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.
Arbitrators, oversighters, bureaucrats do not have the right to *make policy* more than any other Wikipedian. Nor do they have the equivalent right to *veto policy* e.g. by saying they won't enforce a policy. That is what some arbs, etc. seem to be saying here - that they will not enforce the WP:Paid editing disclosure policy, and even that they might punish folks who do attempt to enforce it by providing the needed information on who is a paid editor.
Do please let us know if this is incorrect. If you are willing to enforce our policy and the ToU, please let us know how we are to report violations.
I'll ask that folks who have taken on the responsibility to enforce Wikipedia policy answer in the top section, and regular Wikipedians make any comments in the 2nd section. Smallbones(smalltalk)16:03, 5 July 2016 (UTC)
"Posting of personal information which has not been previously disclosed by the editor in question has never been an acceptable practice"
"Linking one's personal identity to a Wikipedia account is considered harassment. (Unless he or she has disclosed the link on Wikipedia.)"
"It is not acceptable to link undisclosed, private information to an editor's account."
"There is no reason I have ever seen to link to an external link to try to show who one editor thinks another editor is."
"It's harassment. Allowing a little bit of harassment, when you think the person really really deserves it, is not appropriate. "If we suspect you might have violated the terms of use, our volunteers may harass you"? No"
"COI editing is unwelcome, but can be addressed through other than via amateur internet sleuthing and outing attempts. "
"No As others have said, "case by case" is pretty useless and invites linking."
"We have ways to deal with this already through how many diffent policies. I am obsolutely opposed to other Wikipedians investigating other Wikipedians real lives."
to
"linking to material posted by paid-editors on elance is not allowed, even if the posting was and is fully available to the public and doesn't reveal any real world names."
But if you aren't going to take the time to actually read what I and others have said over the last day or so, than I (we) don't owe you (and I mean you specifically) any explanation at all, as the presumption should be it will be ignored. --kelapstick(bainuu) 01:41, 6 July 2016 (UTC)
Yeah, I'm puzzled here too. Personally, I don't think I've even talked about paid editing here, and as far as I'm concerned it doesn't really matter. But to take that ball and walk with it a bit, yes paid editors should disclose, but no, you can't make a suspected paid editor disclose by putting some obvious link on their talk page or anywhere else. I think I said that already. Again, are we doing the whole "some Arb thinks we can't do this, so they're saying we can't do anything"? Drmies (talk) 02:08, 6 July 2016 (UTC)
Oversighter, not an arb. Full disclosure: I blocked Jytdog last November for a remarkably similar but unrelated incident; I was actually rather astonished that he did it again, given that I only unblocked him after receiving assurances that he understood the problem.@Smallbones: Can you give an example of a case where the only possible way to prove that somebody was violating Wikipedia policy would involve outing them? Why would you not be able to demonstrate that their conduct on Wikipedia (for example writing a spam article) was disruptive and ask an admin to take action? If their conduct on Wikipedia is not problematic, why is any action necessary? Note that by "necessary" here I mean something like "literally no other way to prevent harm to the encyclopaedia".
I have to say, I've been an admin on this site for over six years and I've made many thousands of blocks and none of those relied on information about the blocked editor gained from somewhere else on the Internet or on knowing their 'real life' identity. If somebody's writing spammy articles and they continue after you ask them nicely not to do that, by all means get in touch with me; you'd make my day—I enjoy swinging the banhammer at spammers. It's what admins are for. HJ Mitchell | Penny for your thoughts? 12:38, 6 July 2016 (UTC)
@HJ Mitchell: But the two incidents with Jytdog were not identical, even if similar. And that goes to how editors can genuinely misunderstand what is, and what is not, permitted. I saw the earlier incident before it was oversighted. It was in an angry post about another editor in a dispute. On the other hand, the present incident, insofar as I can tell, was not a post directed at another user during an argument, but rather, the presenting of what was intended as evidence in a COI case. I can see how someone, in good faith, could think something like OK, I will be careful never to attack someone else in that fashion again, without seeing evidence in what seemed like a routine COIN case as being the same thing. Especially because we have had an ambiguous situation about COIN and outing for as long as COIN has existed. It's not particularly unusual for editors to refer to identifying information in COIN cases, and it usually does not result in sanctions. (In fact, a talk section above shows an AfD where a responding editor pointed out similarities between the username of the editor who created the page, and the name of a person associated with the page subject.) This kind of stuff happens all the time, with inconsistent results. Editors can be very confused about it, and it does not necessarily mean bad faith. The reason why I am commenting about this is that we need to make this policy clearer for the borderline cases (and don't anyone tell me that there is no such thing as borderline!), and also have a clearer community consensus about when blocks should and should not occur. --Tryptofish (talk) 22:41, 6 July 2016 (UTC)
GorillaWarfare
Incoming wall-o'-text, you may want a cup of coffee. Split a new section out because this ended up much longer than I thought it would.
I would echo Euryalus' request for you (Smallbones) to be more specific about where you've seen functionaries suggesting they would refuse to enforce enwiki policy. The portion of policy that would hypothetically allow offsite links as it stands right now is completely unenforceable: it consists of one sentence ("Posting links to other accounts on other websites is allowable on a case-by-case basis.") which does nothing to explain under what circumstances this is allowed, and which follows quite a list of ways this is disallowed. I think what you are seeing is a large portion of the functionaries (including myself) who would rather this sentence be removed, rather than allowing it to remain and potentially mislead people into believing outing is acceptable.
If the community agrees that this kind of thing should be allowed, and actually provides some guidance more than "on a case-by-case" basis to inform both people attempting to follow the policy and those attempting to enforce it, I will abide by it in both roles.
This line should be removed from the policy, and it should be clarified that publicly linking an editor to profiles on other websites that they have not disclosed themselves is never acceptable.
The need to actually make this link in order to preserve the quality of the encyclopedia is overblown. In nearly all of the cases that I have seen where a user has been outed to try to "catch them in the act" of paid editing (including this most recent one), they could have been easily been sanctioned under existing and widely-accepted policies. We routinely block people who are clearly editing on behalf of a person or organization, people who are inserting spam links, and even those who are sneakily inserting biased information into editing. It is rare that the only way a person can be exposed as a paid editor is by linking them to a profile on another site.
My recommendations would be the following: Remove the sentence about allowing links on a case-by-case basis. Replace it with an alternative, private method of reporting these kinds of situations, with the emphasis that the need to do this should be infrequent, and only as a last resort. This method can be decided on by the community: either the functionaries team or the Arbitration Committee would make most sense. I have in the past been opposed to the ArbCom taking on the responsibility of dealing with reports of paid editing, but I would be willing to revise that position as it is far preferable to these reports being made in public. GorillaWarfare(talk)02:36, 6 July 2016 (UTC)
However, any admin can place a block like that (I think? or just oversighters?) Only oversighters can place "oversight blocks" such as this, although any administrator can mark a block as only appealable to the Arbitration Committee when private information is involved (edit: see below my comment about a conflict in the blocking policy surrounding this). I realize the difference in this is in practice minimal, although I will note that Jytdog had the choice to appeal his block either to the Arbitration Committee or to the oversight team (of which members of the Arbitration Committee are a subset).
Your comments about Elance made me smile a bit just because they hit close to home. For about six months I had to keep a note on my userpage because someone decided to impersonate me on sites including Elance and Facebook. They created profiles, added a photograph of me, claimed to be User:GorillaWarfare, and sent out offers to companies whose Wikipedia articles had recently been deleted, offering to recreate them for a fee. Fortunately your suggestions regarding this were not in place then, or I'd probably have enjoyed a nice long block.
As I said in response to your inquiry on my talk page, I blocked Jytdog for precisely the reasons that are on his block log and talk page, not because of some long-standing grudge or nefarious oversighter plot. I reported making the block to the oversight team soon afterwards, and many oversighters weighed in to endorse it. I see Mike V has said as much below. Regarding your comments about "discussing whether he should have had more of a chance," please note what many others seem to be forgetting: oversight blocks are only appealable to the oversight team or the Arbitration Committee, which has until now seemed to be a long-accepted and uncontroversial policy.
In response to your more recent posts to my talk page, which you've requested I reply to here: Regarding your request that I or other members of the functionaries team give more insight into what happened: Jytdog posted a link to an offsite profile of another Wikipedia editor. I don't know if you'll trust that that is really the most I feel I can post without compromising the user's privacy further.
Thank you for explaining what you meant regarding the RfC. Your inference was not false; as I've explained above, I indeed disagree with the inclusion of the sentence about case-by-case linking to offwiki profiles absent the editor disclosing them themselves.
An ordinary admin can't make a block based on confidential info The 2010 Arbitration Committee statement on checkuser blocks states The Arbitration Committee has also noted that some administrators (other than Checkusers) have occasionally noted when making certain blocks that the block "should be reviewed only by the Arbitration Committee" or "should only be lifted by ArbCom." This notation is appropriate only when the block is based upon a concern that should not be discussed on-wiki but only in a confidential environment. Bases for such a concern could include information whose disclosure would identify anonymous users, could jeopardize a user's physical or mental well-being, or where the underlying block reason would be defamatory if the block proved to be unjustified. I have, however, noticed that the current version of the blocking policy seems to contradict this, when it states If a user needs to be blocked based on information that will not be made available to all administrators, that information should be sent to the Arbitration Committee or a Checkuser or oversighter for action. These editors are qualified to handle non-public evidence, and they operate under strict controls. The community has rejected the idea of individual administrators acting on evidence that cannot be peer-reviewed. I have brought up this issue on the blocking policy talk page.
Regarding your other points, I agree that it seems we will need to decide on a dedicated place for these concerns to be privately reported, though I strongly doubt that there are actually dozens of instances of paid editing each week that can only be sniffed out by revealing a user's private information. If that is indeed the case, then have they a) been ignored until this date, b) been reported onwiki, and gone almost entirely unnoticed by the functionary team, c) been addressed by individual administrators (despite your statement above that "an ordinary admin can't make a block based on confidential info," which definitely does need some clarification in policy), or d) none of the above? GorillaWarfare(talk)02:36, 6 July 2016 (UTC)
Re: discussion elsewhere that LinkedIn is somehow "public professional information" that is fair game: I cannot disagree with this more strongly. Back in the day when I first started editing (bangs cane) I gave out very little personal information on Wikipedia. I disclosed my gender, and I think my first name, as well as the fact that I'm from the U.S., but not much else. I was fairly comfortable that the average person would not be able to track me down just from "woman named Molly who lives in the U.S." I eventually was thoroughly outed, largely due to information that I did not disclose but which was submitted to 4chan by someone who had access to it, which was a terrible experience. In the end, realizing that this information was out there, I decided to cut my losses and become more open about my real life identity on Wikipedia; largely to discourage those who had outed me initially to try to do so further. The reason I am posting this, however, is that had this not happened, and had I maintained the original level of anonymity that I had intended to with my Wikipedia account, linking to my LinkedIn would have revealed a similarly devastating amount of information. Given that I am now fairly open about my identity onwiki, and that this is easily discoverable elsewhere, I will link it here for illustration. Imagine that I had, as I had initially intended when beginning to edit Wikipedia, only disclosed my first name, my gender, and that I was from the U.S. Someone viewing this profile would know that I was named Molly White, living in Boston, Massachusetts, and that I had attended Northeastern University... They would see the companies I'd worked for, the organizations with which I volunteer... Had I linked more, as many people my age do, they'd see my hometown, my summer jobs when I was in high school, maybe even my address and telephone number. People could infer from my endorsements and connections various people I might be related to. All in all, this is way more information than would ever be allowed (even in the most loose interpretations of the outing policy), and yet people are arguing that this is somehow public information, despite never being linked to a Wikipedia account?? GorillaWarfare(talk)03:43, 6 July 2016 (UTC)
This isn't really an Arb page, so I trust I'm not doing anything wrong by posting in this section. I feel for what you said here, very strongly indeed. At the same time, I would ask you to consider, and I ask this in a friendly and constructive way, whether your experiences not only inform your views about policy (as they should), but maybe also color your reactions to editors. Is it possible that, when an editor posts a link to something that contains far less in the way of personal or traceable information, and does so in good faith, trying to do something good for Wikipedia, and perhaps confused by the poor wording of this policy, you might be quicker than otherwise to see a borderline case as sanctionable outing? --Tryptofish (talk) 03:52, 6 July 2016 (UTC)
Who knows how we're supposed to structure things here; I'm not used to seeing separate "arb" sections outside of ArbCom pages. No harm done in my eyes, at least. Regarding your question, no, I don't think I am more quick to see a borderline case as sanctionable outing. The endorsement of so many other oversighters on this one only solidifies that for me. GorillaWarfare(talk)04:04, 6 July 2016 (UTC)
Thanks for the reply. I wasn't asking specifically about the present case, but rather in general. Is it common for oversighters to disagree? I'm guessing not, but that doesn't mean that the community would also agree. --Tryptofish (talk) 22:45, 6 July 2016 (UTC)
Yes, my attempt at structuring responses into 2 sections has broken down from the weight of the text. @GorillaWarfare: Thank you for a calm reply. I don't think I disagree with you to any real extent. The "case-by-case" wording should be made clear. But we also need a way to enforce the ToU and WP:Paid policy. I don't think we now can send "private" info to ArbCom on obvious paid editors at elance and similar sites. If you can figure out a way to do that, I'd be much obliged if you could let me know. I haven't seen where it is easy to get paid editors to stop their undisclosed editing, but you probably see more of this than I do. What's the secret?
As far as why I see Arbs refusing to enforce policy. It's been just over 2 years now that the ToU was approved via the RfC and the WMF Board. I haven't seen any clear cut path to getting the non-disclosers off Wikipedia. In Jan. 2015, I did see that arbcom (I think you weren't directly involved) state in the Wifione case that they had no mandate from the community to enforce paid editing disclosure rules (which was manifestly false), even though the case did not involve paid editing after the ToU was passed. Since then, I've seen mostly discouragement from admins and arbs whenever there is anything related to stopping undisclosed paid editing comes up. This small part of the policy is just a small part of the discouragement I've seen.
Well, it's getting late for me. But let me ask - do you know of any way of discouraging paid editing that does not require tons of review or long discussions with editors who act like we're trying to take the bread off their plates? Smallbones(smalltalk)04:31, 6 July 2016 (UTC)
It was worth a shot—this whole separation of replies thing tends to break down pretty fast. Anyhow, from your response, I would agree that it sounds like we don't really disagree. The "case-by-case" wording, should it stay, must be clarified. And you are correct that the ArbCom has as yet not been willing to handle paid editing cases. However, I hope that if the community decides that it is the ArbCom/functionaries/oversighters who should take this on, we will listen. I was (even until very recently) opposed to ArbCom taking on paid editing issues, but after discussions surrounding this issue, am beginning to believe that one of the three groups I mentioned before may need to do so. I only hope that we can also put in place the expectation that such a report is only the last resort, if the NPOV/COI/spam policies cannot be applied to solve the issue.
One quick pushback on something you said: I did see that arbcom (I think you weren't directly involved) state in the Wifione case that they had no mandate from the community to enforce paid editing disclosure rules (which was manifestly false). Can you point to where ArbCom was mandated by the community to be the ones to enforce paid editing disputes?
Regarding your question about discouraging paid editing, I think you'll find my answer above. Many paid editing issues can be handled via other policies without involving private information. Due to the nature of it, I expect the ones that rely solely on private information will take considerably longer. GorillaWarfare(talk)04:41, 6 July 2016 (UTC)
In #Need for a better mechanism for private reporting, above (I feel like this talk page has gotten too Balkanized, or too tl;dr), there is also discussion about where to send private information, so it can be kept private. The discussion there seems to be going against using ArbCom for that, and more towards having either the Oversighters or a new OTRS section for it. --Tryptofish (talk) 22:50, 6 July 2016 (UTC)
@GorillaWarfare: I doubt anyone here is in favour of declaring LinkedIn profiles in general "fair game". However, I think the LinkedIn profiles of professional marketing and PR writers should be. AndreasJN46608:57, 8 August 2016 (UTC)
Other Comments
I don't see that any of that is happening. There is an RfC to determine if the relaxation of the Harassment policy WP:BOLDLY added February 2015 should remain. In fact, why would we even discuss enforcing WMF's Terms of Use at Wikipedia's harassment policy page? This is entirely the wrong venue for that.- MrX16:18, 5 July 2016 (UTC)
We can discuss the enforcement of policy wherever arbs say they won't enforce policy - or they can correct my impression here if my reading of what they are saying here is incorrect. BTW, the phrase you want to remove was added after on RfC on the matter. Smallbones(smalltalk)16:49, 5 July 2016 (UTC)
The community has to decide whether it wants Arbs to enforce the 'no paid editing with disclosure' policy. As far as I know, that has not yet happened. BTW, please read my detailed comments above about why the sentence added in February 2015 was not proposed in the RfC, nor did the result of the RfC permit anyone to add it.- MrX17:01, 5 July 2016 (UTC)
"The community has to decide whether it wants Arbs to enforce the 'no paid editing with disclosure' policy." Nonsense, we don't need an RfC to decide whether a policy is enforced. We don't need an RfC to decide whether the Terms of Use are to be enforced. We do, at this point, need some guidance from Arbs on how they think our policy can be enforced since they seem to be saying that it just can't be. Any arbs who agree with Mr. X that we need an RfC to decide whether a policy will be enforced should mention that above (then they should consider resigning). Smallbones(smalltalk)17:27, 5 July 2016 (UTC)
That's a misrepresentation of my comment. I didn't say we need an RfC to "decide whether a policy will be enforced"; I said that the community has not tasked Arbcom with enforcing WMF's paid editing disclosure policy. - MrX18:09, 5 July 2016 (UTC)
Sorry if I'm misunderstanding you, but what do you mean "the community has not tasked Arbcom with enforcing WMF's paid editing disclosure policy" and how would the community "task" arbs to enforce the policy except by an RfC? Smallbones(smalltalk)18:19, 5 July 2016 (UTC)
Let me make it simple: It's not Arbcom's job to enforce WMF's paid editing disclosure policy.- MrX18:42, 5 July 2016 (UTC)
Let me make it simple for you. It is Arbcom's job to enforce all policies on en:Wikipedia (or al least supervise their enforcement). The ToU are policy on en:Wikipedia and the ToU was confirmed on meta by the largest RfC in history by 80% to 20%. WP:Paid editing disclosure is also policy on en:Wikipedia. If you're saying that Arbcom has the right to veto these policies by refusing to enforce them, you are wrong. Smallbones(smalltalk)19:04, 5 July 2016 (UTC)
I don't think that's correct, but let's discuss to see if the differences are due to semantics. When I hear "enforce all policies" I think about an affirmative duty to search out any and all violations of policies. I don't think that's remotely Arbcom's job. That's the community's job. Arbcom deals with a narrower set of issues in two ways — they are interested in the subset of policies dealing with behavior not the subset of policies dealing with content. Second, they deal with issues brought to them; they don't generally go out searching for problems. (In connection with a case there may be a need to do some research into similar situations but that's not the same as an affirmative duty to be searching for policy violations). They certainly are expected to reach their conclusions in accordance with policies — they aren't generally permitted to ignore existing policies or make up new policies. That may be a bit of a gray area; the recent creation of a new protection class feels like treading new ground as opposed to simply enforcing policy, but some may differ.--S Philbrick(Talk)20:06, 7 July 2016 (UTC)
It would be helpful if you could post the diffs supporting your contention; I'm not immediately seeing where arbitrators have said they plan to veto any policy outcomes here, but perhaps I've missed it in the lengthy set of threads. -- Euryalus (talk) 20:27, 5 July 2016 (UTC)
I'm not exactly seeing that either. But what I do see to some extent is functionaries, not just Arbs, saying that they think that a strict outing policy is absolutely the way that things must be, while at the same time, the community is responding to the RfC in a manner that is not consistent with such a consensus. I'm not so sure that anyone needs to explain themselves – but I would very, very strongly encourage those with advanced permissions to be attentive to how their views may be at odds with some community views. We need to work together to resolve those differences. --Tryptofish (talk) 22:17, 5 July 2016 (UTC)
@Euryalus: This is indeed a long discussion, and I have difficulty remembering who is a current Arb, former Arb, CU, OS etc. So I am referring in general to "those with advanced permissions."
I have reviewed the overall contributions of those folks who I can remember as arbs or former arbs and conclude that in general they are saying something like "linking to material posted by paid-editors on elance is not allowed, even if the posting was and is fully available to the public and doesn't reveal any real world names." Folks - please do correct me if I'm wrong on this! A few, e.g. DGG, say something like "in general it shouldn't be done, but let's keep the possibility open a bit."
That definitely puts a crimp into stopping obvious non-declared paid editing. So IMHO it is up to "those with advanced permissions" to let us know how we can report obvious paid editors.
I only see one Arb, @Drmies: who says anything about how to report. See here. I take this as saying, if you have such info, you cannot link to it on Wikipedia, but you can e-mail it to any admin, who can draw his own conclusions based on the evidence and block an obvious paid editor if necessary. My only questions then are whether, if a block is contested to Arbcom, can the blocking admin be punished, or will it be assumed that he is operating in good faith in drawing his.her own conclusions. Can the block be reverted by another admin? Will there be long drawn out discussions on ANI, saying there is no public evidence?
If admins are allowed to draw their own conclusions based on the evidence, and these conclusions cannot be lightly overturned, then I have no problem with that method of reporting. The devil is in the details. Smallbones(smalltalk)23:02, 5 July 2016 (UTC)
I've been trying to figure out what's going on... OK so this is about User:Jytdog... Jytdog was indeffed on June 27 by admin (and ArbCom member) User:GorillaWarfare as an oversight block... His talk page says "You have been blocked for abuse of editing privileges regarding information which has been removed from Wikipedia's public records.... the Oversight team or the Arbitration Committee must be consulted before this block can be removed" so I guess he's gone, Jim.
However, any admin can place a block like that (I think? or just oversighters?) and User:GorillaWarfare being on the ArbCom is pure coincidence -- but maybe the feeling is that since User:GorillaWarfare is on ArbCom any appeal is doomed to fail (or already has? I don't see a public record of one right off). ArbCom members need to be specially careful about making blocks that only the ArbCom can remove, because we all know how human nature works.
And User:Jytdog is a Master Editor III: he's been with since 2008 with 83k edits. And he's got an almost clean block log. And now hes just been shoved into the mass converter? What's going on?
Still, if he said "Editor MyNameIsSecret's real name is Mr Pinckney Pruddle, here's his address and here's the route his daughter takes to school", I guess I guess I can see it. If it was more of the nature of "Hey, look, here we have editor PinkneyPruddle and over at eLance theres a an account "PinkneyPruddle" writing 'please give me dollars and I will write articles in the Wikipedia' and so heads up, and if this person 1) wants to keep his personal info secret but 2) is moronic enough to post his personal info over at eLance and 3) is moronic enough to leave info linking his account here to his account there, well, my willingness to hold the hands of moronic blackguards as opposed to protecting the Wikipedia has a limit".
So which was it? If it was more like the latter then no Do. Not. Want. our Master Editor III's to be shitcanned to cater to somebody's ridiculous pearl-clutching over theoretical (not actual) harm to somebody who is avowedly here to damage, degrade, and if necessary destroy the Wikipedia.
How about User:Jytdog, huh? How about his life? He's poured 83,000 edits into the project over 8.5 years. Think it matters to him much? Maybe he cried. I would if I were banned. I guess we're just Wikipeda editors and not commercial public relations operatives, so maybe we doesn't matter. Is that the deal?
On the other hand, a quick look shows me that lot of people seem to think that User:Jytdog was a tendentious jerk. And if that's the real reason he was blocked let's hear it, and we can discuss that or whether he should have had more of a chance or what. Herostratus (talk) 23:23, 5 July 2016 (UTC)
An oversight block can only be placed and lifted by a member of the oversight team. (This would include the members of the arbitration committee.) These blocks are placed in relation to oversighted edits or non-public information. Given that such material is hidden from view from all but a few users, the details of oversighted material and blocks cannot be discussed openly. Such blocks cannot be placed frivolously or else it would be considered abuse of the tool and will lead to removal of the permission. When the block was placed, GorillaWarefare sent an email to the oversight team to discuss the block and it was supported almost unanimously by a significant number of oversighters. To my knowledge, Jytdog has not appealed the block and is more than welcome to discuss it with the oversight team or the arbitration committee if he so chooses. Mike V • Talk23:58, 5 July 2016 (UTC)
Based on my indirect knowledge from other discussions on-Wiki, I believe that Jytdog is, in fact, discussing the matter privately (via email) with ArbCom, but that the discussions have not yet resulted in ArbCom deciding to lift the block. --Tryptofish (talk) 00:03, 6 July 2016 (UTC)
@Mike V: - following the "oversight block" link twice gets
"Confidential evidence
If a user needs to be blocked based on information that will not be made available to all administrators, that information should be sent to the Arbitration Committee or a Checkuser or oversighter for action. These editors are qualified to handle non-public evidence, and they operate under strict controls. The community has rejected the idea of individual administrators acting on evidence that cannot be peer-reviewed."
This appears to contradict how I interpreted @Drmies:'s suggestion. An ordinary admin can't make a block based on confidential info. There may well be a dozen of these undisclosed paid editors in a week. Is arbcom prepared to deal with all of these? If they are not, then we need a method where undisclosed editors posting their information publicly on elance can be reported. Just saying "we can't deal with this" does not seem to be an option. We need a place to report the info, where action can be taken. Smallbones(smalltalk)00:41, 6 July 2016 (UTC)
Smallbones, my suggestion is not about such a case at all--at all. You're talking about the block of Jytdog; I'm talking about a way to handle possible COI editors. "If a user needs to be blocked based on information that will not be made available to all administrators..." has nothing to do with what I'm talking about, and note that I never said, nor do I think, that in the scenario I sketched there's some demand that some admins be excluded. Drmies (talk) 02:14, 6 July 2016 (UTC)
@Mike V:"the details of oversighted material and blocks cannot be discussed openly" is not at all true. You can say "He linked to an a commercial editors Elance account" or "He linked to long-term editors Facebook page" or "He posted another editors name and social security number" or whatever -- and there's a huge difference between these. You can't post the contents of the oversighted material. You also have to be careful not to leave clues which could reasonably lead to the oversighted material (more or less, depending on circumstances).
Asking us to just trust the functionaries, trust the oversighter to approve the other oversighter, and that's all you need to know -- boy, have you got the wrong project. This is not IBM. We are not employees. I'm not even wearing a suit right now. We are part of a community. We need to be vouchsafed as much information on these contentious decisions as is reasonably possible and stretch the boundaries as far as your reasonably can -- which is a whole helluva lot further than "the details of oversighted material and blocks cannot be discussed openly". You don't do that, you lose the community.
Geogene, I will go with sorta. The reason being, is that this RfC is not at all about paid editing, or Elance, so my opinion on that hypothetical situation shouldn't matter. This RfC is about inclusion of the line "Posting links to other accounts on other websites may be allowable on a case-by-case basis." in the harassment policy. But I suppose, if the question is "Can a user post at COIN a link to job to create a Wikipedia on Elance in order to get it on the radar (and watchlist) of those that work in that area?" My answer would be, "it depends on what is included in that link." Is it just a link to the job. Probably no issue. Does it link to a username of someone who agreed to do the job, but no real name attached to it? Possibly no issue. Does it link to a username with real name, email address, phone number, or any other piece of personally identifiable information? No, don't include it. You would be linking a Wikipedia user (regardless of if they have Wikipedia's best interests in mind) to the information above. As I said before, that has never been accepted practice. And as others have said, the Harassment Policy applies equally to all users. --kelapstick(bainuu) 02:57, 6 July 2016 (UTC)
(edit conflict) But see, the reason why we don't disclose what is oversighted is because that would provide clues that could lead to the oversighted material. Using your example of a Facebook page, if we informed the community that the such a link was oversighted, other individuals will try to find the link for themselves. Some might do this with malicious intent and others for their own personal desires. (Yes, this does happen. I've denied multiple request from term editors who email me and ask for revision deleted material, simply so they they can be "in the loop".) Providing this has minimal benefit to the community and significant harm to the subject of the harassment.
The arbitrators and the functionaries are trusted by the community. Members of both group have undergone vetting by the community and members of the arbitration committee to handle sensitive information and respond to serious concerns not suited for open discussion. As Smallbones pointed out above from the blocking policy, checkusers, oversighters, and arbitrators are entrusted to take action on information that is not publicly available. Mike V • Talk02:43, 6 July 2016 (UTC)
Mike V, where you said (quite correctly, in my opinion) that it's important that oversighters not post anything that would lead other people to track down the oversighted material, that reminded me that oversight blocks of editors who are well-known in the community often result in Streisand effects that lead to the same problem. (I think anyone who has closely followed the block that set off this brouhaha can pretty well reconstruct how to find the private information. And that's a terrible disservice to the editor who may have been outed.) Of the many issues that are getting aired in these discussions, I hope that we will figure out how to reduce those unintended block effects. --Tryptofish (talk) 03:09, 6 July 2016 (UTC)
I am reading above that User:Jytdog is indefinitely banned. I am someone who was unjustly chastised by Jytdog and he encouraged other people to ignore my suggestions. So I am not a fan. However, this must be one of the greatest injustices in the history of Wikipedia. You take an experienced person albeit with faults, who persevered in an important task for many years, and he takes an action designed to protect the community, does no harm at all, and he is banned immediately and forever on a technicality. I am shaking my head in disbelief at the immaturity of the justifications, and head-in-sand naivety that wouldn't even be funny on The Office, yet alone an actual place of arbitration. This is not merely an injustice, but Rome is burning. Some people are going to make a pile of money out of this execution. I'd say the "Wikipedia For Sale" sign up is now up, but it's valuable public space is available to any half-intelligent group of PR types for free. Travelmite (talk) 14:08, 22 July 2016 (UTC)
Travelmite, I very much appreciate that you have expressed those concerns in spite of having had a difficult experience of your own. There have been too many other users who hold on to grudges, which does no good for either them or for Wikipedia. And I, too, am very concerned over how long the block is continuing. In Wikipedia-speak, he actually is not banned, but rather blocked, and the premise is that it is not supposed to be permanent, but rather only until Jytdog and ArbCom can come to some sort of understanding about how he will understand the harassment policy going forward. And yet, very concerningly, all this time has passed without an unblock. I do know that there have been some technical problems with the emails between ArbCom and Jytdog not getting through, so that may be part of it. Beyond that, I don't know, and I don't expect that ArbCom will clarify anything. Perhaps, people are being stubborn, on either or both sides. --Tryptofish (talk) 18:07, 22 July 2016 (UTC)
"This applies to personal information of both editors and non-editors"
I have reverted the removal of this statement, but recognize that it probably needs better wording. The intention of that statement was to capture situations where the insertion of personal information about a "non-public" person who was not otherwise involved in Wikipedia (either as an editor or an article subject) would be captured. It is extremely common to see edits about such individuals that include personal information (sexual orientation, addresses, etc.) or that is extremely derogatory; the purpose of these postings is almost universally to harass the individual, since it will frequently become their #1 google hit. I think we can agree as a community that it is not appropriate to use Wikipedia as a vector for harassment of this type, and it should be considered sanctionable behaviour. Let's work on the wording here. Risker (talk) 17:23, 17 July 2016 (UTC)
The wording bothers me too, because it ends up treating all persons the same as Wikipedia editors for the purposes of the outing policy, and I don't see how one can "out" someone who is not a Wikipedia editor. This has been discussed before, pretty recently, and one can see it in the most recent talk page archive, where I also suggested some alternative wording. Interestingly, the original form of the sentence was added very early on by someone called Jimbo Wales. I think the idea is that one should not post private information about some non-notable person on a talk page. One can imagine someone misusing Wikipedia as though our talk pages were some kind of website where people post nasty stuff about other people. But it seems to me that in article space, as distinguished from talk space, the operative policy is BLP, not the policy here. --Tryptofish (talk) 20:52, 17 July 2016 (UTC)
The issue we have is that some appear to want to have the outing policy apply as broadly as possible. Should no linking to off site information apply to non-editors? IMO no. Doc James (talk · contribs · email) 23:06, 17 July 2016 (UTC)
To note. This is the harassment policy. There is no outing policy, and you both ought to know better. Every time you think "outing", switch the word to "harassment" and you'll start understanding much, much better. Risker (talk) 23:53, 17 July 2016 (UTC)
"You both ought to know better"? If you want to disagree with Doc James, go right ahead, but that should not have been directed at me. I used the phrase loosely to refer to the section of the policy page, and nothing in what I said reflected a lack of awareness of its relationship to harassment. --Tryptofish (talk) 22:45, 18 July 2016 (UTC)
We are discussing "posting of personal information" to which WP:Outing redirects. That some wish to describe all posting of personal information as harassment does not mean that it is. Doc James (talk · contribs · email) 01:39, 18 July 2016 (UTC)
(edit conflict)An interesting case study might be Caitlín R. Kiernan, which is a transgendered author/scientist who had kept her former name non-public. Some time in 2016 people put together the name on the science papers with the current name under which she publishes fiction and added it. I'd made this connection in 2013 but, because of privacy concerns, refrained from including it in the article. So I guess the discussion could be a) was the initial restraint justified and necessary and b) is the current state of the article violating any guideline and c) does the guideline adequately cover cases like this one? I feel comfortable asking this now because the subject appeared at Talk:Caitlín R. Kiernan#Birth name with the account AuntBeast64 to discuss it herself. - Brianhe (talk) 01:41, 18 July 2016 (UTC)
Well, look. This is what the paragraph 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. Personal information includes legal name, date of birth, identification numbers, home or workplace address, job title and work organisation, telephone number, email address, other contact information, or photograph, whether any such information is accurate or not. Posting such information about another editor is an unjustifiable and uninvited invasion of privacy and may place that editor at risk of harm outside of their activities on Wikipedia. This applies to the personal information of both editors and non-editors. Any edit that "outs" someone must be reverted promptly, followed by a request for oversight to delete that edit from Wikipedia permanently.
So, "personal information", as defined in this paragraph, includes legal name, date of birth, job title, work organisation and photographs of both editors and non-editors, and posting such on Wikipedia is forbidden "unless that person had voluntarily posted his or her own information, or links to such information, on Wikipedia". As written, it's an obvious absurdity, making every BLP that includes someone's photograph, date of birth, job title etc. a WP:Harassment violation.
Let's have a policy that actually says what it means, shall we?
A key problem here is that the term "personal information" is overly broad. What we are concerned about is private information, not personal information, because personal information (like date of birth, job title, photograph etc.) is a core part of almost every Wikipedia biography. So let's focus on what is private. For biographies of living persons, private includes almost everything not written about in a professionally published source (like their home address, telephone number and so forth). On the other hand, biography subjects' social-media posts, made from a verified account, are sometimes used as sources. I also think that Wikipedia editors generally agree (although some biography subjects may disagree) that it is not "harassment" if a Wikipedia editor takes a photograph of a notable person and posts it on Wikipedia—yet this is literally what the paragraph as it stands asserts. All in all, the paragraph needs some work. --AndreasJN46615:00, 18 July 2016 (UTC)
Proposal (draft)
Risker and all, here is a draft re-write of the paragraph:
Posting private information about a living person on Wikipedia may be harassment if that person had not voluntarily posted this information, or links to such information, on Wikipedia themselves. Private information includes personal identification numbers, date of birth, job title and work organisation, contact information such as telephone numbers, home, workplace or email addresses, and legal history, including traffic citations and involvement in court cases, unless the information is a matter of public record in reliable sources satisfying both the Biographies of living persons policy and the Verifiability policy. In the case of non-notable people, including in particular all non-notable Wikipedia editors, private information additionally includes legal name and identifying photograph, whether any such information is accurate or not. Posting such information without the consent or approval of the other person is an unjustifiable and uninvited invasion of privacy and may place the other at risk of harm. Any edit that "outs" another editor must be reverted promptly, followed by a request for oversight to delete that edit from Wikipedia permanently. Any administrator may redact it pending oversight, even when the administrator is involved. If an editor has previously posted information about themselves but later redacted it, it should not be repeated on Wikipedia, although references to still-existing, self-disclosed information are not considered outing. If the previously posted information has been removed by oversight, then repeating it on Wikipedia is considered outing.
Posting personal contact information, including phone numbers of the person or his/her employer, email addresses of the person or his/her employer, physical addresses, date of birth, arrest and citation records, job histories, and photographs for which consent has not been given is always forbidden.
Other material concerning a person who has edited on Wikipedia may be posted only where a specific and actual connection between the article edited and the information listed is apparent, and readily found, or posted by that person.
Violation of this is to be considered "harassment" of the person, whether or not it meets the normal definition of "outing."
All violations of this policy may be removed "on sight", and repeating such material is also forbidden on Wikipedia.
Collect, the problem is that posting a photograph for example is not "always" forbidden, nor is it always harassment. For example, people take pictures at Wikimania and upload them; others wish someone a happy birthday on Wikipedia, or use their legal name because they know them offwiki. I'm all in favour of clear language, but then it has to be well circumscribed enough not to include any number of harmless activities that have no harassing intent at all. AndreasJN46614:05, 19 July 2016 (UTC)
In the cases you mention the poster should have a solid idea of whether the person wants them to use their name on Wikipedia or gave content to have their picture posted (I would say "posed" pictures at Wikievents are likely OK but "candid" shots the photographer needs to check with the subjects or take them down on request. It really is simple courtacy.) In the class of cases you describe 99% of the time an "ooppss, sorry" followed by a RevDel/Oversight and no repeating the disclosure is a pretty solid presumption of no intent to harass and should not be punished. Really one needs to be sure another editor will not consider the types of posts you mention problematic and will not complain about them before making such posts. There are of course exceptions but that is why admins have some discretion.
That said I like Collect's wording. More than anything a standard mini-course for admins on recognizing and dealing with "harassment" and actual harassment would help. JbhTalk14:37, 19 July 2016 (UTC)
To cover that cavil, I added "for which consent has not been given." When one writes gaping loopholes in any rules, the loopholes will be abused all too often. If the people wish to post about themselves, then they can do so, and that is quite sufficient. YMMV. It is past time for Wikipedia to recognize some real and ongoing problems which can no longer be excused as "having fun." You likely should note that I do not define the behaviour as "harassment" but as "forbidden." Also note, my suggested wording does not bar people from saying "Happy Birthday, George" or the like at all. Collect (talk) 15:26, 19 July 2016 (UTC)
@Collect: Would the wording you propose replace the entire paragraph from "Posting another editor's personal information ..." to "... is considered outing."?
And when you say, Other material concerning a person who has edited on Wikipedia may be posted only where a specific and actual connection between the article edited and the information listed is apparent, and readily found, or posted by that person., what "other material" are you thinking of? Examples? --AndreasJN46615:37, 19 July 2016 (UTC)
Yes - but subject as always to suggestions for greater clarity.
The object here is to give an exception for, example, an IP which is linked specifically to a given MP or Congressman, etc., or to a specific corporation where the article is about the corporation, etc. without destroying the core principles involved.
And to note that a person who self-identifies as having, for example, a COI may be mentioned as having that COI. "User GGnarph self identifies as George Gnarph" in which case it would be silly to say this could not then be noted by anyone else on Wikipedia, I fear.) I know of no case where personal contact information (such as personal phone numbers, etc.) should ever be posted in a public message. If you have an exception, then we certainly can word this to accommodate reasonable positions. Collect (talk) 15:15, 21 July 2016 (UTC) (sorry for sig delay)
Jayen466 I do not support this text because I am in favor of uploading more court documents to Commons and Wikisource. There are some legal cases which enter public discourse and even more which are discussed by legal professionals, and Wikipedia might link to primary documents. I do this somethings. The revision draft here has several mentions of court records. Court records are not reliable sources, but I do think it is desirable to link to them as supplemental primary sources whenever they are discussed. The point of uploading them to Commons will almost never be to draw attention to personal identifying information, however, court documents do routinely include such information. I think that the text you have above would have the unintended consequence of decreasing accessibility to all kinds of court documents which would be intuitive to include in Wikipedia articles discussing particular legal issues. When legal issues arise, some court documents become public, and I am not convinced that Wikipedia should have a ban on including the most relevant primary source material when it can be either included as a document in Commons or linked as an external link. It is not traditional to call the routine public record creation of a court "harassment". Many public records are not wiki reliable sources, are not addressed in BLP policy, and are not addressed in V. Blue Rasberry (talk)22:16, 28 July 2016 (UTC)
@Bluerasberry: What's proposed is a Wikipedia policy only; it does not affect Wikisource or Commons. Linking to primary sources (including court records) is fine if the matter has been discussed in a professionally published secondary source. BLP policy explicitly states, and has done so for a long time, "Do not use trial transcripts and other court records, or other public documents, to support assertions about a living person". The point of that paragraph of BLP policy is, if there has been no prior public interest in someone's dealings with the courts, Wikipedia should not be the vector for bringing the matter to public attention; that's the job of journalism. AndreasJN46600:32, 30 July 2016 (UTC)
"Private information includes personal identification numbers, date of birth, job title and work organisation, contact information such as home, workplace or email addresses, and legal history, including traffic citations and involvement in court cases, unless the information is a matter of public record in reliable sources satisfying both the Biographies of living persons policy and the Verifiability policy." The problem with this is that job title and work organization are both directly relevant in whether somebody should be considered an expert or not, and existing content policy assumes that it's possible to determine this in a transparent manner. This sets up a potential conflict between content policy and behavioral policy. Which do you think we should give priority? Geogene (talk) 01:36, 29 July 2016 (UTC)
@Geogene: Yes exactly you describe the situation I was trying to address.
@Jayen466: Talk me through this. Suppose that there is a Wikipedia article about a court case. The Wikipedia article talks about the legal significance of whatever issue went to court, and perhaps the participants in the court case are hardly identified in the Wikipedia article, or perhaps they are just named. If somehow court documents are public records, and they are in Commons or Wikisource or posted online in any place which rightly identifies them as intended for public review, then is it okay to link the Wikipedia article to the court documents hosted elsewhere? Suppose further that no one is trying to add primary information to Wikipedia based on the court documents, but rather that the court documents are only being presented as media related to the subject of the Wikipedia article.
Making the link would comply with "Do not use trial transcripts and other court records, or other public documents, to support assertions about a living person" but simultaneously fail "Wikipedia should not be the vector for bringing the matter to public attention" by linking to the documents, which might include all sorts of personal identifying information. The real intent would be to link to the court documents which are of general instance, but in doing so, it just happens that court documents contain personal information. Does the desire to avoid sharing personal information override the desire to share relevant legal documents, whether in Commons or hosted anywhere else as a public record? Blue Rasberry (talk)18:25, 1 August 2016 (UTC)
@Bluerasberry: Maybe a slight tweak would address this concern, as follows: "Private information includes personal identification numbers, date of birth, job title and work organisation, contact information such as home, workplace or email addresses, unless the information is a matter of public record in reliable sources satisfying both the Biographies of living persons policy and the Verifiability policy. Private information also includes personal legal history, including traffic citations and involvement in court cases, unless the cases concerned have been covered in reliable sources satisfying both the Biographies of living persons policy and the Verifiability policy."
This should take care of it, because if we have a Wikipedia article about a court case, this presupposes that there are reliable sources about it. Alternatively, I am happy to drop any reference to legal history/court documents at this point in time and leave this for another day, or leave it up to BLP policy.
I should mention that the references to WP:BLP and WP:V are deliberate, as these are the policies governing the use of self-published sources. So, for example, a paid PR writer's offer of services would be a self-published source falling under WP:BLPSELFPUB in BLP policy, and WP:SELFPUB in the Verifiability policy. --AndreasJN46619:34, 1 August 2016 (UTC)
Jayen466 I am not sure. I have doubts that the text you provided addresses the issue that I wished to raise. I think that I disagree with the idea that all of this information is private unless it appears in what Wikipedia identifies as "reliable sources". I agree that the text of Wikipedia articles should not contain information from primary sources, but I am not ready to agree that information in primary sources is private. The kind of information that you list is publicly available through organization homepages, the annual reports of organizations, the affiliation disclosure section of every academic paper, public records including court documents, and self-disclosed in many public places online. I think it is reasonable to have a rule saying not to bring such things into Wikipedia's text, but if the rest of the world calls all these sources "public" then I do not think that Wikipedia should call them "private".
I am not sure how to advance this discussion. The example that I have in mind is that there is a court case well-covered in reliable sources and with coverage on Wikipedia. The case happened in US federal court, and a range of documents about the case are in the public domain for being created by US federal government employees. There are some select documents which concisely describe the details of the case, and are the primary sources from which all secondary sources and commentary are derived. Those primary sources reveal everything you suggest is private, and they are also part of the public record and were distributed in press releases by the federal prosecutor. I am not suggesting that anyone take primary information out of these sources and put it in Wikipedia, but at the same time, it seems right to consider that Wikipedia's overview of the issue would link to the original source and perhaps even the text documents themselves as thumbnails in the Wikipedia article linking to the Commons file.
Suppose that we kept your text exactly as you provide it. Would there be some kind of addendum which communicated that readers might be one click away even within Wikipedia itself from accessing all of this prohibited content? Why is it useful to have a rule like this when editors may in fact provide an external link or possibility even original documents which do give all of this information? What would a privacy rule change, when the major regulation is still whether a source meets WP:RS? Blue Rasberry (talk)20:40, 5 August 2016 (UTC)
@Bluerasberry: Harassment can happen both in article space and in other namespaces. Let me give you an actual case study. A borderline-notable academic terminated a tenancy, because the lodger appeared to engage in unlawful activities on the premises. The lodger's response was to go to the academic's Wikipedia biography and introduce a section "Brushes with the law" that listed traffic citations and described a private legal dispute (a non-criminal matter) that academic had been engaged in, citing court records. It came to dominate the biography, even though no secondary sources had ever taken note of these matters. The subject removed the content from their biography, and it was twice reinstated by other Wikipedians because the lodger had cited court records. Were the lodger's actions harassment? I think we'd agree they were. So, how do we reflect this in the policy?
Actually, I wish I had not brought this matter of court records into my rewrite. Never a good idea to try to do two things at a time! The main problem I was trying to address with the rewrite is that according to policy as written right now, almost every biography Wikipedia contains is "harassment". Let's go through the current wording. 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. Personal information includes legal name, date of birth, identification numbers, home or workplace address, job title and work organisation, telephone number, email address, other contact information, or photograph, whether any such information is accurate or not. Posting such information about another editor is an unjustifiable and uninvited invasion of privacy and may place that editor at risk of harm outside their activities on Wikipedia. This applies to the personal information of both editors and non-editors.
Note first that there is no sharp dividing line between editors and non-editors, as a single click can turn any member of the public from a non-editor into an editor. Indeed we have seen that biography subjects often do become editors, even if only to edit their own biography. (Moreover the paragraph explicitly states that everything said applies to both "editors and non-editors" alike.) Now, almost every Wikipedia biography lists personal information as currently defined in that paragraph that the person himself or herself has not disclosed on Wikipedia, beginning with a legal name and a date of birth. Many biographies similarly include job title, work organisation and a photograph, something that this paragraph as written explicitly forbids. Hence I brought in reliable sources: it's okay to post personal information on Wikipedia if it is a matter of record in reliable sources satisfying WP:BLP and WP:V. Do you follow me so far? --AndreasJN46607:47, 6 August 2016 (UTC)
You've posted this meme in several places now, and honestly I'm at a loss how you've reached that conclusion. Writing a BLP-compliant biography of someone who happens to be an editor is not (usually) harassment. Saying "User:W.Coyote is really Wile E. Coyote and he's a celebrity spokesperson for Acme Corporation" is, unless he's disclosed that. Opabinia regalis (talk) 08:14, 6 August 2016 (UTC)
@Opabinia regalis: I'm at a loss to see how you avoid that conclusion. It's what the policy says, in black and white. Try to read WP:OUTING with novice's eyes, forgetting your Wikipedia experience, forgetting that you know what is meant. The policy even emphasises that what it says applies to editors and non-editors alike. That's why it needs re-drafting.
As for Wile E. Coyote, if Wile E. Coyote publishes a press release saying that he has created the article on Acme Corporation this month, and the edit history shows that User:W.Coyote started that article this month, it is not in any way, shape or form "harassment" to infer that Wile E. Coyote is W.Coyote. But this is a point on which reasonable people may differ, and it really has nothing to do with the point I'm making in the preceding paragraph. The point I am making in the preceding paragraph is about the meaning of plain English words. A policy should say what it means, and mean what it says. This one does not. AndreasJN46616:01, 6 August 2016 (UTC)
And why are we taking at face value what a press release of unknown provenance says? Joe jobs and impersonation are common; maybe it's a Road Runner plot. Maybe the press release misprinted the name. Maybe the editor misunderstood and didn't realize their information would be included in the public text. It's easy enough to ask first.
As for the policy text, I disagree,; reading with "novice's eyes" does not suggest that writing biographies is harassment. No one who is not deeply steeped in wikipolicy would entertain that reading. Novices mostly don't read policies at all, and when they do, they don't finely parse them based on their background understanding of prior community disputes. Opabinia regalis (talk) 17:34, 6 August 2016 (UTC)
(edit conflict) The policy says that posting the legal name, date of birth and job title of another person (editor or non-editor) on Wikipedia is harassment ... Surely we can do better than that. Instead of pretending there is no problem, it would be more constructive to work on improving the text.
Joe Jobs: Taking the recent press release brought up by Smallbones below as an example, do you really think it likely that this was a Joe Job? Yet this policy as written forbids our talking about it, or linking to it. Joe Jobs happen, just as miscarriages of justice happen. That is not a reason to shut down the courts.
Note too that on those occasions where editors were outed in the press (off the top of my head, I recall Wiki-PR, Hari and Young), community consensus was to include the information in Wikipedia, even in mainspace. The decision to do so was not even controversial. So sourcing matters. Basing an allegation on a pseudonymous social media post that anyone could have made is as incompatible with BLP policy as using such a post as a source for a statement in a biography. But if a high-quality source asserts that Editor X is Person Y, community consensus has been against censoring that information. Clearly there are many shades of grey in-between, but I believe it's wrong to say you can't even raise the question if there is good evidence. AndreasJN46618:29, 6 August 2016 (UTC)
The sentence about both "editors and non-editors" goes way back in wiki-history, and was actually written by Jimbo himself. It's an anachronism. Whatever the fine print about press releases and whatnot, I think that the policy here should be what applies to editors and talk space, and the content of BLP pages (whether or not the BLP subject also happens to be an editor) should be governed instead by the BLP policy, not by this policy. --Tryptofish (talk) 18:03, 6 August 2016 (UTC)
Thank you. Yes, this policy should govern discussions of active Wikipedia accounts, and BLP policy should govern what is written about named people.
It occurs to me that quite possibly Jimbo had people like editors' spouses in mind when he wrote about "non-editors". But if that is what we mean, we should make that clear. AndreasJN46618:43, 6 August 2016 (UTC)
I'm under the impression that the intention was to prevent what we would now call a BLP violation: essentially using content to attack a person. It was like saying don't use Wikipedia to attack anybody. It was a good thought at the time, but BLP policy didn't even exist then. I think it's appropriate to extend the outing policy to apply to something like using a talk page to post doxing information about a non-notable non-editor – we clearly don't want that. But once we get into the realm of notable non-editors, it no longer makes sense to call it outing. --Tryptofish (talk) 18:49, 6 August 2016 (UTC)
FWIW, here is the original edit Tryptofish is referring to: [1]. At the time, the paragraph read: Posting another person's personal information (legal name, home or workplace address, telephone number, email address, or other contact information, regardless of whether the information is actually correct) is almost always harassment. This is because it places the other person at unjustified and uninvited risk of harm in "the real world" or other media. This applies whether the person whose personal information is being revealed is a Wikipedia editor or not. (Jimbo added the words marked in bold.) It was really the changes and additions made since then that made the passage nonsensical. AndreasJN46618:59, 6 August 2016 (UTC)
I think you're right above that the usual current use for the "non-editors" thing is in reference to editors' off-wiki associates, and to importing disputes from off-wiki. That said, it doesn't really matter whether you consider it a form of outing or not; edits revealing BLP subjects' private information routinely get suppressed. Opabinia regalis (talk) 04:09, 7 August 2016 (UTC)
@Opabinia regalis: "Personal information" as defined by the current wording of this policy includes "legal name", "date of birth", "job title", "work organisation" and "photograph". None of these, the policy says, may be posted on Wikipedia unless previously revealed by the person, on Wikipedia. Are you willing to oversight every Wikipedia biography that contains the subject's "legal name", "date of birth", "job title", "work organisation" and/or "photograph"? AndreasJN46608:32, 8 August 2016 (UTC)
Although OR can and doubtless will respond for herself, it seems kind of obvious to me that posting such information about a BLP-notable person as part of the page about that person is not (or should not be) what the policy here is about. Conversely, posting those things about an editor who has not chosen to reveal them voluntarily is outing. --Tryptofish (talk) 22:24, 8 August 2016 (UTC)
@Opabinia regalis: You come on. A better use of our time would be to come up with a paragraph that does not contain obvious logical absurdities. This kind of policy writing is (1) slovenly and (2) insidious, because it implies that policies can be assigned whatever meaning someone wants to assign to them. If by "non-editors" we mean editors's off-wiki associates or relations, as you assert above, then let's simply say so. Any objection? AndreasJN46600:21, 11 August 2016 (UTC)
Sounds excellent to me. I would support having this replace the current sentence "This applies to the personal information of both editors and non-editors." --AndreasJN46614:09, 12 August 2016 (UTC)
While the whole of the proposal at the archive page is, at least on initial inspection, very sound. I would not support a simple replacement of the current sentence as proposed immediately above. It is not acceptable to reveal even public personal information of an editor, where that information has not been linked by that editor to their Wikipedia account(s), and we need the policy to be clear on that. - Ryk72'c.s.n.s.'14:33, 12 August 2016 (UTC)
Editors are not permitted to reveal the personal information of other editors on Wikipedia. This is unaffected by suspicions of conflicts of interest, including suspicions of paid editing, or indeed by suspicions of anything else. This does not prevent inclusion of personal information of article subjects (subject to compliance with WP:BLP, WP:V, etc), even if those subjects are also editors. And, for mine, it does not preclude highlighting that requests for paid editing have been made (such as in the Elance example above). It does preclude revealing, on Wikipedia, personal information about editors, even if those editors have made that information available elsewhere. Misunderstandings of this are starting to seem deliberate. - Ryk72'c.s.n.s.'18:02, 12 August 2016 (UTC)
It appears some doe not agree with "it does not preclude highlighting that requests for paid editing have been made (such as in the Elance example above"
I am certain that there would be increased support for highlighting that requests for paid editing have been made if it was not alloyed with harassing editors by "outing" based on suspicions. - Ryk72'c.s.n.s.'20:33, 12 August 2016 (UTC)
Ryk72, here is how the complete paragraph would read after implementation of Tryptofish's proposal:
Posting another editor's personal information is harassment, unless that person had voluntarily posted his or her own information, or links to such information, on Wikipedia. Personal information includes legal name, date of birth, identification numbers, home or workplace address, job title and work organisation, telephone number, email address, other contact information, or photograph, whether any such information is accurate or not. Posting such information about another editor is an unjustifiable and uninvited invasion of privacy and may place that editor at risk of harm outside their activities on Wikipedia. It is never acceptable to use Wikipedia in any namespace for the revelation of any person's non-public personal information without consent, whether editors or anyone else. Posting private information of non-editors may also be a violation of the biographies of living persons policy. Any edit that "outs" someone must be reverted promptly, followed by a request for oversight to delete that edit from Wikipedia permanently. Any administrator may redact it pending oversight, even when the administrator is involved. If an editor has previously posted their own personal information but later redacted it, it should not be repeated on Wikipedia, although references to still-existing, self-disclosed information is not considered outing. If the previously posted information has been removed by oversight, then repeating it on Wikipedia is considered outing.
If you read it as a whole, isn't the point about editors' personal information that you are concerned about as clear and strong as before? The very beginning of the paragraph (marked in bold above) addresses that.
The two sentences Tryptofish proposed (marked in italics above) do not legislate discourse about editors, but make a broader point about any and all people anyone might write about in Wikipedia, including biography subjects.
Tryptofish's wording is to my mind clearly superior to the current wording, which fails to take into account that notable people's legal names, dates of birth, job titles, photographs, even home addresses on occasion (see Bill Gates's house), may be legitimate encyclopedic information to include in Wikipedia and present to readers. The current policy wording simply isn't in line with routine community practice; Tryptofish's is. --AndreasJN46608:54, 13 August 2016 (UTC)
While I agree that the current "editors and non-editors" phrasing causes issues for the literally minded (in whose company I count myself); I should feel infinitely more comforted, and infinitely more comfortable with the "non-public personal information", if the section immediately below this one did not include attempted justification of attempts to link editors with personal information from profiles on external websites on the basis that these are "public". - Ryk72'c.s.n.s.'17:42, 13 August 2016 (UTC)
Thanks. I'm not really comfortable with changing just this wording in isolation. That's why my archived proposal was for a rewrite of the section as a whole. But I think that the events that have transpired between my earlier proposal and now have revealed such a lack of consensus about the most basic aspects of the policy that I think we are a couple of RfC's away from being able to make a useful revision. --Tryptofish (talk) 20:53, 13 August 2016 (UTC)
Word definitions always have subjectivity to a degree. To me, "outing" is "revealing who the editor is behind the computer screen". In other words, if somebody has an account and they don't put any personal information on, a good friend of theirs also edits Wikipedia and says "the person who uses this account is called this", that would be outing. But what about genealogy and company sites that anybody on the Internet can look at, such as findmypast, genesreunited, and companycheck? For example, if the operator of an account put on their userpage what their real name is, another editor looked the name up on findmypast and saw only two people (let's make a situation up), and one person was born in 1896 and the other in 1984. If the person who looked the name up linked to findmypast and said "this is them (inserts link here)" on Wikipedia, is that outing? Plankton55 (talk) 09:55, 18 August 2016 (UTC)
If posting that research reveals the editor's real life identity, then yes, it is outing. But simply citing a website about genealogy as a source for content is not outing, even if the website also happens to include information about people who happen to be editors. --Tryptofish (talk) 19:20, 18 August 2016 (UTC)
Ok. I have another question. Say, for example, somebody who is not famous/only known in the area they live is mentioned in a Wikipedia article. And let's make this up: if they had a Facebook/Instagram account, would it be acceptable to:
use info about them on WP that is in their Facebook profile and is set so it's available to everybody, even those not logged in?
and what would happen if after an editor added information available to everybody (what I've mentioned above) to WP and later the Facebook user decides to make the information private: would we have to remove the content from WP immediately (as soon as possible)?
I think it would become a moot point in an article, because websites like Facebook would not be used as sources in the first place, because they fail WP:RS. If, unlike your example, information about an editor was taken from Facebook and repeated on a talk page, that would probably violate the outing policy, and even more so, if the information had subsequently been redacted. Now there's another twist about that: we have had bad-faith COI editors who posted on their websites that they were violating policies here, but then, after they were called out for it, they took the information down and complained that they were being outed for accusations that were made before they took the information down. Not something that happens often, but I've heard about it happening. --Tryptofish (talk) 23:51, 18 August 2016 (UTC)
Unless there is some issue about Kirsty McCann being a Wikipedia editor, there's no issue at all about outing. On the other hand, I'd say that edit has big problems with WP:NOR and WP:RS. --Tryptofish (talk) 22:21, 19 August 2016 (UTC)
Undisclosed paid promotional editing
When one blocks what appears to be an undisclosed paid promotional editor and someone asks for the evidence what is / is not a person allowed to do?
At this point in time it would obviously be unwise to post any evidence on WP. Can one email it to people? And if so whom? Or is WP:DUCK all the justification that is needed.
Suggest that the best evidence to link to would be diffs of non-neutral POV pushing, failure to accept consensus, re-creations of consensus non-notable articles, or similar. If there is no on-Wiki evidence of disruptive editing, then for what are we blocking? And, as a side question, why is paid editing any more egregious than non-paid POV pushing? - Ryk72'c.s.n.s.'01:46, 10 August 2016 (UTC)
I did not say paid editing was more egregious than say Scientologists promoting their religion across multiple articles.
The problem with paid vs. non-paid anything is that money is a great incentive to fight a fight that you might not otherwise care about. Let me ask you this - would you revert edits on an article for free just because you didn't like them if someone kept putting them back, or would you be more likely to do it because because "your version" got you paid a few thousand dollars (no specifics given, but an actual situation, so don't accuse me of making up a ridiculous number)? MSJapan (talk) 04:31, 10 August 2016 (UTC)
And people will not understand rules when the good name of their faith, creed, doctrine, nation, belief system or identity depends on not understanding them. A quick perusal of ArbCom past cases will reveal any number of content disputes where editors edit disruptively without requiring payment. Conservatives, progressives, radical feminists, MRAs, fringe theorists, people on both sides of both "I-P" disputes are all capable of being as disruptive, as conflicted (COI), and as POV-pushing as paid editors. - Ryk72'c.s.n.s.'18:11, 12 August 2016 (UTC)
I wholeheartedly concur with those editors. But the existence of such an advertisement is neither sufficient nor necessary evidence to block an editor; we would still need to demonstrate, and we would only need to demonstrate, that the editor was editing non-neutrally, promotionally, or otherwise disruptively. - Ryk72'c.s.n.s.'17:23, 13 August 2016 (UTC)
That is just plain wrong — if they are engaged in undisclosed paid editing none of that matters! If you read the policies you will realise it doesn't matter one iota whether the edits are considered neutral or even beneficial to the page or not.Carl Fredrik 💌📧22:04, 13 August 2016 (UTC)
Aside from the issue of blocking, however, I consider the most important point to be that having one's personal information at such a site should be considered public information, and therefore not considered outing if cited here. That is how I would see it. Thus an editor pointing the site out at COIN should not be blocked. And yet – there are clearly plenty of users who disagree with me about that. --Tryptofish (talk) 20:47, 13 August 2016 (UTC)
off-topic
Not trying to nit-pick but I would like to discover reasoning or prove a point. Is there a reason we are cascading these discussions all across the page? This either forces a break (if an editor knows how to do this), that just adds confusion (maybe just to me) or at a point forces an editor to:
Restart on a line another editor was using when it was not necessary. In our utterly confusing discussions, bouncing from one to the other, I may be the only editor that has to reread up and down the discussion(s) several times (old(er) age may be a reason) but it just seems we can do without this line spreading unless there is some reason I do not know about other than confusion assistance. We have an editor;
replying under Doc James then:
way out here under the same section;
and I can't see how we got past possibly this point; that would lead to my comments somewhere around;
here. I just am not clear on the reasoning so if someone would kindly provide this, or humor me, and possibly any other editors unless I am the only one (then maybe revert to the humor me part), so I can more easily follow these discussions (sometime commenting) and maybe we can just list it as being nice to those of us "disadvantage" editors.
Which would look better (again--maybe just to me) than way out here.
The conversation seems to be following a standard layout for replies. I suggest you read WP:THREAD which will explain how to read discussions. In a nutshell, you only indent one further than the comment you're actually replying to. Tryptofish's last comment is less indented than Carl Fredrik's because Tryptofish is not replying to Carl but rather they're replying to Rky72. We don't group replies together because it breaks the chronology of the discussion. Mkdwtalk05:48, 19 August 2016 (UTC)
I'm thinking the real problem here is that Doc James has a degree in medicine, but what he needs to sway Wikipedia is a degree in baby-talk. Any admin will proudly declare that "obvious sock is obvious" as if it were incontrovertible logic. But "obvious paid editor is obvious" is a pidgin, a bit too much English, not enough babble. The solution is that we need a baby-talk term for paid editors who violate the Terms of Service. Though I lack a formal degree in the field, I think perhaps "poo" would make a good term - a Paid Online Operative, perhaps. "Obvious poo is obvious" is the kind of crap that any established Wikipedia admin should have no trouble with, and can be used in a chant, like its parent phrase, to drown out any related or unrelated policy objections that might come up. Wnt (talk) 19:21, 20 August 2016 (UTC)
It seems to me that one of the underlying themes of the support comments in the RfC at the top of this talk page is a concern among editors that it is too easy for an editor who is acting in good faith to get blocked for outing, when in fact, it was more a situation of meaning well, trying to deal with something like COI, and misunderstanding a confusing policy based on "case-by-case" language. And, as I and others have noted, we have a problem with Streisand effects following some blocks for outing, in which attention is drawn paradoxically to the information that was supposed to be oversighted.
As much as I agree that outing, and other forms of harassment, are very bad things, it seems to me nonetheless that there has been a problem with blocks for outing being punitive as much as they are preventative. There is a difference between a user who means harm and is likely to continue the outing behavior unless blocked, in which case a preventative block is well justified, and a user who will immediately regret a mistake and stop the conduct upon realizing what they did wrong, and whose block will predictably trigger a Streisand effect. At Wikipedia talk:Harassment/Archive 8#Flowcharts, I suggested a draft version of how administrators might be more rigorous in adhering to the community norm of WP:NOTPUNITIVE when deciding whether to block for outing. I think there's merit to that.
Ultimately, I think I might start an RfC with the goal of establishing a community consensus that WP:NOTPUNITIVE applies here just as it does elsewhere, as a consensus that admins, functionaries, and arbs, need to keep in mind. For now, I just want to start a discussion here. I realize of course that NOTPUNITIVE has its own problems, in that it is rather porous and lots of blocks really are a mixture of preventative and punitive, but I'm more interested here in clarifying what the community wants with respect to outing. --Tryptofish (talk) 19:53, 23 August 2016 (UTC)
Although the RfC at Wikipedia:Village pump (proposals)#Proposal for a confidential COI mailing list will still be open for some time, it seems obvious to me that the community is strongly rejecting the idea of dealing with personal information pertaining to a COI issue via private email. Editors appear to object strongly to having administrators or functionaries act on a COI accusation on the basis of a private email. And I recollect that ArbCom has said that they do not want to deal with such emails either.
I've been thinking about what that means for the policy here, and it seems to me that some things will likely need to be changed.
The current third paragraph of outing includes these sentences:
If redacted or oversighted personally identifying material is important to the COI discussion, then it should be emailed privately to an administrator or arbitrator – but not repeated on Wikipedia: it will be sufficient to say that the editor in question has a COI and the information has been emailed to the appropriate administrative authority. Issues involving private personal information (of anyone) could also be referred by email to a member of the functionaries team.
And the last paragraph is:
Nothing in this policy prohibits the emailing of personal information about editors to individual administrators, functionaries, or arbitrators, or to the Wikimedia Foundation, when doing so is necessary to report violations of confidentiality-sensitive policies (such as conflict-of-interest or paid editing, harassment, or violations of the child-protection policy). Only the minimum information necessary should be conveyed and the minimum number of people contacted. Editors are warned, however, that the community has rejected the idea that editors should "investigate" each other. Posting such information on Wikipedia violates this policy.
It seems to me that the community is very strongly rejecting most of this. Looking at Wikipedia:Conflict of interest#Avoid outing, it is actually presented in the form of saying that one can email functionaries or ArbCom for advice about how to deal with COI, as opposed to emailing evidence to be used against another editor. It seems to me that we could replace some of the last paragraph with that. Otherwise, I think that all that is likely to remain is "Editors are warned, however, that the community has rejected the idea that editors should "investigate" each other", and maybe emailing WMF about harassment and child protection. --Tryptofish (talk) 19:52, 24 August 2016 (UTC)
Personal comment: the main problem is really the lack of a one-stop shop for dealing with undisclosed COI editors where private evidence is either necessary, or because they've denied it and now you think any further challenge would be outing. (I've had this in my experience: an editor whose user name was the same as a social media page giving their name. I thought that they were basically next-best-thing-to-out as a COI editor until they suddenly panicked and denied everything. You can't really do anything more with someone like that short of saying "sorry, I really don't believe you.") What I'd love is for a system to file AIV reports and append private evidence that could be looked at if necessary if the AIV admins don't see fit to block on the grounds of diffs. Blythwood (talk) 22:43, 24 August 2016 (UTC)
Where you said AIV, you meant COI, right? The proposal at the Village Pump now is for just such a way of emailing that kind of evidence, and it is being overwhelmingly rejected by the community. --Tryptofish (talk) 00:35, 25 August 2016 (UTC)
Right, that proposal seems to be going nowhere. As someone who's done a lot of COI work, here's what COI issue handling looks like. At WP:COIN, if it looks like COI and/or paid editing, it probably is. The usual approach is not to ask for editor sanctions at first. It's to take the COI material out of the article, get more eyes on the article, and wait to see what happens. If the COI material is persistently re-inserted by the same editor, even after reverts by other editors, and they're not making a good argument for it on Talk or COIN and won't stop, they get blocked for edit warring or WP:NOTHERE. This approach doesn't require "outing" editors, and is quite effective.
More and more, I agree with that. I increasingly think that what we need here is a clear and unambiguous outing policy that leaves little leeway for COI investigation, except for carefully considering how not to needlessly punish a good-faith editor working on COI. (I still suspect that in a few years, we will end up wanting more tools to use against bad-faith paid editing.) --Tryptofish (talk) 15:20, 3 September 2016 (UTC)
The current approach works but does not scale. There's an ongoing decrease in the number of active editors combined with an increase in attractiveness of Wikipedia as a vehicle for promotion. At some point Wikipedia will need to come to terms with this but as usual it will take a highly visible crisis to spur action (ref the Siegenthaler incident). Shock Brigade Harvester Boris (talk) 15:28, 3 September 2016 (UTC)
In one of the many very long discussions here, now at Wikipedia talk:Harassment/Archive 8#In defence of paid editing, Blythwood made what I think is a very useful suggestion, and I want to bring it up again here (adding some details of my own, to fill out the concept), to make sure that it eventually gets acted on.
The idea is to have one simple question added to the process of creating a named account, asking if the editor expects to be a paid editor. For good-faith editors (this idea is not addressed to bad-faith ones) who answer "yes", they would automatically have their user page placed in Category:Paid contributors/No listed employer and a helpful message would be placed automatically on their user talk pages, providing information about our guidelines and terms of use for disclosed paid editing.
A lot of new editors mean well, and just need to be pointed in the right track (and again, this isn't intended to work for bad-faith ones). Doing this automatically and right at the start would cut down on a lot of the need for COI enforcement and the accompanying outing issues.
At meta:2016 Community Wishlist Survey in November, it will be possible to propose development ideas to the WMF software creators, and I would like to propose this idea there.
I'd like to hear what other editors think, and also to try to devise the question that new users would be asked. It should be simple, friendly, and yes/no. I'm thinking of something like:
Do you expect to be paid or compensated for your edits to Wikipedia?
Support - I can't think of a reason not to facilitate policy-based interaction with the WP community by good-faith paid editors. - Brianhe (talk) 19:53, 18 August 2016 (UTC)
Support - but the example question suggested above is confusing. Does it mean that you shouldn't expect Wikipedia to pay you to write about yourself? How disappointing. Geogene (talk) 20:44, 18 August 2016 (UTC)
Suggestion - thanks for this! As original suggester, a good question phrase would be
"Do you intend to edit Wikipedia for business or promotional purposes?"
That would cover people who are editing to promote themselves (so not getting paid) or people promoting their idea, non-profit foundation, husband or whatever. I also suggest that once people click "yes" they are invited to give further details - my reasoning here is that I think people are likely to behave better if they feel that their edits are accountable to their employer. Blythwood (talk) 21:50, 18 August 2016 (UTC)
Thanks, all! Yes, I think "business or promotional purposes" is the much clearer question (but if anyone finds out how I can get Wikipedia to pay me, do let me know!). I might tweak that sentence to change "intend" to "expect", because intention carries a very slight implication of devious intent. I think an advantage of a follow-up question is that it could move the user out of Category:Paid contributors/No listed employer, and into a category corresponding to the employer that they name. The disadvantages are (1) it makes registration just a bit more complicated, and we should keep it simple, and (2) it also makes the software development more complicated, which might decrease the likelihood of WMF agreeing to do it. What do editors think? Do the advantages of specifying who the employer is outweigh the disadvantages? --Tryptofish (talk) 22:18, 18 August 2016 (UTC)
Note that promotional editing is prohibited, whereas paid editing is not. All editors, included paid ones, are expected to follow WP:NPOV. - Brianhe (talk) 22:25, 18 August 2016 (UTC)
Yes, I can see that fact either of two ways. One is that we don't want to give new users the false impression that promotion is OK. The other is that we will get new users who are planning to write promotional content, whether we want them to or not, and having a chance to tell them right at the start on their user talk why we have NPOV can be good education and prevent a lot of grief. --Tryptofish (talk) 22:34, 18 August 2016 (UTC)
I realise. But we can educate people on that once they've joined - what I'm getting at is that we need to cover people who aren't expecting to get paid but are lobbying for a charity or whatever. Essentially: I want to encourage people to confess to a COI upfront before they start editing (because they will start editing, whatever we do) as part of managing their expectations of what Wikipedia will allow them to do, and discouraging bad behaviour. Blythwood (talk) 22:46, 18 August 2016 (UTC)
Support. Won't help, can't hurt. ""Do you intend to edit Wikipedia for business or promotional purposes?" is better, since "paid" can mean "paid by a grant from the British Museum" or "Working on the clock, technically, in my university professor's office". Herostratus (talk) 17:18, 19 August 2016 (UTC)
On that subject. I would not lump in someone working on uploading public domain digital images in the British Museum's collection, who is being paid by the Museum or a WMF grant or whatever to do it, with a PR flack. The category then becomes too broad. This conflation of (rare but existant) virtuous paid work with PR flackery has been pursued by PR flacks to deliberately muddy the issue. Herostratus (talk) 00:33, 20 August 2016 (UTC)
I'd like to know what other editors think about having a second question for users who reply "yes" to the first question, asking them to name their employer. Myself, I've thought about it overnight, and I think I oppose a second question. Registering a new account is about a lot more than the specific issue we are discussing here, and we need to keep it simple for new users, with as few steps as possible. Also, there are a lot of technical problems with asking software to recognize answers. What will the software do if the employer name is misspelled, or when users reply "I don't know" or "Don't know" or "Not sure yet" or with two employer names? On the other hand, it would be easy for the automatic message on the user's talk page to explain how to specify an employer if the user wants to. --Tryptofish (talk) 22:17, 19 August 2016 (UTC)
I think on balance that that's fine. In most cases it will be obvious on whose behalf they're editing to be honest. Blythwood (talk) 20:26, 20 August 2016 (UTC)
The questions should be branched if that is possible.
"Are you receiving, or do you expect to receive, compensation for editing Wikipedia?" with yes and no radio buttons.
If yes is the answer: "Welcome to Wikipedia! Per the paid editing policy, you are obligated to disclose your employer (or contractor), the client, and any other affiliation that is relevant on your user page, and you should not directly relevant content, but rather propose content for others to review. Please read the WP:PAID policy. You are of course free to directly edit other content."
If no is the answer: "Do you have any connection with subjects you wish to edit in Wikipedia? (a person who is a relative, friend, or enemy, or a company you work for, or your school or alma mater?" with yes and no radio buttons.
If yes is the answer: "Welcome to Wikipedia! Please be aware of Wikipedia's guideline on conflicts of interest (COI). You should declare your connection on your userpage and you should not directly relevant content, but rather propose content for others to review. Please read the WP:COI guideline. You are of course free to directly edit other content."
If no is the answer: "Welcome to Wikipedia! Please make sure you review the Welcome message that will soon appear on your Talk page. "
I'm afraid that I very strongly oppose anything that rococo. Registration should be quick and simple, and it's about a lot more than COI. Plus, I think the WMF people will react negatively to anything that involves that many steps of coding. Most of what you propose as a response to "yes" can instead be expressed in a boilerplate talk page message, which also has the advantage of making it easy for the user to look back at it repeatedly. --Tryptofish (talk) 21:46, 20 August 2016 (UTC)
I hear that. The above algorithm is what would adequately deal with the prompting PAID and COI disclosures. But to be clear the only time someone's employer should be asked about is if they say they are paid and they are trying to figure out how and what to disclose. Otherwise there is no need to ask that and one shouldn't ask that. The broadest question relevant to COI and paid is: "Your identity is protecting by the OUTING policy, but if you have some connection to subjects you have edited (or will edit), would you please disclose the connection?" And the aim of that is just to elicit the kind of connection - employee, contractor, relative, etc. Not the identity. Jytdog (talk) 21:57, 20 August 2016 (UTC)
Thanks. Although the discussion here is far from complete, and I don't intend to make it sound as though anything has been decided, here is my impression of what the discussion is leaning towards at this time:
1. During registration, a single question with a yes/no radio button will be something like:
Do you expect to edit Wikipedia for business or promotional purposes?
Putting the PAID disclosure template on someone's page is making a claim that they are actually paid in the RW and that claim should not be made unless that fact is first affirmatively disclosed by the user. Placing it solely on the basis of that question is not valid, in my view. Template:Welcome-COI or Template:Uw-coi would be appropriate. Jytdog (talk) 22:24, 20 August 2016 (UTC)
I agree with respect to template messages, but I do not see it as a problem with a user category. I think the category is useful in keeping track of new registrants, and of course a category can always be replaced or reverted. But maybe I'm wrong? --Tryptofish (talk) 22:34, 20 August 2016 (UTC)
It is the same problem with the category in my view. We use the PAID label with care. See here and affirmation here. And the template instructions say it is for people who must comply with the TOU. In general I am all in favor of prompting more disclosure but I am not comfortable with making definite claims about people in the absence of disclosure. Part of the issue here is that there is a lack of clarity in the community about what "paid" means. Pretty much everybody agrees that it applies to 1) freelancers who edit under contract; most also include 2) someone working a PR firm; but it drops off steeply from there - many people say it doesn't apply to 3) a company employee (boss tells secretary to create a WP page, or engineer decides she wants to write about company's product) but many say it does; people are even more sharply divided if 4) the person owns the company (doesn't necessarily even draw a salary) It is that third and fourth categories where applying PAID is dicey in the community but the user might answer "yes" to the question. Messy I know. So it is safer/less controversial in the community/less potentially upsetting to the user if we go with the less definitive COI and not use the very definitive PAID unless the user discloses that they are being paid. Jytdog (talk) 22:59, 20 August 2016 (UTC)
I didn't know that. In that case, there is no need for an automated assignment of a category. And I think it is actually better to use a category worded in terms of having a notice, instead of actually having a COI. So having the software simply start the talk page and add a template would be sufficient. We ought to consider which, if either, of those templates to use, or whether a new one is needed. I notice that the existing templates imply that edits have already been made. --Tryptofish (talk) 00:56, 21 August 2016 (UTC)
@Jytdog: Actually, I just discovered that the category is only assigned for the "uw" templates, not the "welcome" one. Consequently, I think that we will need to create a new tracking category, not that this would be difficult. --Tryptofish (talk) 23:26, 25 August 2016 (UTC)
Alternatively, it would be simple when creating the new template to just incorporate the coding to put the user talk page in the existing category for the "uw" templates. However, a potential advantage of a new category is that the existing category has a huge number of users in it (really a disturbingly large number!). --Tryptofish (talk) 23:36, 25 August 2016 (UTC)
Support and thanks to Tryptofish bringing this forward. I'll just summarise why I think this is important: it's about expectations and managing them, and making filing a disclosure seem easy and the right thing to do. My experience with COI editors is that they almost never want to make a true, formal COI disclosure, and will do almost anything but fill one out, even when informally they're quite happy to admit that they work for a company in things like edit summaries and talk page comments. It's easy to see why: disclosure is confusing (what is this "talk page" I've got to put it on? Why are there all these messages shouting at me that my edit has been reverted and my article's going to be "speedy deleted"? Shouldn't I deal with that first?) and seems humiliating (you want me to admit that I work for this company? But why don't I just not? You can't tell, right?? What's going to happen to me if I don't? Do I really want to admit I did something wrong?). Often, I get the feeling that people would rather just quit their account the second they get messages telling them that they're doing something wrong, and establish another clean one without all these scary warnings (this goes for inexperienced but good-faith editathon editors, too) - oops, now they've just started sockpuppeting themselves and made it look even worse. My suggestion is that we have a radio button choice and a question like "Do you expect to edit Wikipedia for business or promotional purposes?" Then, if you click yes, you go to a form saying "Do you wish to fill out details on this now? Remember, Wikipedia does not require you to state your name". If at this point you've realised that you've made a mistake or that (say) you only want to edit about the Girl Guides because you were in that when you were a kid and loved it - you can click a button saying "I'd rather not give details now". Either way, you get a welcome on your talk page that explains COI issues, and maybe put into a category like "Users who might have COI issues in future". Why does this matter? The problem is that people come into Wikipedia with the expectation, or at least hope, that it's basically a blog, that they can edit their own company's article, and do whatever they like. The way to manage this is at the signup stage. While - yes - it's kind of fun to zap promotional articles with speedy deletion notices and dash COI editor's dreams, ultimately it's corrosive. The fact that someone is a COI editor in their day job doesn't mean that they might not be interested in becoming a constructive contributor in their spare time, or that they don't themselves have the problem of a boss who just wants a specific copyvio text on their article and doesn't want to take "they won't let me!" for an answer. What we need to communicate is what Wikipedia's values are, not how many fun ways to revert edits we have. If you get to that screen, it's easy to click "Yes" and fill out a form. Clicking "no" is the hard step, since you will know that if you're planning to make promotional edits, it's going to look very bad and telling an obvious lie that you clicked "no". People don't want to admit that they might be on Wikipedia for the wrong reason, but they'd probably rather do that than be caught in an obvious lie. Blythwood (talk) 22:28, 24 August 2016 (UTC)
Thanks for the kind words. The multi-step process at registration is pretty similar to what Jytdog and I discussed just above, and the problems with doing it that way remain the same. But I think that I can create a new template that is a modification of Template:Welcome-COI, that would automatically go on the user's talk page if they answer "yes" at the first radio button, and it could elicit all of that, but allow the user to do it in their own time, instead of forcing them to do it right away, and it would also make it easy for them to change their mind. When I get around to drafting that, I'll post about it here. In my opinion, that actually is the least WP:BITEy way to do it. --Tryptofish (talk) 00:41, 25 August 2016 (UTC)
OK. The reason why I'm leaning towards a form is that it's standing right there in front of you - it creates the expectation that you have to type something in - and yes, I do see the difficulty of complexity, which is why I think a simple "I don't want to do this now" button would be a good option to allow people to get out of there. The reason why I want a form is that editing the raw code of a talk page to copy and past in a COI template is clearly not something most COI editors are prepared to do, and in some of the most insidious cases, it will be valuable to other editors evaluating your contributions to know who you work for specifically, not just that you think you might have a COI at some point in future. As a second choice, we could have a form to fill in replacing WP:DISCLOSE that automatically fills out a COI disclosure on your talk page and any other pages you specify and signs it for you. Anyway, we can discuss this with WMF developers when/if this gets taken forward, since they are professionals at UI design and I'm not. For a message, I suggest it starts as a generic welcome but then says something like "You have said you might have a conflict of interest in future. Please remember..." We don't want it to be too strong to discourage editing for people who have clicked "yes" by mistake or because they didn't understand the implications. Blythwood (talk) 11:03, 25 August 2016 (UTC)
I agree with you about the tone of the message. And, in case it wasn't clear, we would not be asking the users to edit anything pertaining to a template. Rather, it would be a template message that the software would put automatically on the user's talk page, to be read. --Tryptofish (talk) 22:23, 25 August 2016 (UTC)
