The Signpost

News and notes

Alleged "outing" of editor's personal information leads to Wikipedia ban

Contribute  —  
Share this
By The ed17 and Andrew Lih

As part of the second major "outing" controversy to hit the English Wikipedia in less than a year, the Chelsea/Bradley Manning naming dispute was dragged into the spotlight yet again when the English Wikipedia's Arbitration Committee ruled by motion to remove the administrator tools from long-time Wikipedia contributor Phil Sandifer and to ban him from the site. The committee stated that a personal blog post by Sandifer, which extensively profiled the real-life name, location, and employers of a user involved with the Chelsea Manning renaming case, went too far—violating a formerly little-used policy corner of the English Wikipedia, "outing", which governs the release of editors' personal information.

Sandifer's ban was surprising in that it was based solely on content published outside Wikipedia; he did not link to his writing or publish any personal information on the English Wikipedia, Wikimedia sites, or associated content areas, such as mailing lists and IRCs. It was instead published solely on his personal blog, focusing on the editor Cla68.

The committee's decision to take action was split into three clauses. The first, which passed 9–2, declared that Sandifer broke policy, while the second removed his administrator tools and had slightly less support at 8–3. The third clause indefinitely banned Sandifer. It passed with the least amount of support: seven in favor, three opposed (Kirill Lokshin, David Fuchs, Carcharoth) and one abstention (Risker).

The Signpost has examined statements by arbitrators on both sides of the issue and contacted specific members for comments, in addition to collecting reactions from Wikipedians on the talk page of the Committee's noticeboard.

What has been revealed?

In his 22 October blog post, "Wikipedia Goes All-In on Transphobia", Sandifer revealed personal information about Cla68's location and occupation to make a point about what he considered a conflict of interest in the Chelsea Manning naming case. The post contrasted the associations—and potential conflicts of interests—of editors such as David Gerard, who was recently topic-banned by the Committee, to that of Cla68, who received no sanctions. (Editor's note: in keeping with the Signpost's practice in covering outing policy, we give no direct hyperlink to Sandifer's blog post.)

One paragraph in Sandifer's lengthy piece brought up Cla68's involvement in the naming dispute and potential conflict with his occupation, noting that Cla68's background was "a fact he has studiously attempted to hide" from the Internet community. When Sandifer's post was brought to the attention of the Committee, it probably raised red flags, on the basis that revealing this information could be considered "outing" Cla68 against his wishes. For his part, Cla68 told the Signpost:


Whether it actually constituted an outing by the definition of Wikipedia policy has been a point of debate. In the past, Cla68 has given his full name to a number of media organizations, such as an interview with the Register (UK), and on-wiki in a 2009 "Not the Wikipedia Weekly" episode. As for location, Cla68 has declared his basic prefecture-level geographic location on his user page since 2006. He has also given the name of the city where he once lived on more than one occasion, including in several early revisions on his userpage. However, Cla has never, on Wikipedia, named his current location.

A second claim, that he has never revealed his employer, falls into a far grayer area. Cla68 did detail his work history on at least one open forum that is easily found with an Internet search engine, and on Wikipedia he has edited on multiple occasions—perhaps inadvertently—from an IP address traceable to his employer. However, the exact details of his occupation (detailed in Sandifer's blog post) did not seem to be previously widely known, and confirming them requires either an inference or some level of off-wiki research.

Sandifer told the Signpost that his alleged outing was discerned from a careful reading of Wikipedia edits and by making inferences from them—which in a careful reading could fall under the "although references to still-existing, self-disclosed information is not considered outing" clause of the English Wikipedia's harassment policy. Still, given the context of the surrounding text, it could be taken in the opposite direction. The policy also provides guidance, though unclear in places: "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."

Previous outing

Oddly, Cla68 was himself accused of revealing the personal information of another editor earlier this year. As in this case, the information was published on an outside site—Wikipediocracy, a well-known forum that is openly critical of the English Wikipedia, where he is a "global moderator." Wikipediocracy has been a host to many personal attacks and outing efforts. For his part, Cla68 told the Signpost that Wikipediocracy moderators "quickly remove overly personal details like phone numbers, home addresses, etc. and usually move the rest to non-public forums. I myself never take part in trying to find personal details on other people. I stay out of it."

For the March story, when the Signpost asked Cla68 how his actions did not constitute outing, he stated:

When prompted with this incident and the above quote, Sandifer commented to us that "I’m more than a little worried that the real objection is the fact that I’ve been so publicly criticizing people, and that my reporting got picked up by the Guardian. Clearly this isn’t really about outing for Cla68. And the shockingly severe sanction makes me worry that it’s not really about the outing for the committee either. I dearly hope I’m wrong, of course, and that this is simply a misunderstanding."

Ensuing debate

In discussions after the case, some of the arbitrators detailed their stance in an extensive, spirited debate with community members.

Newyorkbrad, who voted for all three clauses, commented that "it is unacceptable for an administrator, or for any editor sufficiently experienced to be aware of our policies and project norms, to escalate an on-wiki disagreement by publicizing the real-life identity, employer, and geographical location of a fellow editor, as Phil Sandifer ("Phil") did in this instance." Risker, who voted to reprimand and remove Sandifer's adminship, but abstained from a full ban, elaborated on her role:


In opposition, arbitrator Kirill Lokshin noted the problem of banning a Wikipedian for aggregating existing information together as a case of outing. To the Committee's mailing list, repeated in communications to the Signpost, he was blunt: "It's untenable for us to pretend that someone's identity is private when they're openly making statements to the press under their real name." Fellow arbitrator David Fuchs, who voted against banning Sandifer, was more nuanced, saying that while Sandifer's blog post was "unnecessary, unhelpful, [and] poorly reasoned", "off-wiki conduct alone is generally not enough to provoke on-wiki sanctions."

David Gerard and Morwen, previous parties involved in the transgender individuals dispute, both vigorously disagreed with the Committee's decision. Gerard disputed the nature of the outing: "It wouldn't pass muster as material to out a user on-site, but it certainly passes muster as an exercise in applied journalism, which is what the post was. So the arbcom has now banned someone from Wikipedia for journalism about Wikipedia."

Samuel Klein, a member of the Wikimedia Foundation's board of trustees, was disappointed: "This does look like a successful breach experiment. Extraordinary and heartbreaking that it extended to a ban."

In the extended discussion about the decision, long-time Wikipedia editor Jehochman echoed the views of a number of commenters about the nature of "outing" someone with known information.

Beeblebrox, who was involved with the blocking of Cla68 in March and whom the Committee has granted oversight powers, expressed dismay: "I don't often comment on ArbCom decisions, but this one is so puzzlingly inconsistent and weird I feel compelled to comment ... This is the most schizophrenic thing this schizophrenic incarnation of ArbCom has done yet. Some of you have been consistent as individuals but as a group you have been wildly inconsistent. I can't say it has been a pleasure working for this particular committee and I sincerely hope there are is a substantial turnover in the next election."

Under the committee's interpretation of the outing policy, all external links to Sandifer's blog post on the English Wikipedia have been systematically removed and the revisions oversighted.

Editor's notes: The authors have had several minor interactions with participants on both sides over the years. The ed17 has collaborated on several English Wikipedia featured articles with Cla68.
Wikipedia Weekly's podcast #104, which Fuzheado hosts in concert with other Wikipedia editors, had hyperlinks to Phil Sandifer's blog post when discussing the block of David Gerard. The episode was released 22 October 2013. Though the podcast and its participants did not mention the case of Cla68 or his personal details, the wiki page has since been "oversighted" and a link was removed by arbitrator AGK on 7 November.
While the original Committee motion did not mention the username in question, it was revealed as Cla68 on the talk page of the Arbitration Committee's noticeboard.

In brief

+ Add a comment

Discuss this story

These comments are automatically transcluded from this article's talk page. To follow comments, add the page to your watchlist. If your comment has not appeared here, you can try purging the cache.
Sjakkalle, thanks for the correction. Tony (talk) 08:56, 9 November 2013 (UTC)[reply]

Philip Sandifer details in his latest blog post "Fuller Statement on my Wikipedia Banning" that the arbitrators have misstated and misrepresented many of their claims about him. The post does not mention the name of the outed subject - David Gerard (talk) 10:07, 9 November 2013 (UTC)[reply]

In that case, we can link it directly? Google seems not to have it yet. --Demiurge1000 (talk) 10:30, 9 November 2013 (UTC)[reply]
Google "Philip Sandifer: Writer" and you'll get the blog. Money quote:

The arbitration committee attempted to intimidate me out of reporting critically on them. They've handed out the most draconian punishment in their arsenal according to a novel new reading of a policy, and have publicly lied about my conversations with them leading up to this ban, suggesting that I refused requests on their part that never existed. They have sided with a blatantly hypocritical user who has previously spoken favourably of the importance of transparency and accountability, and who has in the past done the exact same thing I'm banned for. And they have salted the earth on Wikipedia, forbidding all linking to the "Wikipedia Goes All-In on Transphobia" article - an article that, as noted, was widely linked to by serious-minded blogs and newssites and that formed the basis of an article in the Guardian.

Given all of this, an obvious question forms: If this isn't about punishing me for criticizing the arbitration committee and censoring that criticism on Wikipedia itself, what is it about?

- David Gerard (talk) 10:43, 9 November 2013 (UTC)[reply]

How does this PROD on 7 November 2013 at Phillip Sandifer fit into the story? -- Michael Bednarek (talk) 10:55, 9 November 2013 (UTC)[reply]

Presumably Jehochman idly typed in Phil's name, found an article of questionable noteworthiness and PRODed it - David Gerard (talk) 11:05, 9 November 2013 (UTC)[reply]
Sure. That's plausible. --Stephan Schulz (talk) 13:13, 9 November 2013 (UTC)[reply]

What makes the authors of this article think this is a "major" controversy? There are 114,078 active editors on Wikipedia -- a handful of folks complaining does not a controversy make. NE Ent 13:34, 9 November 2013 (UTC)[reply]

A couple of comments:

  • I was not contacted by the Signpost, but I'm happy to answer questions on this matter if asked, either here or at the arbitration noticeboard talk page. I can only speak for myself, not for my colleagues, but I may be able to clear up any points of confusion.
  • As one of the arbitrators opposing these actions, I made these comments at the arbitration noticeboard talk page. I'm linking them here for the benefit of readers of this Signpost article.
  • The point about the external links to Sandifer's blog post on the English Wikipedia being removed needs clarification. It is possible to track where such links are being used by using Special:LinkSearch. An example is the results for www.philipsandifer.com. Monitoring those results over the course of a few days formed part of my judgement that this matter was not making a big impact on Wikipedia in the sense that it was not being linked to from many places (though it may have been a topic of off-wiki discussion).
  • It was suggested during the ArbCom deliberations (on the mailing list) that the links in question would need to be removed. I argued against this, saying that heavy-handed removal of such links can have a chilling effect on discussion. The links have now been removed, but as far as I can tell were not removed until after the desysop and site ban motions were published on 6 November.
  • We received an e-mail from an uninvolved editor on 7 November pointing out one of the links (on User talk:Montanabw). You can see the subsequent discussion between Montanabw and AGK (an arbitrator) following the question left here.
  • Another place that I'm aware of where the link was used (and later removed, by AGK on 7 November) was at Wikipedia:WikipediaWeekly/Episode104. For the record, User:Fuzheado (who organises the WikipediaWeekly podcasts) is Andrew Lih and the co-author of this Signpost article.
  • Two further places where the links occurred and were removed (not by an arbitrator, but as far as I can tell independently, both on 7 November by administrator Bilby) were at two user pages in retirement statements: User:JJARichardson and User:Stealth Munchkin. Update: I missed one, it was also added and removed from User:Vexorian. 00:42, 10 November 2013 (UTC)[reply]

As far as I know, those were the only places the blog post in question was linked from. I would not normally go into this much detail, but I think it is important that it is a matter of public record which pages these links were on, how many links there were, and what actions were taken in respect to them, by whom, and when. If any Signpost writers, editors or readers have any questions on this or other matters relating to this story, I would be happy to comment further if needed. Carcharoth (talk) 14:42, 9 November 2013 (UTC)[reply]

I'm not sure if it is particularly important, but just to clarify my role in the above, I removed the links based on my understanding of the policies governing harassment, outing and external links, on the assumption that if the blog post was deemed serious enough to warrant a ban and desysop, then it would fall under those policies (given that I also did my own digging into the issue, and I considered events around previous occasions of linking to alleged outing). This was completely independent of ArbCom. I've been keeping a very close eye on the discussions regarding the decision before and since, in case a consensus emerges that the blog post ins't a problem - if such does happen I'll immediately revise my actions. My involvement (given that I've had nothing to do with the Chelsea Manning dispute) came as a result of looking into the issue after the ArbCom announcement, and stumbling across what I felt was a problematic link and quote from the blog post.
I hope that is of some use in clarifying things. :) - Bilby (talk) 12:25, 10 November 2013 (UTC)[reply]

Quick question for the Signpost people. Is there any reason why this is being referred to as an alleged outing? That it was an outing/doxing seems to me beyond dispute. ("My reasoning for outing [user] was and is simple: it's in the public interest" [Blog, 6 Nov 2013] and "Regardless, yeah, I doxed a dude" {Blog, 9 Nov 2013].)  Roger Davies talk 15:00, 9 November 2013 (UTC)[reply]

Hi Roger, the ArbCom decision did not use the word "outing", preferring harassment, personal details, et al. Therefore, with such an amorphous concept of what constitutes "outing" and the debatable nature of whether the information constituted outing (in the Wikipedia policy sense; WP:OUTING does not tackle off-wiki content), we felt it was important to not to take a firm stance. Ed [talk] [majestic titan] 15:41, 9 November 2013 (UTC)[reply]
Hi Ed. Thanks for the comments. First, outing is part of the harassment policy, and needs to be seen in that perspective. Our policy's definition of outing is not vague at all and linking the legal name (which is more or less disclosed) to an employer name (which hasn't been) and a precise workplace (which is also undisclosed) is clearly covered. Publication on an external site is covered by the another section of the policy. The combination protects editors from being outed or worse off-wiki (someone lately announced a scheme to publish people's social security numbers) and that stuff ending up on wiki, because it was published on the internet. So, sensibly, policy says that if the material has been disclosed or linked to on-wiki by the potential outee, then it's not outing. However, in this instance, the publisher has unambiguously acknowledged that they outed and doxed someone so I'm not seeing where the problem in saying so lies. This is a very important point: the defence is purely that it was in the public interest to out/dox.  Roger Davies talk 16:16, 9 November 2013 (UTC)[reply]
Hi Roger - I feel like I may wind up regretting asking this question, but would you consider this to represent a similar violation of WP:OUTING? Kevin Gorman (talk) 19:39, 9 November 2013 (UTC)[reply]

I don't see how this is even worth so much attention. It's a very straightforward case, and I think the reaction to it hardly qualifies as a controversy. There are always plenty of people getting banned for violating policy, and in many cases there is a serious controversy to report. Everyking (talk) 16:56, 9 November 2013 (UTC)[reply]

I don't know anyone involved in this case, and I was not even aware of it, but this sounds like an important issue, because, even though Sandifer's disclosures off-wiki were (1) bad behaviour (and he should have been willing, at least, to remove the most private ones about location, etc.) and (2) he should be held to a high standard because he is an admin, I think that the remedy applied was too severe. It would seem that a topic ban would be a more appropriate remedy. We don't have such a thriving editor community that we can kick productive editors off the site entirely when their bad behavior pertains to one topic area. It seems like a bad precedent to set. -- Ssilvers (talk) 17:25, 9 November 2013 (UTC)[reply]
Over the past four years, Phil has made about 290 edits. Of these, 193 were about Chelsea Manning; so that's less than 25/year about everything else. In contrast, the editor that Phil brags about outing "in the public interest" has written 30+ featured articles.  Roger Davies talk 19:09, 9 November 2013 (UTC)[reply]
So far in 2013, Cla68 (the subject of this news story) has made precisely 118 content edits to wikipedia - David Gerard (talk) 19:34, 9 November 2013 (UTC)[reply]
I think a more significant metric would be how many malicious outings occurred on sites that Phil is a global moderator on, vs sites that Cla68 is a global moderator on. Kevin Gorman (talk) 19:39, 9 November 2013 (UTC)[reply]
Interesting stat, David Gerard, surprisingly low, but I think fewer main space edits is more a sign that an editor has been on Wikipedia a long time and does a lot of admin work. I was more surprised that this outing information wasn't already present on his user page as Cla68 does reveal more details about himself there than I would ever be comfortable with. But, from what I've read, ARBCOM did a search to see if this specific information had been shared and found it hadn't been. Liz Read! Talk! 21:44, 9 November 2013 (UTC)[reply]
This is a factually incorrect claim on their part; I've seen the edits, they're still public. I sent Ed a list of links for this article. And the arbcom has also claimed Phil used private information, but that's just casual defamation (which is an unfortunate thing for the arbcom to be getting back into when challenged) because a fairly trivial web search shows it to have been completely unnecessary. (I was particularly struck by the publication from his employer that showed on page one of Google for me.) - David Gerard (talk) 22:10, 9 November 2013 (UTC)[reply]
This Arbcom seems especially capricious and prone to questionable decisions. I too hope for a massive turnover, followed up by all the hideously bad decisions they've made this year being thrown out. Adam Cuerden (talk) 23:49, 9 November 2013 (UTC)[reply]

I don't think CLA was "outed" at all, actually, although I must confess taking a certain perverse satisfaction in watching a POV warrior go down in flames. Carrite (talk) 01:01, 10 November 2013 (UTC)[reply]

As much as I don't approve of doxing at all (even in this case, where I agree with 99% of what else Phil said), I note that senior WMF staff have shared a link to the blog article in question on Facebook. Using this same logic, should they also be banned or sanctioned for outing the user on a third-party website? Lankiveil (speak to me) 02:17, 10 November 2013 (UTC).[reply]

Perhaps they can use that made-up offence of "conduct unbecoming a staff member" again. 'Cos the arbcom actually trying to get Foundation staff fired for daring to defy them was such a good idea last time David Gerard (talk) 08:32, 10 November 2013 (UTC)[reply]

Well, that is a harsh sentence, given that Phil only posted information off-wiki. Epicgenius(give him tiradecheck out damage) 15:05, 11 November 2013 (UTC)[reply]

While I was surprised by the ban, I think Phil was wrong to post location details, and said so in his blog comments last week. – SJ + 09:15, 13 November 2013 (UTC)[reply]


"Challenging Arbcom's authority"

Yeah, pretty much. We've seen evidence of Arbcom blocking before just because they didn't like that someone didn't kowtow to them. Also, Cla's identity was a secret? After he published it all over the place and linked to it multiple times on wiki? If you say so.

Meanwhile, i'll just leave these links here to these news articles (Link 1) (Link 2). You know, if anyone wants to read them and all. Public information and all that. SilverserenC 19:48, 10 November 2013 (UTC)[reply]

  • Which means that, if someone can be blocked for doing something like this on their personal blog, not anywhere connected to Wikipedia, then that means we can ban anyone else involved in such outing on sites like Wikipediocracy, correct? So I should expect you to get to banning any contributor to Wikipediocracy that is directly involved in outing a Wikipedia editor on that site, right? Good to know. I'll make sure to cite the Arbcom ruling in any future instances of off-wiki outing. It should be possible to get quite a few people banned with that. SilverserenC 00:40, 13 November 2013 (UTC)[reply]

Phil's post-facto self-censorship of his blog post is quite duplicitous and is clearly meant to mislead people reading about his ban. Shii (tock) 18:51, 13 November 2013 (UTC)[reply]

@Shii:, you don't see that as him acknowledging that he made a mistake? – SJ + 01:48, 17 November 2013 (UTC)[reply]
I guess I would if he sounded more apologetic. Shii (tock) 06:10, 17 November 2013 (UTC)[reply]



       

The Signpost · written by many · served by Sinepost V0.9 · 🄯 CC-BY-SA 4.0