Wikipedia: The Giano Problem and the Site's Seedy Underside

To many, Wikipedia's mantra of "the free encyclopedia anyone can edit" smacks of democracy, with a little bit of anarchy thrown in here and there. But in fact Wikipedia professes to be neither of those things; it is not a place where any editor is allowed to go around willy-nilly and do what he or she pleases. At the same time, Wikipedia isn't a democracy; in theory, decision making derives from consensus, not head counting.

What is consensus? The nutshell description for Wikipedia's page on the policy states three things:

*Consensus is about how editors work with others.

*Consensus is Wikipedia's fundamental model for editorial decision-making.

*Policies and guidelines document communal consensus rather than creating it.

In short, rather than voting on proposals or editorial decisions, editors are supposed to find common ground and reach decisions based on the strength of arguments, not numbers. Local consensus on a single page or a WikiProject cannot override consensus on a larger scale; administrators and regular editors alike are supposed to abide by consensus (for an explanation of WikiProjects and the Wikipedia hierarchy, see this article).

This is all well and good on paper, but how does it work in practice? The result, like the contents of Wikipedia itself, varies wildly. Certain constant areas of friction on the 'pedia (topics relating to nationalism, for example, or pseudoscience) are often determined by either the strength of whichever side fields the most editors, or by fiat from on high by restrictions of sanctions placed by the Arbitration Committee. Elsewhere, WikiProjects set project-wide rules that are commonly accepted, and issues are civilly solves on talk pages. Wikipedia proves that the consensus model can work, and can also fail in its application on articles.

But what about editors? It is editors who are the lifeblood of Wikipedia. While it is the encyclopedia anyone can edit, evidence suggests that it's a relatively small cadre of users who form the bulk of its content, specifically it 1s highest-quality audited material, that of Featured Articles, Lists, Pictures, Topics, et al. In 2008, just 10 people (a group of which I am a member) nominated 20% of all 719 FAs promoted that year. This is on a project where there are tens of thousands of active editors. Similarly, the top three nominators at Featured List Candidates have produced 15% of the encyclopedia's featured content in that area. Even in maintenance areas, it is often a small group who do much of the work--clearing backlogs, clerking, reviewing articles, sorting stubs and making fixes.

That isn't to say Wikipedia can do without its less-active or prolific users; quite the opposite. But it does show that in the egalitarian landscape Wikipedia lends itself to, there are naturally going to be some who rise above the others. And in disputes between editors, how are these different editors going to be judged and handled?

Take Giano, for instance. This editor specializes in architecture-related topics, and has a respectable 14 featured articles under his belt, according to the official list of Wikipedians by FA nominations. But he is perhaps better known by controversy that he has either caused, been the recipient of, or otherwise embroiled in.

Giano, formal username Giano II, has a sizable block log to his credit. Much of these blocks have to do with civility, and Giano has had the dubious distinction of being blocked (and unblocked) by Jimmy Wales, Wikipedia's founder. Giano is known for saying exactly what is on his mind, including calling people "idiots", "liars", "deluded" (or a "dellusionist [sic]"), along with a stream of profanities.

While few people would want to be on the wrong end of such comments in real life, the internet is known for looser tongues. Wikipedia has two policy pages outlining relevant behavior; on "Wikipedia: Civility", the nutshell description prescribes the following:

*Participate in a respectful and considerate way.

*Do not ignore the positions and conclusions of others.

*Try to discourage others from being uncivil, and avoid upsetting other editors whenever possible.

Furthermore, the page titled "No Personal Attacks" tells editors to "Comment on content, not on the contributor" and warns repeated attacks may lead to blocks. Clearly, Giano has violated these restrictions. Or has he?

Here's the first issue with the consensus model. What constitutes a personal attack? Is calling someone an "ass" a personal attack? Or intimating they are liars, cheats, or slanderers? What if there is evidence to back up accusations? Should Giano be allowed to plainly call a spade a spade?

Unsurprisingly, there are people on both sides of the fence on this issue, one that affects not only Giano but many other editors brought to administrator or community attention. The major difference is that generally the distinction is not dwelled upon. Giano, however, built himself a reputation as an excellent article writer. The result is that the community (or aspects of it) are loathe to drive away or permanently ban a chronically uncivil user who also produces good content.

The second issue in regards to the consensus model as applied to users is that because power is distributed across many users, one action is easy to apply, revert, and reapply. Constant reversions in blocking is one facet of a "wheel war", something that happens quite often where Giano is concerned. Let's take a prime example.

Back in late December 2007, Giano made several edits to the Wikipedia project-space article on IRC, a form of internet chat that is off of wikipedia (and considered out of the wiki's jurisdiction), but which is still used by Wikipedia editors. In the admins-only IRC channel, another user made an insulting comment that Giano found out about. He posted onwiki about it, which the offending user blanked repeatedly. He was blocked for personal attacks. Giano, in response, added a paragraph airing his grievances to the admins IRC page. Some of Giano's "friends", for lack of a better word, who were privy to the disagreement also made changes, and soon an edit war on the wording of the page. It was at this point that I became involved, after noting that Giano was violating revert guidelines, I warned him on his talk page. Mere seconds afterwards, he finished an edit, and noting the change (seemingly in disregard to the warning), I blocked him for an hour. As a timeline would show, I was incorrect in applying the block. [A note moving forward: the blocking policy prescribes "preventative" blocks rather than "punative" ones; that is, a block should be no longer than needed to prevent disruption, rather than being used as punishment; this was the reason for my short block.] Soon afterwards, another editor protected the IRC page, perhaps not realizing the main offender had been blocked.

A few minutes later, Giano was unblocked in order to provide a longer block of 24 hours. The blocking admin did so due to incivility on Giano's talk page after he was blocked. After being unblocked, the pattern of edit warring, heated discussion, and blocking continued. Eventually the conflict spilled over into a large discussion on IRC (not Giano) by the Arbitration Committee [a timeline of events I based this info on can be found there: Timeline].

The ArbCom put forth findings of fact that reminded Giano that he had been counseled on collaborative editing but had ignored this advice. He was subjected to an editing restriction which stated: "Should Giano make any edits which are judged by an administrator to be uncivil, personal attacks, or assumptions of bad faith, Giano may be blocked for the duration specified in the enforcement ruling below."

Here in enters the fact that certain admins are willing to unblock; a total of seven blocks in line with this remedy were made, with many shortened or overturned.

As the block log states, this has done nothing to dissuade Giano. In early April 2009, following offwiki comments to a newspaper blog that disparaged Giano, he took his grievances to the editors' request for adminship page and derailed the discussion. Despite being blocked (with as is custom, much warring about it), Giano remained unapologetic: in a note on his personal talk page he wrote that "Defending oneself from an unprovoked, lying and vicious personal attack on a highly public national newspaper blog is justified. I would do the same thing again. Please don't come here telling me I was wrong - I was not!"

So where does this leave the community? Giano, in being intransigent, ends up creating drama, as well as attracting it; the many kilobytes of discussion spent on requests for comments and arbitration cases and filed noticeboard reports... do they outweigh his or any user's positive contributions? At what point does an editor become a net loss for the project? And how does one determine that?

A similar case to Giano is Sceptre, a frequent contributor to Doctor Who articles who got himself blocked for months after some poor choices (including attempting to anonymously vandalize other editors' user pages). After being given another chance, he proceeded with name-calling and incivility (not to mention creating fake accounts) that proceeded to get him blocked several times (though never for good). In many ways Sceptre's use of sockpuppets while banned is far more egregious than Giano's, but Sceptre has been allowed to return. On a different tack, the user ScienceApologist, in many ways a militant editor like Giano (except dealing with the sticky subject of fringe science), was blocked for three months. Whether or not he comes back slightly earlier like Sceptre is up in the air.

As these examples demonstrate, actions on par with Giano's, or even beyond it, are simply not enough to allow for a total ban from Wikipedia. The consensus model means that enough users who support the user can lobby for decreased sanctions or check the administrator actions of others. What's the end result for the project? A lot more drama, to be sure, but any other results are hard to substantiate. Damage to the project appears to be confined to those who actually read filings or Arbitration reports; the casual readers who are not embroiled in the backend of Wikipedia seem not to notice. Discussion of this damage is confined to limited discussion on Wikipedia and its associated venues, as well as anti-Wikipedia sites such as the Wikipedia Review or Wikipedia Watch. Perhaps the greatest damage can be to the editors themselves, who have libelous smear pages dedicated to them on places such as Encyclopedia Dramatica. Ultimately, the consensus model fails to deal with all stripes of troublesome users, and it looks like a chronic condition that Wikipedia will have to live with.