. There is no consensus, that I know of, anywhere on Wikipedia that says anyone must be blocked for doing something. To make blocking mandatory would require a full rfc. And in this particular instance who decides there is no evidence? If you think there has been a misuse of the check user tool then raise a complaint with arbcom or meta:Ombudsman commission. Making changes to blocking policy without consensus is not the way to go. - GBfan14:58, 10 February 2017 (UTC)
