The balkanisation of Wikipedia

I will begin this article with a controversial proposition: the Wikipedia community is fundamentally dysfunctional, and unless something drastic is done the great progress made by Wikipedia to date will grind to a halt.

At this point you might expect some tiresome comment about cabalism, process wonks, rogue admins or some other such tripe. Those ‘problems’ aren’t really the problem at all - the problem is that a community of people writing an encyclopedia even has such terms at all.

When I started editing Wikipedia ((My first edit as a registered user - I had made a few anon edits before then - was correcting a misplaced bracket on the 22nd of May 2004)), everyone was concerned with the quality of the articles we were writing. There was no need for a ‘process vs product’ flamewar, because nobody was behaving in such a way that we needed product-restricting processes to control them.

At some point in the intervening three years ((OK, I’m out by a month, who cares)), vocal elements within the Wikipedia community - both ordinary users and the hierarchy - have decided that they don’t want to write an encyclopedia any more. What they really want to do is spend all their time promoting or shouting down each other’s RFAs, deleting or saving each other’s articles, arguing about the said behaviours, arguing about arguments, and arguing about arguments about a policy controlling arguments. Still following me?

At the centre of this is the ‘instruction creep’ which has become overwhelming under the banner of ‘helping Wikipedia cope with a larger userbase’. Before, a Vote for Deletion ((Yes, I was editing Wikipedia back when a vote was still called a vote - now it is called a ‘Discussion’, and calling it a ‘Vote’ is doubleplusungood. This despite the fact that the process involves users giving the Wikipedia equivalent of Yea or Nay, and an admin counting those for and against deletion)) was a simple matter - was the article good or bad. People might have differing opinions as to when a poor quality article (or a very short one) became so bad as to warrant deletion, but on the whole everyone agreed that bad articles should be deleted, and good or mediocre articles (or those which had a reasonable chance of being improved to such a state) should be kept.

Now, on the other hand, everything is codified. What used to be ‘I don’t think this person, product etc. is important enough for a Wikipedia article’ is now an acronym soup of ‘notability policies’. What used to be ‘this is obviously some kid writing an article about his band, which plays in his garage’ is now ‘DELETE. national TV broadcast about them only 29 minutes long, therefore fails WP:MUSIC’ ((I’m really not kidding about that one!)). When did Wikipedia users stop being judges of article content and start being cybermen?

Hopefully you are now aware of the problems ((If you weren’t already)). The catch is that I don’t know what the solution is. A start would be to take a blowtorch to all the excessive policy, notability fluff, and other quasi-legalese which does nothing but hamper article editing. Delete everything except the core policies that are actually necessary to run Wikipedia - the Five Pillars ((Of Wikipedia, not of Islam)), and possibly a cut-down version of the deletion policy.

The wikiwarriors, of course, would not allow this. Take away the notability policies and you take away their excuse to argue over whether Latvia constitutes a ‘medium-sized country’ or not for the purposes of music notability. Take away most of the responsibilities of adminship ((In a properly-organised Wikipedia, it would be the community that decides whether a particular article or image is deleted, and all that the admin would be required to do is what the community tells him)) and you take away their excuse to argue over whether 68% of the unvote is sufficient for ‘promotion’ to admin. That is why I think that, while Wikipedia might have more than a million articles ((Actually approaching two million now - about 1.7 last time I looked)), it is in a noticeably worse state than it was in 2004, or even a year ago, and will not go much further without some radical changes.

3 Responses to “The balkanisation of Wikipedia”

  1. David Gerard Says:

    I wrote Practical process as an attempt to provide tools for those hacking through excess red tape. It doesn’t seem to have solved the problem all by itself :-) Time for a followup essay?

    It is important to note that (a) most articles are not controversial at all - so excess bureaucracy for the problem cases will just hurt the rest - and (b) newbies write most of the actual text, so jargon directly harms the process of writing an encyclopedia.

    There’s no way we’re going to do without process - with 4500 frequent editors any month and 45,000 occasional editors any month, you’re going to have politics and need bureaucracy. But there’s a number of problematic attitudes that need addressing.

  2. David Russell Says:

    Do we really need all this bureaucracy though? Surely if we actually enforced the five pillars then a lot of the fluff could be put out to pasture without any negative effect? For example, there would be no need for many different notability guidelines if we actually enforced WP:V and WP:RS. Anything which is not notable will not be verifiable by means of reliable sources, so we can do the job with two policies instead of twelve.

    That’s just one example of how overgrown Wikipedia’s policy has become - we have reams and reams of legalese which don’t actually do anything that the ‘core’ policies (the Five Pillars, et al) don’t already do.

  3. David Gerard Says:

    You appear to have missed the raging flamewars over “reliable sources” - people inappropriately fitting articles to the Procrustean bed as a substitute for just deleting stuff they don’t like, its abuse as a bludgeon by POV warriors, etc.

    WP:LAP is promising.

Leave a Reply