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.
Risks/benefits of posting employer/client/affiliation?
As User:Opabinia regalis noted here and others have noted elsewhere, a discussion about the potential risks and benefits of allowing editors to post "company/client/affiliation" on behalf of paid editors would be useful.
Setting this up to get that rolling, with various categories to help keep things clear, based on what various folks have already said on this issue here and elsewhere.
This is all with respect to allowing posting of employer/client/affiliation on behalf of undisclosed paid editors (UPE) and is all about potential risks and benefits. If we actually do this, unexpected harms and benefits could emerge that nobody anticipated.
People wrongly accused of being paid editors, or who fear being wrongly accused, do not produce encyclopaedic content for people to read. Thryduulf (talk) 16:07, 17 February 2017 (UTC)
The focus on paid and suspected editors means non-encyclopaedic contributions from editors not suspected of being paid are overlooked. Thryduulf (talk) 16:18, 17 February 2017 (UTC)
the loss of reputation and income is likely to occur only if the person is a confirmed persistent violator, and is a positive result of policy enforcement, not a risk. DGG ( talk ) 05:21, 11 February 2017 (UTC)
In risk/benefit analysis, risk is a neutrally appraised cost to an actor (a party). You're adding a judgment about that outcome here, which is IMO best accomplished in a separate phase when we weigh the risks against one another. It also helps us to understand that we are in control of some risks ourselves, e.g. we could choose actions that have a greater or lesser impact on actors' reputations. Brianhe (talk) 17:12, 22 February 2017 (UTC)
might be used as bludgeon to win content disputes; editors unfortunately "cry COI" too often in content disputes. Jytdog (talk) 17:19, 31 January 2017 (UTC)
if identification of employer or affiliation is accurate and there was no unpaid editing, editor has been OUTED with no benefit to anyone and only harm to the editor Jytdog (talk) 17:19, 31 January 2017 (UTC)
"Suspected" should not be included, only confirmed undisclosed paid editors should be named: risk of error and therefore legal action and defamation if unconfirmed — Iadmc♫talk 21:23, 31 January 2017 (UTC)
Editors who make edits about entities/people they have a COI with but who are not paid (directly or indirectly) to make them could become suspected of UPE and outed. e.g. Students making edits about their university, family members making edits about their spouses/parents/etc, employees in roles wholly unrelated to marketing making edits about their employers. Outing can then lead to loss of job, harassment, stalking, violence, and any other "usual" risk of outing. ~ Rob13Talk11:58, 4 April 2017 (UTC)
Increased risk of real-life harassment, stalking, and violence as a result of being wrongly accused. (Important to be clear how severe the "harm" can be here) ~ Rob13Talk11:53, 4 April 2017 (UTC)
Risks to other Wikimedia projects
Editors on en.wp who stop editing because they have been or fear being wrongly accused are likely to stop editing other projects as well. Thryduulf (talk) 16:16, 17 February 2017 (UTC)
Risks to people, institutions, etc not named above
the possibility of misidentification is why we need actual procedures , as the alternative is our current way of doing things, which amounts to guessing, and is even less accurate and thus more risky. DGG ( talk ) 05:25, 11 February 2017 (UTC)
Whether content is promotional (whether or not it is disguised) is independent of whether the contributor is (suspected of being) paid.
Whether content is promotional (whether or not it is disguised) is independent of whether a paid contributor disclosed their being paid. Thryduulf (talk) 16:21, 17 February 2017 (UTC)
I'm not sure if these sections are really for threaded discussion, but since you have asked me questions below, I will also respond to you here. Please really engage with what I wrote in #Break 3. Is changing the numerical value of the average temperature in a given location really self-evident as promotional? The world is entering into a new era, of alternative facts. --Tryptofish (talk) 00:33, 18 February 2017 (UTC)
If done correctly and sensitively, it can provide an opportunity for a falsely suspected editor to be able to set the record straight. --Tryptofish (talk) 22:58, 31 January 2017 (UTC)
How? It is very difficult, at best, to prove a negative. Even if they were to out themselves as having a relationship with a competitor to the organisation suspected (company A) that proves nothing either way about whether they are receiving money from company A. Outing them doesn't help. Thryduulf (talk) 16:25, 17 February 2017 (UTC)
By giving the accused editor an opportunity to see the website that is the basis of the accusation against them. And please note that I said "if done correctly and sensitively", not as willy-nilly outing. Upon seeing that information, the accused can submit rebutting information privately, if that is better. --Tryptofish (talk) 00:37, 18 February 2017 (UTC)
Benefits to editors who post
Benefits to other editors
allows other editors to see disclosure that should have been there from the beginning, so that edits can be evaluated in light of the COI. Same reason that COI disclosures are required in academic publishing. Jytdog (talk) 17:15, 31 January 2017 (UTC)
allows editors concerned with undisclosed paid editing over time, to see patterns of company activity and identify articles that are especially subject to COI editing, to better maintain article quality there. Jytdog (talk) 17:15, 31 January 2017 (UTC)
If any Wikimedia project loses the perception that we provide useful information, that instead of having gone back to the early Pokemon days, we have become an outlet for "alternative facts", our legitimacy will be lost, and Wikimedia will go the way of daisy-wheel printers, to be replaced by something else. No, I am not exaggerating. --Tryptofish (talk) 23:02, 31 January 2017 (UTC)
Further information: #Break 3 and Alternative facts. By decreasing the amount of bogus information on Wikipedia. I think bogus information on Wikipedia leads to negative public opinion of Wikipedia, and I would hope that you agree about that. --Tryptofish (talk) 00:41, 18 February 2017 (UTC)
Benefits to people, institutions, etc not named above
provides groundwork on which WMF Legal might then choose to take direct action with regard to paid editing companies Jytdog (talk) 17:16, 31 January 2017 (UTC)
if list is made and is reasonably public, would allow people/entities who want to hire paid editors to see companies with a history of UPE. Could come to function like a Beall's list of Predatory open access publishing (a word of caution, Beall discontinued his list, apparently due to legal troubles or other harassment but the reason he discontinued it is not clear. ) Jytdog (talk) 17:32, 31 January 2017 (UTC)
This is a fascinating list, but more for what it doesn't include than what it does. You've started a list of the "risks" of posting others' personal information without their consent, and didn't include a heading for the risk to the editor whose personal information is exposed. It's pretty clear the thinking here began with risks to the people already inside the Wikipedian "empathy circle" - mostly the editors who want to be permitted to post this stuff - and therefore has overlooked the most serious risks, which are the possibility of harming someone who's done nothing more than ignore a terms-of-use document on a website (a behavior that is ubiquitous online), and the possibility that you might be wrong and irreversibly expose some entirely innocent editor's personal information. These are much more serious risks than someone getting temporarily blocked while a problem is investigated and resolved. Maybe this is a consequence of my real-life work, but I spend a ton of my time being wrong about stuff. It is part of the nature of research to develop a model that seems to fit the available data well and then discover some new information that reveals a flaw. I am very accustomed to assuming I might be wrong about things, even when I feel subjectively confident in the evidence. Unfortunately I don't see that kind of mental habit reflected in most of the conversations surrounding the "outing" of paid editors.
Importantly, I want to amplify what I mentioned above in my previous post on this page. It is really weird that so much of the conversation about managing paid editing is taking place on the talk page of the harassment policy and is focused on this one specific, controversial, and time-consuming means of approaching the problem. Personally, I'll be much more inclined to believe the arguments that this is somehow necessary if I start seeing the people who claim they need to be posting personal information instead investing serious effort in developing other approaches to the problem that are more consistent with our usual working practices. And by that I mean the traditional methods of "comment on the content, not the contributor". Notability standards, new-article quality control, sourcing. See where the WMF stands on adapting ORES to spammy edits as well as vandalism. Consider amending deletion processes to facilitate the removal of promotional articles. If necessary, people can then present evidence of specific sets of circumstances in which those more conservative preferred approaches are not sufficient. Opabinia regalis (talk) 20:43, 31 January 2017 (UTC)
Wouldn't "the editor whose personal information is exposed" be included in the "Risks to confirmed UPE" and "Risks to suspected UPE" groups (even if innocent)? Mojoworker (talk) 21:05, 31 January 2017 (UTC)
Not to totally speak on Opabinia's behalf, but I think she's talking about the framing. Calling them UPE and not Editors puts them out of the empathy circle. --Kyohyi (talk) 21:08, 31 January 2017 (UTC)
If - and it is a huge if - you want to read Opabina as being rhetorical in writing that there is no consideration to the editor whose information is exposed, that kind of rhetoric is unacceptable to me for a sitting arb in a highly charged discussion. I would like to think that Opabina simply read too quickly before they posted. Jytdog (talk) 21:32, 31 January 2017 (UTC)
User:Opabinia regalis please redact your misrepresentation of what I wrote. I provided two sections for risks to editors whose information would be posted (and in one case I am not even sure the community would ever arrive at consensus to post if it was only suspected). Two. Not zero. Jytdog (talk) 21:27, 31 January 2017 (UTC)
Kyohyi is correct. I'm objecting to both the framing and the consequences of that framing, namely that "the person whose personal information is exposed" is not necessarily the same as "the suspected/confirmed paid editor". That's why I posted in the "general comments" section rather than just adding to the headings or lists of risks. There are many other possible risks to the owner of the personal information that aren't accounted for in the construction of the discussion - for example, it may be that information belonging to the wrong Joe Bloggs gets posted, or that the investigator falls for a joe job and posts accurate information about an editor's identity based on inaccurate beliefs about their activities, or that posting information about Joe also reveals information about Jane, or.... etc. Opabinia regalis (talk) 21:34, 31 January 2017 (UTC)
User:Opabinia regalis The cases you mention about making a mistake or falling for a joejob fall squarely under "suspected UPE". And again, this is about posting, with regard to a specific user account that is a suspected or confirmed of UPE, the "employer, client or affiliation". There are only four "fields" under discussion, 1) account username; 2) the three pieces of information named in the ToU - employer/client/affiliation. There is not one section for suspected/confirmed - there are separate sections for suspected and confirmed, and I did it that way exactly to deal with the kinds of things you mention. At this point you appear to be almost willfully trying to distort the discussion and I ask you to a) stop doing that, and b) redact your misrepresentation. Jytdog (talk) 21:47, 31 January 2017 (UTC)
This is a real question, not a rhetorical one: do you really not see that "the person whose personal information is exposed" - the person, the real-life human being - is a different class being asked to bear different risks than "the suspected UPE", the (IMO rather dismissive) term you chose to refer to the Wikipedia identity you suspect of editing for pay? The difference lies in the possibility that something went awry in your "investigation".
In principle, doing a risk analysis on this problem is a good idea. In practice, I think it's unlikely to produce any clear conclusions, for a number of reasons. The simplest is the issue in the preceding paragraph, the need to correctly identify all of the stakeholders and their associated risks. More subtle is the problem of framing: how we (readers of and participants in this discussion) evaluate these risks largely depends on the extent to which we identify with the hypothetical members of each group. This is what I'm getting at with the "empathy circle" comment - the discussion is framed so that readers of it empathize with the hypothetical editors who might get blocked for posting their concerns, but not with the hypothetical real-life person who found themselves in unpleasant personal circumstances, took on some crappy online freelance work, and now has that decision publicly associated with their real name on a major website. I'm not saying you're deliberately constructing it that way; it's just the expected consequence of a discussion framework written by someone with a known strong opinion on one side of a divided issue. The broadest reason this type of risk analysis is unlikely to be beneficial is simply that it's incomplete: as I mentioned above, there is a very noticeable trend in discussions of "paid editing" to focus narrowly on the perceived value of posting allegedly paid editors' personal information, to the exclusion of managing paid-for content using other community mechanisms. A risk analysis that doesn't compare the risks and benefits of such posts with the risks and benefits of other approaches is insufficient as a basis for drawing any conclusions about how the community can best manage paid editing. Opabinia regalis (talk) 23:09, 31 January 2017 (UTC)
(edit conflict) Jytdog, you are surely a grown-up dog, and you don't really need to have anything redacted. Opabinia, no it's not weird to discuss the issue here (and does it really surprise you to see something weird around here anyway?), because the outing policy is the policy that will be used against any editor who makes a mistake in how they try to deal with undisclosed paid editing. --Tryptofish (talk) 23:12, 31 January 2017 (UTC)
Trypto and Opabina. The misrepresentation is very serious to me. I am very disturbed by the utter lack of good faith, and the assumption of stupidity and thoughtlessness on my part in what Opabina is writing here. The assumptions make actual dialogue impossible. I am not writing this lightly. Jytdog (talk) 23:32, 31 January 2017 (UTC)
Does anybody ever intentionally misrepresent things? It doesn't matter why. It is a blatant misrepresentation that what I wrote above ignores the risks of OUTING to people suspected of UPE who aren't doing that or leaves no space for discussion of that. It is right here for pete's sake. User:Tryptofish how does your point here differ from that in any significant way, reading mine with even a modicum of good faith? Jytdog (talk) 00:09, 1 February 2017 (UTC)
Trypto, that's exactly the problem I'm trying to push back on. You can make lots of mistakes in trying to deal with paid editing - say, by not making a good case for deletion when you take a fluff-filled article to AfD. The only way this policy is relevant is if you want to post the personal information of someone you believe to be editing for pay. That should be a very small part of the broader activity of "dealing with paid editing", relevant only in the most extreme situations, yet the discussion is disproportionately dominated by that topic.
Jytdog, I didn't say you are stupid or thoughtless. I said that I disagree with what appear to be the underlying assumptions motivating this discussion, and I wish you would reconsider those assumptions in more depth. I think the people advocating management of paid editing in part by loosening the outing policy are underestimating the risk of error and making assumptions about the likely failure modes that are inconsistent with my experience dealing with harassment and outing victims. The fact that I disagree with you doesn't mean I think you're stupid or thoughtless or acting in bad faith. Like I said, I'm wrong all the time, so I may well be wrong now - but calling my thoughts on the subject "misrepresentation" doesn't convince me I am. Opabinia regalis (talk) 05:31, 1 February 2017 (UTC)
Only a small part of dealing with undisclosed paid editing involves the use of off Wikipedia material. Most has and always will entirely involve on WP evidence. What we are doing now; however, is not enough when we are dealing with companies that primarily do undisclosed paid editing. There are a number of organizations out there, that number is growing, and we have been failing to adapt to the changing times. Doc James (talk · contribs · email) 02:57, 2 February 2017 (UTC)
I should clarify that when I referred to an editor making a mistake, I was thinking more specifically about an editor who posts something that might seem to you to be personal information and seem to me to simply be an advertisement for paid editing. I think that editors who are trying hard to act in good faith can get in trouble over the current version of the outing policy because the policy is unclear in so many ways, and it may be another kind of mistake to sanction such an editor. For that matter, "case-by-case" strikes me as a colossal mistake. So obviously there are indeed many different kinds of mistakes. --Tryptofish (talk) 23:44, 1 February 2017 (UTC)
And I will add that Opabina's repetition of the claim that "there are other ways - go work on them" expresses a lack of awareness of the hours and hours of work that people put in all the time - the ongoing efforts -- to address paid editing/COI issues, operating very much within the limits of OUTING, and gives no room for them to even begin to understand why there has been this continual effort to come up with better ways to navigate the tension between the values of integrity and privacy, much less a space from which to productively contribute to the discussion instead of trying to derail it and shut it down. It leaves no where to go. Jytdog (talk) 23:49, 31 January 2017 (UTC)
While I have no doubt that people focused on COI issues work hard, I mentioned the "comment on content, not contributors" maxim because I was specifically referring to approaches that focus on content, not COI investigations. The issue I am interested in is the implicit premise that evaluating bad content on its (lack of) merits is inadequate to solve the part of the "paid editing problem" that matters to our readers: the part where we get bad (misleading, promotional, poorly sourced, etc.) content as a result.
User:Opabinia regalis I realized I did not directly reply to the following: } This is a real question, not a rhetorical one: do you really not see that "the person whose personal information is exposed" - the person, the real-life human being - is a different class being asked to bear different risks than "the suspected UPE", the (IMO rather dismissive) term you chose to refer to the Wikipedia identity you suspect of editing for pay? The difference lies in the possibility that something went awry in your "investigation". Of course I am aware that every person operating a user account is a real person. My entire approach to dealing with COI (when I could do that, before my TBAN) was dialogue-based and the mistake I made that got my blocked was a departure from that. What I am trying to communicate to you is the following - your responses to my posting have been based entirely on your assumption that I don't respect other editors and their privacy. Your question is the exact same as "Does your mother know you beat your wife?" The only answer is: I don't beat my wife. And of course I take everyone's privacy seriously. Why do you assume I don't? And why is the effort to even discuss this so threatening that you keep trying to shut it down? Those are real questions back to you - and they are real questions from me to you. Jytdog (talk) 02:36, 1 February 2017 (UTC)
Thanks. However, you're wrong about what assumptions I may be making. You argued that the category I said was omitted had in fact been included. I disagree, and asked whether you saw the distinction I was drawing. As for shutting down discussion, why would you assume that was my intention, instead of the simpler hypothesis that I just disagree with you? Considering this exact subject is currently being discussed on at least three separate pages, whoever is trying to shut down discussion is apparently doing a piss-poor job of it anyway ;) Opabinia regalis (talk) 05:31, 1 February 2017 (UTC)
Let's try to actually discuss the objection to the section header "suspected UPE editors". In the discussion about amending OUTING to deal with UPE, we need some way to talk about useraccounts that people begin investigating in good faith, to make sure folks commenting are clear that this is not something that applies to just anybody. This is also written with the assumption there wlll be an actual process in place to authorize posting "employer/client/affiliation" (I cannot imagine there is a chance of consensus without that) for user accounts of suspected or confirmed undisclosed paid editors (who are, yes, people). Keeping all that in mind, User:Opabinia regalis what would you prefer that we call this subsection so that we can actually talk about it? Jytdog (talk) 01:01, 1 February 2017 (UTC)
My point is more subtle than "we should change the section header", or I would've just said that ;) What I am encouraging you to do is question the assumptions that led to your original choice of wording and structure. Changing the wording doesn't change the assumptions. Opabinia regalis (talk) 05:31, 1 February 2017 (UTC)
Tryptofish has "framed" this is in difficult way, as a "disagreement". This is not a disagreement about where to go with policy, this is a disagreement about what is going on in my mind. That is not a "disagreement" in any sense we use that word in WP. You have declared me incompetent to participate this discussion - too caught up in fervor about COI matters to be trusted to think and write clearly about OUTING and needing to have what I write here deconstructed, the conversation about the issues stopped dead in their tracks, and the flaws in the place I am coming from unpacked, as it is just "fascinating" to see a warped mind at work. And Opabina, you actually stand by this behavior and defend it as appropriate, anywhere in WP. I cannot wrap my head around that. Jytdog (talk) 16:25, 1 February 2017 (UTC)
All I mean by "disagreement" is that OR takes a very strict position on "content not editor", whereas you are (correctly, I believe) pointing out that it isn't as simple as that – and that you consider OR's language to be a personal attack on you, whereas she (correctly, I believe) is trying to explain that The fact that I disagree with you doesn't mean I think you're stupid or thoughtless or acting in bad faith. My hope is that the discussion will turn away from editor motivations, which is very much of a distraction, and towards finding a consensus about what does or does not need to be posted about undisclosed paid editing accusations. And as for "framing", I can try to explain what I mean by going back to your earlier comment about my edit at #Risks to suspected UPE. I did not mean it as a criticism of what you had previously put there, and I do recognize that there is some redundancy between what we each said. But I gave it a different framing, in the sense of a difference in affect. You referred to no benefit to anyone and only harm to the editor. Quite correct, and said in a businesslike way, and nothing wrong with that. I, on the other hand, put an emphasis on the more emotional aspects of that harm, which I think is useful in drawing attention to something that will be important for editors to pay close attention to as these discussions continue, and also in demonstrating to editors who feel as OR does, that we understand their concerns. --Tryptofish (talk) 23:44, 1 February 2017 (UTC)
Following the implementation of this: [2], it seems to me that this new, more precise, language supersedes the "under discussion" line Posting links to other accounts on other websites is allowable on a case-by-case basis. That being the case, I propose that the case-by-case sentence should be deleted from the outing section. --Tryptofish (talk) 20:43, 16 March 2017 (UTC)
Just providing a permalink to the RfC and its close at WT:COI -- it is here. It was on the basis of that, that the diff above was done. Jytdog (talk) 21:32, 16 March 2017 (UTC)
And I'm just adding a note that the closure is being reconsidered, so editors may want to wait until the dust clears before answering my question. --Tryptofish (talk) 22:53, 16 March 2017 (UTC)
Yes, we have one specific kind of link that the community (tentatively, depends on the revised close) has endorsed. We don't have consensus for linking in any other situation, do we? So there is no reason to imply that there are additional kinds of links that have consensus, but they are determined "case-by-case", whatever that means. If you review the talk archives here, there have been a lot of discussions about how the "case-by-case" language is so imprecise as to be useless in a policy to which editors will look for guidance. And after all, it still caries that "under discussion" tag. --Tryptofish (talk) 22:18, 17 March 2017 (UTC)
Much of this policy is controversial such as "This applies to the personal information of both editors and non-editors." Started a RfC to address this sentence. Doc James (talk · contribs · email) 23:51, 21 March 2017 (UTC)
I'll note that a recent edit has combined the "exceptions" language into a single paragraph, such that the "case-by-case" sentence becomes an introductory topic sentence for the paragraph. This is a very good improvement, and partially resolves my original concerns. At the same time, I continue to feel that "case-by-case" is such a vague term that it is unhelpful to have here at all. So I am going to revise my original question as follows: can anyone explain to me what "cases" are allowed, that are not now described in that paragraph? If not, what purpose does the sentence serve? If so, why not spell it out instead? --Tryptofish (talk) 00:17, 22 March 2017 (UTC)
Change Posting links to other accounts on other websites is allowable on a case-by-case basis. to Posting links to other accounts on other websites is allowable under specific conditions. --Tryptofish (talk) 00:33, 22 March 2017 (UTC)
