Wikipedia:Wikipedia Signpost/2011-09-26/Opinion essay

''Jan Eissfeldt is a German Wikimedian, holding down the position of administrator on the German Wikipedia and administrator and bureaucrat on the cross-community Outreach wiki. Here, he writes about the nature of the German Wikimedian community, and what lessons might be learned from its response to the recent proposals for an opt-in image filter.''

''The views expressed are those of the author only. Responses and critical commentary are invited in the comments section. The Signpost welcomes proposals for op-eds. If you have one in mind, please leave a message at the opinion desk.''

Something is different on Meta these days. Substantial parts of the German community, usually focused on itself and the three German language Wikimedia chapters, are going up the walls. Why is that?

Well, to keep an extremely long story very simple: they really hate the image filter and therefore the German projects are the only corner in the Wikimedia universe (it seems to me) heavily resisting this feature following the recent global referendum (see previous Signpost coverage).

According to stats.wikimedia.org, approximately 35% of the core community of the German Wikipedia (357 users) voted to totally reject the idea of the proposed image filter as they interpreted it in a local poll that publicly passed with 86% support. Traditionally, the community sorts out differences in binding crucial votes and nearly half of the core community took part in this one. Additionally, I attended their annual community conference this month and can tell you: many of those who did not participate in the poll aren't happy either. As a result, the Terms of use update is mainly under fire there right now for this reason.

What is different
So, what's different about these folks? For one, their way of building and running their projects. The German community mainly lives in a concentrated geographical area, smaller than Texas, and is extremely homogeneous even by Wikimedia standards. As a result, their community life is mainly conducted via real life meet ups, called Stammtische, and users are on record for knowing up to 350 other users personally – in a core community of approximately 1000 people. Fine, that makes them something The Guardian's Michael White would most likely call a tribe and the academic studies on this community largely agree. But since the image filter is a tool for readers, that doesn't really count, right?

Right, but take a look at their readers. Public debates about the Internet in Germany have centred around the idea of filters for years now. The so-called Grand Coalition introduced them back in 2009, in an attempt to combat child pornography, but the public reaction was disastrous and filters were accused of being tools for protecting the abuse they were supposed to attack. For the very first time a national government was forced to not enforce a law it had already enacted, and the constitutional court had to deal with the question; on a more localised level, major German sites don't use filters either. The law, called the Zugangserschwerungsgesetz, was finally abolished in 2011 in response to wide-ranging criticism.

It seems that filters aren't a popular concept in Germany generally and you can largely count Austria and the German-speaking part of Switzerland there as well. How does this translate into the view of Wikimedians? Well, the German Wikipedia reached an unusually high degree of penetration and reputation relative to the related language community. In 2007, Stern magazine claimed that the German Wikipedia was more reliable than the Brockhaus, the largest traditional German encyclopedia. Additionally, the German project already had its share of practice tests (Signpost coverage) and there was no major public outcry. If we are ready to take that as a given, they are doing all right and their content doesn't seem to be controversial to their particular audience. The obvious exception is their hostility towards popular culture, which is often cited as a reason to use the English Wikipedia rather than the German one. Coincidentally or not, so far the biggest flare-up of this disagreement between German readers and community was triggered by the deletion of an article (Signpost coverage) about an anti-filter organization protesting the bill in 2009. In essence, German Wikimedians mirror the population they are drawn from.

What follows out of difference
That said, what does that mean for us? Well, we as participating community agreed with the Wikimedia entities back in the strategy process 2009/2010 that spreading free knowledge in the foreseeable future will mean reaching out to the Global South and to do that means among other things – in my understanding – going into the messy business of balancing out arising value conflicts, which are non-binary by definition.

The image filter tool, like it or not, can potentially ease such debates by free choice-diversification of viewed images (which are different from text for all sorts of well-known anthropological-epistemic arguments I can't name here), while preserving our "NPOV/one article" approach. All-in-all a solution far more clever than what was tried on Commons back in 2010 and it takes into account what we know about people without a proper voice inside our debates – and the readers and contributors we all are looking for as well. This looks quite settled to me, since I got the impression out of the referendum that low-level contributors are significantly more enthusiastic about the whole idea than the core community. These occasional contributors without a proper voice in our usual decision-making processes are the nearest we can get to readers and, by taking into account their views, the Board of Trustees was designed to push for unpopular proposals if necessary. But the global conclusion misses the point regarding what I wrote above about the structurally different German case and dealing with them regarding any given topic – as the response to the terms of use update shows – will become even more difficult as long as their identity question dominates minds. There is no voiceless German-speaking global south – or east, in other cases – to be safeguarded.

What should be done
I could, of course, be wrong about this local audience as a whole and this local community in particular. But taking the outlined unique tribe and the global position both as a given: it seems reasonable to me to look at this special case, maybe within a few months after implementation to check whether or not I'm right regarding readers, for a proper local response to this local problem.