Wikipedia:Wikipedia Signpost/2013-06-12/Op-ed

Op-ed

The tragedy of Wikipedia's commons

Gigs is an English Wikipedia editor who has been on the site since 2005.
The views expressed in this op-ed are those of the author only; responses and critical commentary are invited in the comments section. The Signpost welcomes proposals for op-eds at our opinion desk.

I've long thought that we should get rid of the Commons as we know it. Commons has evolved, through the actions of a tiny group of people, into a project with interests that compete with the needs of the various encyclopedias that are the primary users of Commons, and the reason it was created. It's also understaffed, which results in poor curation, large administrative backlogs, and poor policy development.

First, some background information. Commons was primarily created so we could share media between various wikis, with a secondary goal of being a free media repository. When Erik Möller proposed the idea of Commons, he also proposed an inclusion criteria, "Material would be eligible for inclusion in the Commons if it is useful to at least ONE Wikimedia project [including potential future use]."

At no point during initial discussions was it proposed that the inclusion criteria basically be the mere fact that an image was free. There was an implicit assumption throughout that the files would be free, and also encyclopedic in some way.

From inception until 2008, the main inclusion criteria at commons was the media be "useful or potentially useful" to a Wikimedia project, reflecting Möller's initial proposal comments. In 2008, a replacement policy was proposed and implemented by User:MichaelMaggs, with half a page of feedback from about six other editors. These six editors (some seemingly unwittingly) redefined the scope of Commons from a repository of files useful to Wikimedia project, to files "useful for an educational purpose".

This unchallenged action by a tiny group of people changed the scope of the project such that any media file with a free license can be included, since it is extremely easy to argue that any media is "useful for an educational purpose", no matter what it is. Even a file of white noise could be a study in tests for randomness.

In proposing Commons with a dual mission, Möller conflated solving a technical problem with a new project that turned out to be of very niche interest, a image/media gallery project that's similar to wikisource for images. If it weren't for the interwiki sharing function of Commons, it would probably enjoy a similar level of success as Wikisource; limited. So now we have a niche project with limited volunteers effectively dictating policy across all our projects.

Commons has come under attack many times in the past for hosting low-quality images of nudity and sexual acts, mostly uploaded by apparently exhibitionist editors, with the subjects being themselves or their partner. To even discuss removing them is framed as censorship. I'm not offended by nudity. What I am offended by is people abusing the encyclopedia for their own ends, to the detriment of the project.

Nearly all our policies are driven by the need to prevent this sort of abuse of Wikipedia. Policies on biographies of living people are driven largely by those who would abuse Wikipedia for purposes of defamation. Policies on neutrality and verifiability have been largely driven by the need to address those who were here to push a political agenda or promote their fringe viewpoints. What Wikipedia is not is pretty much a chronicle of all the things that people have tried to use Wikipedia for that the community has decided are detrimental to a quality encyclopedia.

Preventing uses of Wikipedia that are detrimental to our mission is the entire reason that most of our content policies exist. This isn't censorship, it's curation. There is no reason we should indulge exhibitionists who spew copious nude or sexual pictures of themselves or their partner across dozens of keywords for their own gratification, any more than we should tolerate the link spammer who spews their links across Wikipedia.

Commons has consistently failed to develop a reasonable policy on this matter. To me, this is just one more example of the failure of Commons due to lack of participation and conflicting mission. A small group is more likely to develop a self-reinforcing delusion that their position is reasonable, even when a large number of people outside the group are telling them otherwise.

Some have challenged my titling of this work "Wikipedia's commons", pointing out that the proper name is Wikimedia Commons. The encyclopedias have a dependency and tight integration with Commons, which would like to govern itself as a separate project with a completely different, and potentially conflicting mission and policy. Whether either project likes it or not, right now Commons is part of Wikipedia.

There is a fairly simple technical solution to the problem. Commons, for the purposes of interwiki sharing, doesn't need to be a project with separate administration. What we know today as "commons" can simply be a feature of Mediawiki, it doesn't need to exist outside of some computer code. When a picture or media file is uploaded to an encyclopedia, if its license is compatible, it will be registered as sharable, but still "owned" by that wiki. If the media violates that wiki's policies, it can be deleted. If another encyclopedia makes use of it in their article space before it's deleted, they will become the owner of the media. Search should easily allow the option of only searching locally owned media, or all media including interwiki media. I'm sure there are some details of this proposal that are not fully fleshed out, but I don't think any are insurmountable. The point is that the interwiki media function is a technical problem that can be solved by technical means.

The free repository function of today's Commons could still exist as a separate project, call it WikiGallery or something. In no case should it be a part of the encyclopedia interwiki media system, however, which should only include encyclopedic media.

The title of this is an homage to the tragedy of the commons. I don't know if it's hilarious, ironic, or sad that Erik Möller chose the name commons, citing the "positive connotations" of the word. Little did he know it would foreshadow the eventual lack of maintenance and resulting lack of policy maturity that would result.