Wikipedia:Wikipedia Signpost/Single/2015-10-21

Last month marked the first anniversary of Wikipedia's involvement with the GamerGate (GG) controversy and its seething throngs of partisans. For those who remain as yet unaware of this confusingly -gate-suffixed brouhaha, it can be summarized thus: a social-media-based slug-fest between two groups:

• The GamerGaters/Pro-GamerGaters, who described themselves as an ethics-oriented consumer movement of video game enthusiasts.
• The anti-GamerGaters, who characterized the GamerGaters as a monolithic gang of misogynists.

The depiction of GamerGate in the mainstream media has generally aligned with the anti-side's assessment, which the pro-side has taken as confirmation of their suspicions regarding the unethical bias of the mainstream media. The disagreement has been raging for more than a year and seems intractable. Surprisingly, Wikipedia's article on the topic sees little if any involvement from editors affiliated with Wikipedia's Video Game WikiProject (WP:VG). Instead, the bulk of the edits have come from a small cadre of prolific editors and a long tail cavalcade of fly-by-night accounts that pop up and vanish—or even re-animate, golem-like, after years of inactivity—to support one side or the other.

In this special report, we look back at the development of the GamerGate controversy article from its earliest appearance to the present. The emphasis is on the difficulties of editing in controversial and topically overlapping areas, and on the ways that editors with strongly divergent perspectives can work together or at least alongside one another effectively, despite their differences. No Wikipedia article is ever considered to be in its final form, and this particular work in progress remains an object of considerable attention as new and related events in the larger (non-Wikipedia) world fan the flames of partisanship and prompt new edits. This retrospective should in no way be understood as an endorsement of the current form of the article.

GamerGate is generally acknowledged to have begun, or at least reached critical mass, when Eron Gjoni posted a 9,425-word blog post containing serious allegations regarding his former girlfriend, independent game developer Zoë Quinn. It took less than three weeks for the first manifestations of one of the most vigilante-abetted public breakups in modern memory^{[nb 1]} to slouch its way like a rough beast into our editing grounds. The general sense of reluctance to host an article on this topic was palpable in the early days of community discussion, and the associated WP:VG community discussion was heading slowly toward a consensus of "wait and see" when the inevitable happened: on 5 September 2014, Mckaysalisbury created an already trending CamelCase redlink and the GamerGate article was born.

As of publication time, the article has been the subject of one RfC, one RfM, 25 AN/AN/I filings, three arbitration requests, one Arbitration Committee case, four ARCA requests, and 48 arbitration enforcement requests. Much about this case presents a mixture of the good and bad: on the one hand, Wikipedia's "GG controversy" article has probably received more attention from mainstream reliable sources than any other WP:VG article; on the other hand, much of this attention is critical of Wikipedia.

It's probably fair to say that GG as a topic has attracted a large number of new editors to Wikipedia, but in the same breath it's probably fair to say that many of them came here to do battle and right great wrongs. As a controversy that falls along gendered lines, GG's relation to women in tech reflects and magnifies Wikipedia's own shortcomings in this area, but with the GamerGate and Gender Gap Task Force ArbCom cases behind us we can at least hold onto the fact that the issues are starting to receive the measure of the attention they deserve, and that they've played a role in the ripening of this important part of the discourse.

But is GG really an unmitigated disaster for Wikipedia? In speaking with editors active at the "GG controversy" article, we certainly don't hear glowing reports of collaboration and camaraderie, and the 47 archived talkpages pay sad tribute to this fact. Still, let's take a quick look back to see how we got from there to the present.

The GG article grew rapidly during its first two months, with more than a thousand edits each in (most of) September and (all of) October 2014. As eager to draw first blood as ever, the Wikipedia critic site Wikipediocracy posted a GG-related Wikipedian-outing article on September 8, which soon found its way to AN/I. Community fabric was further tested when GG editors Ryulong and Loganmac set up a tilting ground at AN/I in a pair of filings aimed at one another.

The "GG" article became the "GG (controversy)" article and then the "GG controversy" article in quick succession, and in late September the Gamergate ant made its way onto the main page, prompting entomologists across the region to give silent fist pumps. By early-to-mid-October, Jimbo Wales received his first GG-related talk page posting and the first GG-related mediation and arbitration requests were filed. Swedish online distributor GamersGate made an appearance on the main page late on October 14 and piqued enough curiosity to warrant a listing at DYKSTATS. GG associates TFYC got a main-page appearance shortly afterward.

By late October, the controversy had metastasized into an RfC, a second arbitration request, and a notorious list of alleged GamerGaters was making the rounds on the administrator's noticeboard (AN). Editors at AN spent October hammering out the first set of community sanctions (now enshrined at WP:GS/GG) and rounded out the month with its first two GG topic bans. The blistering 1000-edits-per-month pace slackened somewhat in November as editorial disagreements over content increased and positions hardened. The battle-lines were drawn and ArbCom finally consented to hear what the participants at "GG controversy" had to say for themselves.

December and January 2015 saw considerably fewer edits than the previous three months, as involved editors expended all of their efforts writing novellas for the arbitrators' collective amusement. Back at the article, page protections continued to stack up, and a large number of general sanctions were handed out.

Further removed from the locus of contention, Jimbo was questioned over his laissez-faire attitude toward a notorious competing list of alleged anti-GamerGaters making the rounds at GG's very own Wikia, and Wales was called on to settle a dispute between a "Gamergate controversy" editor and Slate writer David Auerbach, whose writing about Wikipedia would later be covered in a series of articles for the Signpost. Off-wiki, Reddit users established the WikiInAction and WikipediaInAction subreddits to track perceived injustices against GamerGate on Wikipedia.

The GG ArbCom decision was handed down at the tail-end of January, sanctioning several prominent editors at the GG article for violations of Wikipedia's behavioral policies. The decision was highly anticipated by Wikipedians and non-Wikipedians alike—so highly anticipated that a passel of reliable sources pre-empted it with their own versions of how ArbCom was sure to decide. In an attempt at damage control, ArbCom took the unusual step of issuing a statement that was soon followed up by an explanatory post at Wikimedia's official blog.

When the final decision was rendered nearly a week after the first prognostic reports of the decision had been published, it turned out that the reliable sources were not entirely correct on all of their specifics, but with truth serving as handmaiden to verifiability at Wikipedia, a disingenuously RS-sourced article entitled "ArbitrationGate controversy" was soon created to tweak our collective nose and then deleted to smooth our collective brow.

February saw an increase in edits as second-wave "Gamergate controversy" editors of both persuasions signed up to carry on the good fight now that a number of the previous champions had been unceremoniously curb-sided. The first few arbitration enforcement (AE) requests were successful in obtaining sanctions. In mid-March, AE imposed a "500 edit/30 day minimum account qualification" on the GG controversy article, which slowed editing by new editors to a comparative crawl, and Wikipedia's GG-specific watchdog-watching group, SeaLionsOfWikipedia, was established to track and maintain lists of Wikipedians adjudged to hold public or closeted pro-GG sympathies.

Slow editing speeds continued through April and May, although they experienced a slight rise for third quarter, perhaps related to the publication of a small number of off-Wikipedia retrospective articles reminding readers of the issues that had originally led to GG in the third quarter 2014. And that brings us up to the present date in the middle of the fourth quarter. The rate of editing now hovers around 200 edits per month—considerably lower than this same time last year, but still high enough to be a hurdle for editors who may not be interested in devoting large portions of time catching up to speed on the article.

Wikipedia ledes are intended to summarize the contents of articles, so by tracing the changes in the lede we can get an idea of the changes in the body. Due to significant levels of revert-warring and campaign-like efforts to alter the lede, this report focuses specifically on the first sentence of the lede, where "GamerGate" is defined. Definitions have ranged in length from 54 to 506 characters, and the tone and tenor have drifted in either and both directions throughout the past year. Below are a set of three graphs depicting the history of this definitional sentence's many alterations from 5 September 2014 to 5 September 2015:

• The first graph shows how many lede-sentence edits were made each day during this first year, and is intended to examine the partisan character of edits at periods of increased and decreased editing.
• The second graph is a timeline portraying the degree of dynamism in the tone of the lede sentence.
• The third graph depicts the range in lengths of the lede sentence that the article has gone through in the last year. These latter two graphs are intended to explore the issue of stability in article content.

The graph displays the number of times the lede sentence was edited per day, and the nature of the edits. Anti-GG edits are here defined as edits that introduce words such as "misogyny", "sexism", "harassment", "astroturf", etc. to the lede. GG-friendly edits, by contrast, insert words such as "ethics", "movement", "journalism", etc. Edits identified as "GG-neutral" represent alterations to the lede that either contain nothing of partisan significance to either side (e.g. edits that correct punctuation and spelling) or that introduce or remove language of partisan significance to both sides of the dispute.^{[nb 3]}

Large numbers of edits with essentially evenly matched colors generally represent edit- or quasi-edit wars. The lede sentence has played host to at least 15 edit wars in the past year. The most significant issues leading to edit-warring concern whether or not GG is a movement—the orthodox GG position is "yes", and the orthodox Anti-GG position is "no"—and whether concerns over journalistic ethics or misogynistic motivations are the driving forces of the campaign (GamerGaters claim ethics, and Anti-GamerGaters claim misogyny). Other more minor issues include disagreements over whether or not GG can be properly described as a criminal or terrorist organization (the reader is here left to guess the orthodox positions).

In this graph and the one below, we examine the durability of the ledes. The dark-blue bar represents one year's time (5 September 2014 – 5 September 2015). Vertical red lines divide the timeline into months, and the vertical white lines represent periods of editing activity related to the lede sentence. The multicolored line above the dark-blue line corresponds to the tone of the lede sentence. Segments colored teal contain only negative words like "misogyny", "sexism", "harassment", "astroturf". Segments colored fuchsia contain only words that GG supporters endorse (e.g. "ethics", "movement", "journalism"). Yellow-colored segments contain both kinds of words and olive-colored segments contain neither kind of word. The resulting graph seems to show that a GG-friendly (or GG-ambivalent) tone seems to have been more commonly accepted during the early periods of the article's construction, and that this tone has been generally dropped since the start of November 2014. The graph below demonstrates the durability of the exact wording of the lede.

Graphs are unavailable due to technical issues. There is more info on Phabricator and on MediaWiki.org.

Here we see a depiction of variations in the length of the lede sentence over time. In comparing this graph to the timeline above it, a general trend can be observed: the longer the lede, the less likely that a single ideological stance will dominate. Particularly long ledes representing only one position tend to reflect particularly partisan approaches to the topic, and are often characterized by "piling on" language. Such ledes tend to have very short lifetimes as other editors quickly revert or otherwise modify them. Shorter ledes, however, tend to be more stable even when they contain language that conflicts with the position taken by some editors. For controversial topics such as GG, this is likely due to editorial adherence to the "due weight" policy.

Anyone interested in participating at the GG controversy article has a lot of work in store to catch up to speed: the article talk page and its 47 archives clock in at nearly 10 million bytes. The participants have strong personalities and little tolerance for those with apparent agendas or whose suggestions repeat previous suggestions buried in the talk page archives. In part, this is understandable. In a situation where all aspects of editorial discretion must be the result of local consensus, new participants whose suggestions challenge the current consensus can be perceived as a threat to the stability of the article. This is doubly true when new participants arrive in numbers.

The result is perhaps inevitable: the main participants are those who are thoroughly steeped in the culture, who have actively participated for the longest, and who possess an intimate understanding of the history of the article and the social turmoil underlying the article's topic. But this set of "GG controversy"-article experts are often quite far from ideological alignment, and persistent tensions have boiled over more than once. At present, at least three of the top 10 article editors have been topic- or site-banned as a result of their participation at the article.

We discussed the article with some of its most frequent early contributors; In anticipation of problems arising from contacting and publishing responses from prominent banned editors, we contacted a member of the Arbitration Committee, who clarified that neither the printing of answers nor the responses of the banned individuals would constitute violations of their sanctions. Responses are reported here exactly as written except for potential BLP issues, and questions have been interpolated in some cases to allow for the addressing of common themes. Efforts were made to interview others from the late-2014 editing period, but contact information was lacking in several cases, and in other cases we either received no response to our request for an interview or knock-back. We decided early in the interviewing process to focus primarily on editors who had been active in the later 2014 editing period, and to avoid contacting sanctioned users in any way on-Wiki.

When, where, and how did you first hear of GamerGate?

• Mark Bernstein – August or September 2014, through Stacey Mason who studies games at the University of California, Santa Cruz.
• Masem – I may have read on some of the initial complaints that Quinn had received in the few days after the Gjoni post, but I absolutely remember actually seeing Phil Fish's Polytron Twitter get hacked and reading on that the next day of being connected to the whole situation.
• The Devil's Advocate – I have a difficult time recalling where I first learned of it. The first time I caught wind of the controversy was back in August when [feminist media critic] Anita Sarkeesian got a death threat on Twitter that contained her home address. Having spent over a year monitoring her article it quickly came up there, but it was brought up on Wikipediocracy at essentially the same time in connection with the Zoe Quinn controversy, and I also had a habit of Googling for coverage whenever Sarkeesian's latest video dropped.
• Ryulong – I had heard rumblings of it though social media when it first started but I didn't really know what was going on until I saw the mess on Wikipedia.

In 2 or 3 words maximum, how would you describe GamerGate? In your view is the Wikipedia article on GamerGate essentially neutral and accurate?

• Mark Bernstein – 2–3 words: juvenile criminal conspiracy
  The Wikipedia article is not entirely satisfactory, but on the whole it reflects the consensus of reliable sources.
• Masem – "A complicated situation". I would argue that the article is accurate (in that we cover the history of the situation reasonably close to how events unfolded) but not neutral (in how that history is approached).
• The Devil's Advocate – A1: Charlie Foxtrot.
  A2: HAHAHAHAHAHAHAHAHAHA!!!!!! Okay, in all seriousness, there is a fundamental conflict with how Wikipedia's policies approach a subject and what is required to have a truly neutral and accurate description of GamerGate. Given widespread sourcing bias and general lack of media competence in covering niche controversies, a false view of GamerGate may always be the majority view and thus get more space than I think it deserves. The most we can hope for with the article is to have it comply with Wikipedia's policies and guidelines. Just looking at the lede it is clear to me the article falls far short of that standard.
• Ryulong – "Entitled reactionary clusterfuck" is as simple as it can get. The Wikipedia article is as good as it's going to get really. Gamergate supporters are going to be like the Scientologists, climate change deniers, and LaRouche supporters that came to Wikipedia before them, never satisfied that the article is "neutral" enough to their liking.

Had Wikipedia's article on the topic improved, degenerated, or remained about the same during your period of activity at the article? Has it improved, degenerated, or remained about the same since your last edit there?

• Mark Bernstein – The article has grown considerably, and is somewhat improved. The use of Wikipedia to harass Gamergate victims has abated although attempts persist.
• Masem – It has gotten degenerated since the article started, not helped by the actual situation of GG get more nastier as time progressed. Strictly in terms of the scope of WP, it has gotten more difficult to talk on a neutral basis due to a combination of issues: the topic has attracted editors that feel very strongly against the GG situation (particularly after the ArbCom decision and the outside attention that got); and the steps that had to be taken to remove sockpuppets and similar new accounts/IPs (the 500 edits/30 days editing restriction for example) had also caught a few false positives of well-intended editors looking to bring other viewpoints in but with too few edits to participate. It has also not helped that the reliable sources have done a poor job in covering the story in a neutral or non-opinionated manner as the events have progressed.
• The Devil's Advocate – I think when I was active there I made a good effort at moving it towards a more accurate and neutral take on the issue despite a good deal of resistance, especially when I dared to include a single well-sourced paragraph saying supporters of GamerGate had been harassed as well.
  The article has become significantly worse since I was topic-banned. Unfortunately, admins aren't really doing their job to restrain POV-pushing by editors who have a hostile view of GamerGate.
• Ryulong – The article certainly isn't awash with baseless accusations against the people victimized by Gamergate as was most certainly the problem when everything started. It was decent when I stopped contributing and it's better now than it was then.

Mark Bernstein, you joined Wikipedia as an editor all the way back in September 2006, but your level of editing has increased exponentially since 2013. The number of your edits from 2015, for example, is close to three times that of your number of edits from 2013 and 2014 which in turn are more about four times your rate of contribution in 2012. To put it differently, if we extrapolate a comparable rate of editing for you through the end of the year then your edits from 2015 will be nearly twice as many as all of your edits from the previous 9 years. Part of this probably reflects the fact that you are editing at one of the more contentious and fraught regions of the encyclopedia. In the interest of transparency we should note that you have received a few blocks during this time (5 blocks of varying duration – 3 of which were reversed or shortened), and you have been brought before ARBCOM/AE/ANI/AN/3RR a dozen or so times since your last block. But sanctions from these venues were never applied and you have now been editing successfully and uninterruptedly for nearly 6 months. Can you tell us how you have avoided some of the pitfalls that other editors operating in this arena ran into? Have you spent more time reading up on the rules before running into them? Have your persecutors' efforts backfired by fatiguing the administration and were you given greater latitude because your efforts were more clearly well-intentioned?

• Mark Bernstein – Wikipedia's disciplinary processes are designed to protect it against childish vandals and isolated cranks. Wikipedia was and remains unprepared for Gamergate's coordinated assault — an assault ultimately assisted by eminent Wikipedians and which could command the support of a legion of zombie and brigaded accounts. One change in recent months is that it is now permissible to say what we have long known to be true.
  When Gamergate demanded that five editors they disliked be banned, Wikipedia acquiesced. Within days, Gamergate returned with a fresh enemies list. I wasn't one of their Five Horsemen Of Wiki Bias, but I think it's safe to say that securing my removal or extorting my silence has been a daily priority since the Arbcom decision.
  There's no symmetry here. Gamergate relentlessly sought to discuss the sex lives of their intended victims at Wikipedia; putting an end to that is entirely in keeping with Wikipedia policy. Gamergate relentlessly sought to smear Wikipedians whose opinions Gamergate found inconvenient, spreading filthy sexual slanders, demanding they lose their jobs, threatening them on-wiki and off. The only escape was to let Gamergate have its way.

WP:VG Newsletter – You have posted at your blog several comments that cast Wikipedia in a very negative light. The Guardian notably based one of their articles on your claim that Wikipedia had "purged [feminists] en bloc from the encyclopedia", and that ArbCom would allow "GamerGaters [] to rewrite their own page (and Zoe Quinn's, Brianna Wu's, Anita Sarkeesian's, etc.)". Has your view of Wikipedia changed since you posted those comments? Do you feel that Wikipedia's structure and culture has changed on its own, that you have contributed to bringing about such a change, or that it has remained the same but that you have learned more about it? If it's better than your initial blog posts suggest then how can Wikipedia better go about eliminating the concerns of outsiders (i.e. those that Wikipedians consider to be the readership) in matters like this?

• Mark Bernstein – Though my writing was quoted by The Guardian (as well as by writers in Gawker, The Mary Sue, Neues Deutschland, der Standard, Vice, Think Progress, de Volkskrant, Social Text, and others), that article was reported by Alex Hern and the staff of The Guardian. I cannot take credit for their hard work.
  What I wrote in [my blog posts] "Infamous", "Thoughtless," "Careless," and "Reckless" was correct: Gamergate launched a long-planned Arbcom case to secure the ban of five feminist editors, and the preliminary decision did ban them. (The final decision made a token gesture of only admonishing TheRedPenOfDoom; his ban was secured a little while later.) What surprised me (and, I believe, also surprised the arbitrators) was the courage a new group of editors showed in continuing to defend Wikipedia and its policies. Their success was costly, but thus far they have proven adequate to the task.
  The key effect of the public outrage over the infamous Gamergate decision was its demonstration that Gamergate could not conquer Wikipedia. If Wikipedia were to accede to Gamergate's demands and once more began to serve as a platform for harassing female software developers, public outrage would force Wikipedia to uphold its policies. Nonetheless, Gamergate's supporters at Wikipedia have persisted in their fervent — and futile — efforts, exacting a tremendous toll on its volunteers and its reputation.
  Despite many urgent calls, neither administrators nor Arbcom have been willing to address the effects of rampant harassment of Wikipedia volunteers or to find a solution to Gamergate's continuing assault. That Wikipedia is open to manipulation by other groups, similarly organized but with greater resources, goes without saying; I am aware of no realistic effort to address or remedy the problem.

As a researcher and expert in hypertext, would you say it's accurate to describe Wikipedia as a work of hypertext? Have the events arising from GamerGate surprised you? Inherent to crowd-sourcing is the tension between the extremes of a no-holds-barred anonymous anarchic society and a nanny/police-state society where real-world-linked accounts are required and all contributions are monitored for heterodoxy by teams of censors. It seems inevitable that loud protests will be heard from either or even both of the extremes whenever an organization adopting this content-growth model takes a step in either direction. How does one gage where to draw the line and should the line be continually in flux to match the vagaries of societal norms?

• Mark Bernstein – would you say it's accurate to describe Wikipedia as a work of hypertext? – Absolutely. That's how I came to be program chair of Wikisym, after all.
  Have the events arising from GamerGate surprised you? – Gamergate has surprised, astonished, and deeply dismayed me. When I was in graduate school, nearly half my colleagues were women: we weren't quite there, but we were getting close. Throughout my career, nearly half of my colleagues in my particular corner of software research have been women – fewer than half if you count noses, though on balance women held more of the most influential positions. We weren't quite there, but we were getting close.
  Gamergate's crusade to expel women from the software industry shows that we're not even close; we've gone back to the 19th century.
  Inherent to crowd-sourcing is the tension between the extremes – You draw a false dichotomy here in assuming that opposition to criminal efforts to use Wikipedia to punish women for daring to pursue their profession must necessarily require nanny-state censors, and you assume that the appropriate response to harassment is to seek 50% less harassment.
  What Gamergate made clear – Gamergate's true innovation – is that Wikipedia is nearly defenseless against an organized group who work together to advance their aims. Wikipedia's governance mechanisms are designed for protection against children and isolated cranks; faced with a concerted attack, Wikipedia would have been helpless had a handful of editors not stepped in to uphold Wikipedia policy. When Gamergate insisted that those editors be banned, Wikipedia meekly complied; fortunately for Wikipedia, others stepped into their place. Wikipedia has not even been able to muster the courage to approve a token resolution against sexual harassment, or the decency to express sympathy with people against whom Wikipedia has been used as a weapon.

Masem, you are one of a small handful of WP:VG editors engaged in this. Do you feel stranded or out of your comfort zone? How can the WP:VG sub-community do more to make life at that article easier (less partisan, more pragmatic and consensus-based, etc.)? Do you have any ideas concerning what Wikipedia can do to better dissuade POV-pushing and encourage neutral third parties to enter a fray like we see at the GamerGate article?

• Masem – The one thing that has taken a while for the GG situation to come to but is now apparent is that the VG industry has come to realize that GG is a product of the overall community (gamers, industry, journalists) that has been lingering for a long time, and having these issues pushed into the mainstream spotlight due to GG coverage by the major press outlets. Despite all the bad stuff that has happened, it is a moment of enlightenment that was needed for the video game industry, and we can already see the effects this has had on the industry in terms of addressing sexism, diversity, and other issues. I don't think the resolution is done yet, but from the WP:VG side of things, we should be treating the GG article as a keystone moment in the VG industry and trying to write it from that stance. We know a lot of bad stuff has happened, and that has to be documented, but we should be less about the blame game as the current article does, and instead look towards what this has meant in changes to the industry (even if these weren't the changes that GG really wanted).

You're an admin and on this topic you're pretty clearly an "involved" admin. Can you talk a little about the difficulties of handling violations in areas where your actions could be perceived as biased?

• Masem – It can be very tempting at times to use admin powers to block a voice you don't like, but I have been extremely careful to even think of considering that here. As an admin, in general, you have to wear one hat while contributing as an editor to an article, and a different hat when you are looking at preventing disruption of WP, and that's the stance I've been taking for the GG article, in that I'm just one voice, and do not expect my admin status to mean anything special.

You have been described by the online name-and-shame group, SeaLionsOfWikipedia as a kind of leoni marino de tutti leoni marini. What is a "SeaLion" in this context and why do you think you have been labelled this way?

• Masem – "Sealion" is a general term that those opposed to the GG movement have adopted to call anyone associated with the movement, based on a Wondermark webcomic [1]. (In the comic, a sea lion randomly breaks into a conversation when one character mentions "sea lions" and continues to pesters the characters for some time, and was done as a allusion to how GG proponents were reacting on social media.) I know I have been labeled as a sealion because while I do not side with the GG movement, I also don't take the line that the anti-GG would like to have drawn (one that is a highly critical, condemning take on the GG movement) and instead want a more neutral stance; in such a polarizing situation as GG is, this type of attitude makes me appear as a GG supporter or a sealion. Further, there were claims earlier that I was engaging with GG supporters to get them to help edit the GG article on WP, leading this charge. I know my name has been bantered around the various GG forums as "based Masem", and before the SPA controls were added to the articles, those new editors would often iterate the points I made about neutrality, so I can certainly see how this appears that I am leading them. But I have avoided any interactions with this groups on purpose, much less to use the groups as a tool to challenge the WP article. Everything I've done on the WP article is simply my own purpose of writing a neutral article on a difficult situation to write neutrally about.

You have made the case that in areas where the reliable sources may have a vested interest in skewing the truth (e.g. in cases where sources are covering the propriety of their own actions), it may be important to emphasize truth at the expense of verification. Is that accurate, and can you elaborate on that a little?

• Masem – My intention is not so much about emphasizing the truth but, going back to my short summary of GG, it is "a complicated mess", and it should be readily clear that the truth is unknown. There are things that we simply don't know because of the nature of GG's apparent disorganization; what their true motives are (if they are different from what is claimed), and who has engaged in harassment, for example. We also have a situation of where the press and their actions are put into question, and with no evidence to judge on, it's difficult to take the press's word they have done no wrong as fact. So much of the GG coverage are details that we cannot validate as the truth as a third-party. To that end, my stance for most of the article to apply WP:YESPOV's advice, and report much of what can't be clearly verified or have been argued against by others as claims rather than fact as other editors of the page insist we do. This does not weaken our verifyability policy, since we still are reporting and sourcing these claims and not trying to substitute those with a specific truth that cannot meet WP:V. At the end of the day, this remains a controversy and like many other controversies there is no one right answer. Wikipedia has covered these in the past with fairly neutral coverage, there is no reason that the GG cannot be handled in the same way.

The Devil's Advocate, your user page displays a "private name" user template indicating that you would like your name to remain confidential. Do you feel any internal conflict over your participation in the Wikipediocracy forum when WO is responsible for the doxxing of multiple Wikipedia editors including those who might be reckoned to be aligned with your perspective on the GamerGate issue? In your estimation are external watchdog groups like Wikipediocracy effective in their efforts to counteract what has been called Wikipedia's "House POV"? Why or why not?

• The Devil's Advocate – To some extent I see parallels with the coverage of GamerGate and how Wikipediocracy is perceived by many in the Wikipedia community. I once had to deal with an editor routinely referring to me as a member of the "boxcutter crew" because of a, long-deleted, remark from someone on WO about wanting to "slit some nerd throats" with a boxcutter because of corruption concerns at a Wikimedia chapter organization. Biggest difference is that the person who made that remark was a moderator at the time while none of the threats, hacks, swattings, or doxings, covered in the media have been provably connected to anyone active in GamerGate let alone a significant figure.
  I have never taken seriously anyone's attempt to try and associate me with other people's behavior on Wikipediocracy, which is why my user page contains a joking homage to the "box-cutter crew" attack. There are many times when I have strongly disagreed with the course the site and other forum members have taken with a given issue and GamerGate has proven to be one of those issues. For me it has never made sense to associate a group with the worst of its actors or even to tarnish a person with their worst actions and I suspect those doing it are just looking for an excuse to avoid covering something favorably that they don't want covered favorably. After all, you don't see many (as in any) hit pieces about Wikipediocracy in major media, despite them referencing and even working with the site regarding Wikipedia controversies.
  As to Wikipediocracy's effectiveness, I am sure you can ask other members who will be able to give a far better and more detailed explanation, but I do believe it has been effective in many cases at exposing abuse of Wikipedia's vulnerabilities as a relatively open editing community. That exposure has on many occasions led to those abuses being addressed.

To what extent do you believe underdog politics is at play in the GamerGate arena? Is it your sense that the kind of editor who is drawn to the GamerGate article is either a dyed-in-the-wool GamerGater or feminist, or is this more of a conflict between otherwise ideologically disengaged editors who simply disagree with the balance that is struck between objective neutrality and RS-proportional coverage? Are there (m)any actual GamerGaters or Feminists editing the article?

• The Devil's Advocate – I do not think it is a distinction between GamerGaters and feminists as there are feminists who support GamerGate and not all opponents of GamerGate are feminists. As is the case in every controversy that makes its way onto Wikipedia, people actively taking part in the GamerGate dispute outside Wikipedia on all sides have also been taking part in it on Wikipedia. What matters is that the people involved are willing to collaborate and work towards an article that presents the most comprehensive and neutral view of the topic. I can't speak for everyone, but I personally got involved because I wanted to help give GamerGate a fairer shake than they were getting on Wikipedia. Also, I happen to agree that the news is full of shit.

The outcome of the GG ArbCom case landed with particular heaviness on two editors: Ryulong and you. Ryulong was outright banned, and you received a total of four sanctions including behavioral prohibitions and topic and noticeboard bans. The majority of the (reliable) press, when considering the issue, seem to expressed disapproval or discomfort with the ArbCom outcome. Some have characterized it as an ostensibly even-handed ruling with anti-feminist consequences insofar as several feminists editors at the article were removed while the GamerGate accounts that were removed were primarily throwaway and sock puppet accounts. Is that an accurate assessment?

• The Devil's Advocate – Nope, none of the editors named as parties in the case could be fairly described that way. I do think it came down harder on editors taking a harsh stance against GamerGate, but that was partly because those taking a more moderate or sympathetic stance had already been getting taken down or driven away in droves. ArbCom simply took a step towards balancing the scales.

What is your take on the fact that only around 1 in 12 editors at the GamerGate article are female? (For comparison, female membership of the four WikiProjects under whose aegis GamerGate falls stand at around 1 in 33 for Video Games, at around 1 in 9 and 1 in 12 for Internet Culture and Journalism respectively, and at more than 2 to 1 for Feminism). Is the gender gap a problem at Wikipedia? Do you suspect that squabbles like that which has occurred and which is currently occurring at the GamerGate article are more likely to increase female participation (e.g. by providing an issue around which females might be inclined to rally) or to decrease female participation (e.g. by providing a toxic arena of partisanship and legalism rather than one of compromise and dialogue)?

• The Devil's Advocate – The gender gap is definitely a problem. Whenever one group dominates a space it can create a systemic bias and I have seen some clear instances of that bias affecting Wikipedia. However, there is also a problem with people taking issues that affect all editors or all content and making it all about one gender. Not every action that negatively affects a woman on Wikipedia is caused by the gender gap. I think the controversy over the ArbCom case and the overall GamerGate dispute on Wikipedia is one of those situations that has been wrongly treated as a gender gap issue and that has provided a cover for a lot of bad behavior.

One of the more interesting examples of the crossover between Wikipedia and other online communities comes in the form of Auerbachkeller, a tech writer for Slate whose recent articles on Wikipedia have been dissected in The Signpost (See "additional readings" below). Auerbach's October 2014 "Divide and Conquer Plan" article for Slate (advancing the idea that GamerGaters are a heterogeneous group that might be neatly divided by responding to reasonable concerns over journalistic ethics while ignoring unreasonable misogyny-based concerns) raised the hackles of many who are opposed to GamerGate. A subsequent response in Salon by Elias Isquith (arguing that any effort to find common ground with those who self describe as GamerGaters is a dangerous mistake) was then used off- and on-Wikipedia to criticize Auerbach's article. Auerbach soon joined Wikipedia, demanding the removal of summaries that misconstrued his position, and provoking commentary from Jimbo Wales on the involvement of editors at the GamerGate article page. These and other actions taken on Wikipedia have earned him the title of "professional Sea Lion" at the online name-and-shame group, SeaLionsOfWikipedia. There is a small number of ostensibly or professedly neutral parties like Auerbach, including administrators involved in content-edits (e.g. Masem), and in sanctions enforcement (e.g. Gamaliel). This kind of editor is frequently vilified by the ultra orthodox on either side of the debate, but is this fair to them? How do you view their treatment in relation to this issue? Is it possible to remain neutral on a topic this polarizing?

• The Devil's Advocate – I wouldn't describe Gamaliel as neutral, even if he insists otherwise, and I believe both Auerbach and Masem consider themselves to be anti-GamerGate. That said it is revealing how certain people are so readily branded as sympathizers with GamerGate simply for not taking the harshest stance imaginable. You run into serious difficulties trying to handle an issue objectively when people on one side treat any hint of sympathy for the other side as spirited support, especially when they unironically view the other side as terrorists. Seems that is more of a problem for opponents of GamerGate than supporters, though I have seen it from supporters as well. Neutrality is certainly possible, but people who are neutral are more likely to avoid these kinds of polarizing situations. The real focus should be on objectivity and fairness and that is where I believe Auerbach and Masem differ from their critics.

Ryulong, can you explain the most significant ways that Wikipedia's dispute resolution structure frustrates or constrains the efforts of well-intentioned editors? Should Wikipedia have special fast-track channels (perhaps akin to WP:PROD and WP:SPEEDY in the AfD realm) for "raid" cases where large numbers of non-Wikipedians arrive en masse to force a change? Should Wikipedia provide specialists (perhaps from the among the clerks) with informed perspectives on the content rather than treating each case as nothing more than an examination of behavior?

• Ryulong – I think that the 300/50 (or whatever it is) rule that came about certainly helps matters, but there definitely should be some process to deal with issues like these before they reach the level of arbitration. It's unfortunate that these topical restrictions and honor system restrictions come after the fact, because when it comes to these extremely controversial topics (or at least topics where there's external pressure being put on Wikipedia), it helps to keep those who know what they're doing around while also restricting contributions from new people who may not be acting in good faith.

Is there really a debate here, or is it just manufactured by a radical fringe? Has Wikipedia elevated the equal treatment of both sides of this debate at the expense of passionate editors who have little patience for positions far from the mainstream?

• Ryulong – There was never a debate. It's a group of people in a them vs. the world situation, as has happened with other infamous arbitration cases in the past (Church of Scientology, LaRouche, climate change, etc.). It's a small group of people with loud voices and vicious behavior behind them. Any discussion of Gamergate should at the least say what they want people to say about them (which is that they're about "ethics in video game journalism") but then use the dozens of analyses that prove that they're anything but what they claim to be.

Are there any positives that have come from your involvement in Wikipedia's GamerGate article? Do the positives outweigh the negatives in this case? Do you view your involvement with the topic as a mistake? How would you advise others to approach their own involvement?

• Ryulong – There's not much positive that's come about from my involvement in the article. The only thing that's happened is that I have a larger group of acquaintances on Twitter and I have to sit back and watch messes I would usually clean up on Wikipedia get worse. And I don't really know if my involvement was a mistake. The Wikipedia page is as good as it can get but I do have a peanut gallery watching my every move on Reddit now, which is worse than it was before when it was only 4chan commenters disliking translation choices. And my advice is don't do it unless you're prepared to have your entire online history dredged up to be determined if you're on their side, cause if you're not, I hope you like being listed on all their hitlists.

Wikipedia has struggled in the past with issues where Wikipedians have online presences that extend beyond Wikipedia. Some believe that off-wiki behavior should be considered and even sanctioned on-wiki. Others would maintain a strict separation of on-wiki and off-wiki behavior even going so far as to sanction those who draw connections between on- and off-wiki accounts for "outing". Do you have any thoughts regarding this issue in light of the various off-wiki actions of participants on either side GamerGate divide?

• Ryulong – There certainly comes a point where actions off of Wikipedia should be taken into account with regards to the person's behavior on Wikipedia. If it's obvious that someone is part of a group operating in the open off-wiki to affect people and content on-wiki, then their behavior as part of that group should be taken into account. Agendas are very easy to discern when it comes to topics as divisive as Gamergate, and people with mostly identical screennames acting in identical ways on two different websites should be taken into account.

Speaking as someone who has been the target of cruel and offensive off-wiki profiling in response to your involvement with the article, can you speak about the extent to which this controversy has become personalized? In your view is this the "new normal" for the social media generation?

• Ryulong – This is simply the nature of the reactionary assholes of the Internet, borne of 4chan and now spread throughout the rest of it. It's not really an issue of social media as much as it is the assholes who ruin things for everyone else.

You are among a small group of editors that were named the "five horsemen of wiki bias" by off-wiki forum commenters. What is a Horseman in this context and why do you think you were singled out for this label?

• Ryulong – Tarc, NorthBySouthBaranof, TheRedPenOfDoom, TaraInDC, and I were simply the five most active editors responding to the all of the Gamergate supporters, editors who came to the page to try to push for more dominant coverage of Gamergate's claims, which were never found in reliable sources (and as far as I'm aware still aren't). After we were banned in various ways, they turned their attention to Mark Bernstein, Gamaliel, and others who came to the article to edit it in good faith to continually respond to the same talking points that have been going around since August 2014. There's no real threshhold to be considered a "horseman of wiki bias". You simply have to edit the page in a way Gamergate supporters don't like.

Warmly received at RationalWiki following Wikipedia's ARBGG case, you seem to have made something of a home for yourself there. At a combined 1900 edits (and growing) on Wikipedia and RationalWiki, your personal contributions to the encyclopedic coverage of GamerGate is staggering. At RationalWiki your (at times) turbulent Wikipedian past has not proved to be an impediment. Would it be fair to say that this is because RationalWiki's 4 Purposes (see RW:ABOUT at RationalWiki) emphasize "documenting the full range of crank ideas" (purpose 2) whereas Wikipedia's 5 Pillars demand collaboration (pillar 4), neutrality (pillar 2), and encyclopedicness (pillar 1)? Without putting too fine a point on it, can you explain the most significant ways that Wikipedia constrained your efforts and whether you feel that these constraints are generally (e.g. in non-GG-related areas) helpful or harmful?

• Ryulong – Well "neutrality" is really the only thing that has caused any problems when it comes to Gamergate. There's a difference between neutrally and accurately covering a subject and giving undue weight to claims in order to push a false balance. As I've alluded to already, Gamergate supporters are on the same side as climate change deniers and others who think some mainstream entity is inherently biased against them and that their opinion on the subject should be given just as much credence as the one that dominates the discussion. For climate change denial it's the insistence that anthropogenic global warming is a myth. For 9/11 truthers it's that the government was involved in the attack on the Pentagon and World Trade Center. For birthers it's that President Barack Obama wasn't born in the United States. For Gamergate supporters it's that there's rampant corruption in the video game industry and their proof is a diatribe written by Zoë Quinn's ex-lover which only ever gained a foothold on known Internet troll den 4chan after it got banned from everywhere else. And yet they wonder why no one believes what they have to say (not that people haven't looked in on their activities which they do in public and reported on that, often at the behest of Gamergate supporters). The fact that RationalWiki's page on Gamergate called out all of Gamergate's regular lies (discovered from personal experience) just like they call out the lies that other entities peddle is what led me to join.

Having spent considerable time editing at both the Wikipedia and at RationalWiki versions of the GamerGate article, one wonders if you have had the opportunity to examine the coverage of this topic at other wikis. You may be aware that large articles on this topic exist at both Conservapedia and Encyclopaedia Dramatica. GamerGate supporters have at times operated their own wikis, and then there are at least 14 different non-English versions of the article at Wikipedia (often containing completely different content than the en.wikipedia version). How do these different versions compare in your opinion?

• Ryulong – Conservapedia and Encyclopaedia Dramatica are lost causes to everyone. From what I can tell, coverage on non-English Wikipedia projects is for the most part decent. Some focus too much on one miniscule aspect of the topic as a whole. But there is one which is just plain garbage because it's on a low traffic language project and no one (as I'm writing this) has edited it other than an IP editor and some local admins who tagged it with cleanup tags.

Editors like your author look forward to the future of this article with very little trepidation. There is a sense on both sides of this controversial topic that there is no way to compromise with the other side. This is understandable and, given the nature of the debate and the fact that for some the middle of the road is already the other side, one hesitates to even suggest that compromise is possible.

Nevertheless, Wikipedia has a roughly 150,000-byte article on this topic that is relatively stable despite regular acts of disruption. This article wasn't created in a vacuum by one or the other side without the input of the other side, so there are indications that on some of the most basic elements a thin agreement exists. This isn't the same as a compromise, of course, but it's perhaps a useful mental starting point.

Wikipedia places a great amount of trust in its policies, guidelines, and community infrastructures to advance through intractable problems. Although the details may be somewhat subject to pettifoggery, the basic premise that content should be based on reliable sources instead of original research is one that both sides agree upon. The methods of determining reliability are also a matter of general agreement with dubious cases subject to review at RSN. For the rest of it (i.e. for those matters more within the realm of editorial discretion), it seems clear that time will act as the great panacea.

Despite the darkly muttered warnings and forecasts of doom arising whenever Wikipedia covers a contentious issue of pop culture given to spinning by groups of passionate and self-righteous editors, there are still reasons to be hopeful. The coverage of pop culture minutia depends on RS-conferred notability. The lowness of this bar for inclusion may be especially apparent when the topic is especially irritating, yet if multiple RSes cover a topic in sufficient depth to craft an encyclopedic article then the topic has become a cultural reference point however shameful it may be. Wikipedia here sets itself up to collect a kind of "cream" of our memetic "crop"—topics which have broken through and claimed attention from the gatekeeping institutions of cultural relevance, like newspapers, journals, academic papers, books, and other sundry reliable sources. In doing so, it establishes a firm toehold in the future as a reference work for researchers seeking to make sense of the output of our increasingly encoded and meme-driven society.

Where previous generations had to rely on a common pool of classicism for their allusions to Aeschylus and a moderately literate society for their oblique references to Yeats, future generations may well turn to Wikipedia for their explanations to ancient "Dancing baby" and "All Your Base" references etched on fading digital papyri in archival Geocities grottos. In the final analysis, your author is convinced that Wikipedia's coverage of low but culturally-referenced topics like GamerGate matters. And if it is any comfort to those who believe their side of the content wars is receiving far more push-back than is due, it is worth considering that there is nothing the historian loves quite so much as a correction that needs to be made for the record.

The article history at the GamerGate article records every non-BLP-violating previous version of the article and will presumably continue to do so while this encyclopedia retains its digital form. Prior edits are not lost, and the seeds for potential vindication have been sewn and are part of the record. Only time will tell if they are sufficiently viable to germinate. This battle within the greater culture war is winding down and moving into other realms, and it will be interesting to see how history treats this topic in the future. There is no question that the cultural anthropologists who would be interested in sifting through the background material on just the Wikipedia side of this controversial issue will have a true embarrassment of material to sift through.

• Auerbach, David (5 February 2015). "The Wikipedia Ouroboros". Slate. The Slate Group. ISSN 1091-2339.
• Chituc, Vlad (11 September 2015). "GamerGate: A Culture War for People Who Don't Play Video Games". The New Republic. Chris Hughes.
• Dewey, Caitlin (29 January 2015). "The Intersect: Gamergate, Wikipedia and the limits of 'human knowledge'". The Washington Post. Fred Ryan. Retrieved 12 October 2015.
• Flöck, Fabian; Laniado, David; Stadthaus, Felix; Acosta, Maribel (2015). Towards Better Visual Tools for Exploring Wikipedia Article Development — The Use Case of "Gamergate Controversy". 2015 International Conference on Weblogs and Social Media (ICWSM). Association for the Advancement of Artificial Intelligence (AAAI).
• Hern, Alex (23 January 2015). "Wikipedia votes to ban some editors from gender-related articles". The Guardian. Guardian News and Media. Retrieved 12 October 2015.
• Auerbach, David (28 October 2014). "How to End Gamergate". Slate. The Slate Group. ISSN 1091-2339.
• Isquith, Elias (1 November 2014). "Gamergate's infuriating myth: Why searching for common ground is a big mistake". Salon. Salon Media Group. Retrieved 12 October 2015.

• Gamaliel. "News and notes: Superprotect, one year later; a contentious RfA – Contentious RfA draws nearly 300 participants". The Signpost. Volume 11, Issue 32, 12 August 2015.
• Harry Mitchell. "Arbitration report: We've built the nuclear reactor—now what colour should we paint the bikeshed? – Other business". The Signpost. Volume 11, Issue 7, 18 February 2015.
• Gamaliel. "In the media: Wikipedia on 60 Minutes, Kickstarter, and in the classroom – Another victim of notability?". The Signpost. Volume 11, Issue 14, 08 April 2015.
• Gamaliel and Andreas Kolbe. "In the media: Gamergate; a Wiki hoax; Kanye West". The Signpost. Volume 11, Issue 10, 11 March 2015.
• Gamaliel. "In the media: Is Wikipedia eating itself?". The Signpost. Volume 11, Issue 6, 11 February 2015.
• Gamaliel. "In the media: Gamergate and Muhammad controversies continue". The Signpost. Volume 11, Issue 5, 04 February 2015.
• Harry Mitchell. "Arbitration report: Slamming shut the GamerGate". The Signpost. Volume 11, Issue 5, 04 February 2015.
• Masem and Protonk. "Forum: Evaluating the Arbitration Committee's handling of GamerGate". The Signpost. Volume 11, Issue 04. 28 January 2015.
• Go Phightins! and Harry Mitchell. "In focus: Thirteen editors sanctioned in mammoth GamerGate arbitration case". The Signpost. Volume 11, Issue 04, 28 January 2015.
• Harry Mitchell. "Arbitration report: As one door closes, a (Gamer)Gate opens". The Signpost. Volume 11, Issue 03, 21 January 2015.
• Gamaliel. "In the media: Wales on GamerGate". The Signpost. Volume 10, Issue 50. 24 December 2014.
• Guerillero. "Arbitration report: In brief". The Signpost. Volume 10, Issue 40, 15 October 2014.

This special report first appeared in the WikiProject Video Games newsletter. It has been formatted and edited for publication in the Signpost. The views expressed here are the author's alone and do not reflect any official opinions of this publication; responses and critical commentary are invited in the comments section.