Wikipedia:Wikipedia Signpost/2016-07-04/Op-ed
Two policies in conflict?
Over the past week the English Wikipedia has seen a reactivation of the controversy surrounding the intersection of two key policies: those for paid-contribution disclosure and outing. It began on 27 June 2016 with the blocking by arbitrator GorillaWarfare of Jytdog, with “outing” being the justification. Jytdog is a long-standing editor on medical topics who also works in the area of conflict of interest. The aftermath has raised a number of issues that the community is struggling to address including: who gets to decide policy, what is and is not “outing”, and whether it is permissible for Wikipedia volunteers to discuss and link to publicly posted job offers for Wikipedia PR work and the public professional profiles of writers undertaking such work.
Who gets to decide?
The first issue raised is who gets to decide policy. Many editors view the decision of GorillaWarfare to indefinitely block Jytdog and ArbCom not unblocking as of publication as a misapplication of our outing policy, or borderline at best. The discussion of what the policy actually is and where we draw the line between "outing" and "not outing" has been closed by Mike V, a checkuser, with claims that these discussions are outside the community's remit.
The functionaries have gained a head start in the discussion through their use of private channels. Two of them—DoRD and Thryduulf—tried to remove important wording from the harassment policy: "posting links to other accounts on other websites is allowable on a case-by-case basis", apparently with the intention to leave the impression that posting links to other accounts is never allowed. The clause in question has been in the policy in one form or another since early 2015.
The removal was without community consensus, and it appears that the majority of the functionaries have decided that their interpretation of WP:OUTING trumps that of everyone else.
On the question of who gets to decide, my objective is to stand up for the community. When I was on the board of the WMF, I pushed hard for broad community involvement in determining our long-term strategy; I saw it as inappropriate for a small group of individuals to make decisions behind closed doors. The same applies in this case: I believe transparency and community involvement is vital in determining policy as it is in determining strategy.
Legal perspective
Some editors have argued that a very strict interpretation of WP:OUTING is a requirement coming from the WMF. However, the Foundation's legal department has published a statement on this question, indicating that this is not the case:
I've been asked to clarify how this discussion fits with the Wikimedia privacy policy. It is not a violation of the Wikimedia privacy policy for editors to post links to public information about other editors. The privacy policy applies to how the Foundation collects and handles personal information, as well as users who have access to nonpublic information. The underlying principle in our privacy policy is that respecting and protecting anonymity and pseudonymity is essential for encouraging free expression. Posting links to public information on other sites is a question of balancing this underlying principle, not a direct violation of the privacy policy. It's up to community consensus here to decide when the harassment policy should allow editors to reasonably link to public information on other sites. Thanks, Stephen LaPorte (WMF) 20:50, 30 June 2016 (UTC)
It appears people are taking a policy that applies to non-public information and attempting to force it onto all information contained in personal and corporate accounts outside Wikipedia.
Our purpose: stating the obvious
Jimmy Wales summarized the purpose of our efforts well: "Imagine a world in which every single person on the planet is given free access to the sum of all human knowledge. That's what we're doing."
Our purpose is not to create an anonymous online utopia but to create sources of knowledge, and our rules need to be created and maintained to achieve this end. To the degree that anonymity achieves our goals, it should be supported—but this does not mean it should be supported absolutely and in all situations.
Just address content and ignore the terms of use
The Wikimedia Foundation’s terms of use, and the English Wikipedia's paid-contribution disclosure policy derived from them, disallow undisclosed paid editing. Some wish that we exclusively concentrate on content and in essence ignore the terms of use. While this might seem perfectly reasonable in theory, we simply don't have the volunteers interested in following around obviously paid editors and addressing the concerns they create.
Our readers expect our content to be written by people independent of the topic in question. By making it exceedingly hard to address undisclosed paid editing we put our shared brand at significant risk. By allowing promotional or unbalanced material to persist in Wikipedia, we put our readers at risk. This is evidenced for example by the Wifione arbitration case, as reported by Newsweek. The promotional editing in that case negatively affected thousands of students' lives. There is no point in having terms of use if we do not enforce them. They need to be more than simply a public relations measure.
Others have stated that we should never need to use links to justify concerns of undisclosed paid editing. They feel that we can simply express the concerns on a person's talk page without providing evidence; but what we end up with is the casting of aspersions. Our essay on the topic in fact states, “An editor must not accuse another of misbehavior without evidence ... If accusations must be made, they should be raised, with evidence, on the user-talk page.” Those accused also deserve the right to defend or justify themselves openly and transparently.
Specific cases
We currently have an open RfC regarding a statement I posted to WP:COIN in September 2015, and whether we feel this type of comment should be allowed or disallowed. In part, it concerns a job offer for Wikipedia editing publicly posted on Elance, a website frequented by those intending to commission and those willing to undertake paid online PR work:
Checking Elance
[...] Here, we have someone who is buying an article on Anthony LaPine. They have already bought an article on HipLink, created by a sock account, User:Juliecameo3, that is already blocked. Doc James (talk · contribs · email) 04:36, 4 September 2015 (UTC)
Most of the functionaries who have weighed in appear to believe that posting a link to a public Wikipedia-related Elance job offer should not be allowed and is a potentially blockable offense. This is despite the fact that the company behind Elance is willing to collaborate with us and doing this sort of work is an effective method of enforcing our terms of use with respect to undisclosed paid promotional editing.
By extension are we going to say that linking to pubmed, a well-known database of biomedical abstracts, is disallowed because such links could have the real names, employment, or location of Wikipedia editors which those editors may not have disclosed previously on our sites? This link, for example, contains comments from Wikipedians with personal details. Not only will such an extreme stance of “no linking to any accounts outside of Wikipedia” make enforcement of our terms of use nearly impossible, it will interfere with our policy on verifiability.
Public and private domains
The privacy of volunteer contributors is a vitally important value. But public-relations work is, as the very name implies, public, not private. The LinkedIn profiles of marketing managers as well as the job and service advertisements on freelance PR writers' sites like Elance are posted there with the explicit intent to catch the eyes of the world.
Many editors may conclude that it is wrong that the entire world should be free to review, discuss and use that information, except for the unpaid volunteers whose work is directly affected and not infrequently compromised by the efforts of PR writers—people whose primary allegiance is not to the Wikimedia project but to those who pay them.
James Heilman
MD, CCFP(EM), Wikipedian
Discuss this story
Very thoughtful and well written, Wikipedia needs to do more to stop undisclosed payed editing. The actions against Jytdog are entirely out of line with the aim of writing an unbiased encyclopaedia. Carl Fredrik 💌 📧 19:22, 4 July 2016 (UTC)[reply]
I know that, being an op-ed, this article is intended to advocate for the particular point of view held by its author, but I'd like to bring up a few points that Doc James either doesn't mention or that he presents differently here than others have done in the other discussions going on about this.
A fluffernutter is a sandwich! (talk) 20:04, 4 July 2016 (UTC)[reply]
The overarching point I've been trying to make in these discussions is that when we talk about linking to an offsite account, we need also to consider what type of content and details have been posted by the person to that offsite place, because by providing the link you're encouraging people to click it and read what the account behind it has done/said; it's not enough to say "well, it's just a resume, people distribute resumes all the time" or "oh, it's just a reddit account, those are anonymous" and ignore that a resume usually includes significantly more personal (and sometimes private) detail than most people would be comfortable sharing on a website where all edits are permanently and publicly stored (and where it's being linked to in a threatening or retaliatory context), and a reddit account, exactly because it's pseydonymous, often includes comments or posts on topics that would be incredibly TMI or personal if they were connected to a specific individual (for instance, by someone on Wikipedia saying "here's this editor's profile on reddit [link]"). A fluffernutter is a sandwich! (talk) 20:44, 4 July 2016 (UTC)[reply]
I believe we need to do more to protect those like the thousands of students who were harmed when our content was corrupted in this case of paid editing[1]
Jytdog was in good faith attempting to address undisclosed paid promotional editing such as this. He make a mistake in that the link he added to the person's talk page was simply not needed as the case was so obvious. That this minor infraction and one he has given assurances he will not repeat has resulted in an indef block is unfortunate for our project. It also sends a chill through those working in a difficult area who already get plenty of threats due to the work they do.
During the Rorschach ink blot controversy a fellow editor who disagreed with my position dug through my imagine uploads and wrote a letter of complaint to my licensing body claiming that because one of my images was of a patients who had a congenital condition that is occasionally associated with intellectual disability that the person in question could not have given informed consent. The patient was in fact a highly educated professional. I forwarded this to arbcom around the time it happened and more important got a lawyer. The editor was never blocked. That was actual harassment.
Doc James (talk · contribs · email) 21:11, 4 July 2016 (UTC)[reply]
I remember when Jytdog said I was wrong for suggesting that a supposedly neutral editor's claim that a fake literature review of the effect of neonicotinoids on bees commissioned and paid for by Bayer Cropscience by researchers-for-hire was a high-quality source suggested bias and reeked of the epitome of paid advocacy, disclosed or otherwise. That still seems like a mistake to me, but on balance I thought he generally upheld the reliable source criteria well, and would like to petition for his unblocking. EllenCT (talk) 21:57, 4 July 2016 (UTC)[reply]
There will always be a conflict between the concept of producing a high-quality encyclopedia and the concept of permitting anonymous editing. This was less of a problem in past years, because initially the goal was not really to produce a high-quality encyclopedia, but a rough-and-ready encyclopedia that would have at least some minimal usefulness. But regardless of our original modest intentions, the public--even parts of the public such as the Library of Congress devoted to archiving proven reliable information about authors--are using us as a source. Journalist do, courts do. Studies have shown that in at least some field this is not ridiculous, because our quality in those fields compares with more traditionally reliable sources. In other fields it does not. These fields are ones that attract inexperienced editors, and also those fields which attract people having a conflict of interest. While we never will be able in our system to deal adequately with advocates of a cause (unless they are greatly outnumbered, the increasing and most currently significant problem is the financially motivated editing of articles on commercial and non-commercial organizations, and the people associated with them. Those of my colleague who have said otherwise are probably not familiar with the current inflow of such articles. I think there is a purpose in a non-anoymous open content free encyclopedia as a complement to WP, but there is no plausible way WP itself is likely to develop into this, nor would I even advocate that, for there is also a need for the sort of encyclopedia that we do have here.
But I think it has been shown both by experience and by our current practice that we need to make some compromises with the concept of anonymity in order to maintain the quality that is expected of us. We do have our terms of use, even if we lack mechanisms for enforcing them. We do have ways of communicating private information to those who are authorized to deal with it,and we do have ways of taking this into account in dealing with particularly troublesome editors. The actual question is whether we ought to compromise it further. My experience from years of patrolling new pages and drafts and articles for deletion is that we need to do two additional things.One,which I think nobody disagrees with, is to find a better way to communicate to well intentioned new editors what is expected of them. The difficulty is to deal with those not so well-intentioned. We can in fact deal with them, once we figure out who they are--that's the difficulty that brings us to this discussion. The current method pretty much relies upon (occasionally) people making obvious and transparent errors that reveal their status and intentions, and (much more often) guesswork. It's easy to say we should judge by the article, but the flood of promotional articles in some areas is so great that even disinterested volunteers often copy the style, thinking that it's what we want, and we have a unfortunate record of finding false positives and discouraging or driving away potentially good editors. There are paid editors who boast outside WP about their success in evading the terms of use and avoiding detection--some of this is probably mere puffery, and they remain here because of our lack of ability to question those producing what are likely to be paid articles. It's been claimed there are no cases where its needed--such a claim can have no basis, because we only know what we have actually detected.
I am not clear just what rules we should have, and I therefore have not made any specific proposals. It would be easier to find workable rules that would not overly offend the anonymity-absolutists if we did not regard this as a taboo subject where any questioning of the rules is denounced as subversive of our values. (It should be obvious from discussions elsewhere that I am writing in a personal capacity, and that few of my colleagues at arb com or functionaries see the issue the way I do--I certainly welcome Fluffernutter's ability to see the need for at least some changes. And to respond to a question elsewhere, while the matter is unresolved I will enforce the rules in the conventional manner.) DGG ( talk ) 22:08, 4 July 2016 (UTC)[reply]
A few thoughts:
WP:CORP has said (for years) that you can write an article about your business, org, or product if you can scrape together two (2) newspaper articles about it, at least one of which must not be your own local neighborhood newspaper. (As a side note, I'm amazed how often established editors(!) complain about that very minor restriction, and insist that their favorite school/restaurant/business/non-profit/academic organization must get an article even though the only publication that ever wrote anything about it is the newspaper owned by their next-door neighbor.)
But if we really wanted to reduce the flood of COI marketing, we would go in the opposite direction, and say that if you want anything more than a barebones list entry on a List of non-profits in My City or List of windows manufacturers (or whatever), then you need to have coverage amounting to a minimum of a thousand words, from at least four unrelated publishers, and spanning multiple years. And then we could start something like {{subst:BLPprod}} project for such subjects, and specify that if the article doesn't actually cite sources that comply with the minimum requirements within a month or two, then we'll delete it (without prejudice to re-creation in a compliant form, of course). I think that a system like this would discourage the promotional COI stuff. WhatamIdoing (talk) 18:18, 6 July 2016 (UTC)[reply]
You're probably right that not all small-town or neighborhood newspapers pay more attention to the people they know than to people they don't. However, I can't think of a single newspaper where that's not been true in my (more than trivial) experience. Perhaps it's just the small American papers that have this very human quality? Or maybe there's a reason that we've had editors declare those papers to be "reliable" (in the sense of getting their facts straight) for years, but also "indiscriminate" (in the sense of not being a good indicator of "attention by the world at large", to quote the nutshell at WP:N, precisely because they do tend engage in neighborly promotion and civic boosterism). WhatamIdoing (talk) 05:54, 7 July 2016 (UTC)[reply]
While writing for the Wikimedia Research Newsletter, which is published first on meta, but then republished by Signpost, on occasion I have might have identified a number of researchers/educators, connecting the dots between, for example, a teacher who has described in their academic article that "students in my class have edited articles X and Y", their course wiki page, and the course creator's (i.e. said teacher's) Wikipedia userpage, even through said userpage may not even exist. In my review I would not hesitate to clearly state that "John Smith, whose course page was at... and who seems to edit as User:XXX....", even through said user page mage not mention their real name, link to their course page, nor would their research article mention their user name. I have never thought this may be outing, as it seems to me that by engaging in certain actions like teaching a course on Wikipedia and publishing an article about it people are abandoning their anonymity, by pretty much saying in a public forum intended for dissemination that "I created this Wikipedia article", which in turns means for anyone with basic wiki skills "this is my Wikipedia username". I do think that similar "connections" are habitually made by other reviewers, as well as on blogs by Wiki Education Foundation and such (ex. [3] mentions a real name of a user who does not use it on her userpage on Wikipedia, even trough she links to said article). In [4], we again see the real name of a user who does not reveal their name of their userpage, through they link to their course page where said name is present, through that page was not created by them, but by a Wiki Ed helper. Isn't Wiki Ed "outing" teachers? Could a reviewer like myself out somebody by reviewing their paper and mentioning their user account (or should I just link their wiki course pages, and let people see who created it or is listed there as the instructor...? Seems stupid, really). And how what Wiki Ed / I have been doing is different from what is discussed here? --Piotr Konieczny aka Prokonsul Piotrus| reply here 18:53, 6 July 2016 (UTC)[reply]
All of the above just being my random thoughts. But at the end of the day, the penalty has to fit the crime. A paid editor should be blocked and indeffed from editing wikipedia. But even a paid editor does not deserve to be outed or harassed offline. Montanabw(talk) 19:59, 8 July 2016 (UTC)[reply]
Slippery slope argument.
Come on, User:Doc James. You know better than to use a slippery slope argument, don't you? "By extension are we going to say that linking to pubmed, a well-known database of biomedical abstracts, is disallowed because such links could have the real names, employment, or location of Wikipedia editors which those editors may not have disclosed previously on our sites?" Please stop torpedoing your own reputation by continuing to defend and associate with this bullying. --Elvey(t•c) 09:04, 23 July 2016 (UTC)[reply]