Writing laws through a wiki

So an unnamed member of an unnamed government contacted the Wikimedia Foundation recently, asking for staff guidance on setting up a wiki so that unnamed citizens of the unnamed jurisdiction could collaborate on new laws. Since the staff mainly handle the operational aspects of the various wikis, the question got passed off to the volunteers — or, more specifically, me.

I sent a detailed response, advising strongly against anonymous editing and whatnot. The idea has been kicking around my head ever since, though, so I thought I’d post something about it and solicit further input on behalf of the unnamed elected official.

Long story short, I think that the proposed structure (”let’s get everybody together to write laws!”) is doomed to horrific failure, thanks to vandals, savvy agenda-pushers, and the fact that most people find most laws tedious and boring (and therefore would avoid watching the really important bills).

As evidence that vandals would pretty quickly make a nasty bitter mess of things, I offer the LA Times’s “wikitorial” experiment. Need more? Hang out on Wikipedia’s “recent changes” list and check out stuff in near-real-time; how many of those edits are actually useful?

Savvy agenda-pushers? I could refer you to any of those lovely articles on Wikipedia on controversial topics, but instead I’ll point you to this section of a Missouri law:

Services related to pregnancy, persons holding ministerial or tocological certification may provide.

376.1753. Notwithstanding any law to the contrary, any person who holds current ministerial or tocological certification by an organization accredited by the National Organization for Competency Assurance (NOCA) may provide services as defined in 42 U.S.C. 1396 r-6(b)(4)(E)(ii)(I).

This was added to a health insurance bill and didn’t attract a whole lot of attention until after it had already hit the books. Then somebody finally got around to looking up “tocology” — it means “midwifery”, a practice which was at the time very sharply limited in Missouri. It was a brilliant trick, and it worked (if only temporarily) because everybody was too busy looking at the Big Provisions to notice this one wee alteration.

Now go to Wikipedia and start looking at articles on topics you don’t understand: math, physics, why people keep opting for short-term gain and long-term loss, whatever. How quickly could you spot very subtle vandalism? I’m not talking about pictures of penises on [[Johannes Kepler]]; I’m talking about a number changed here, a date there, a minor turn of phrase… If Wikipedia relies on its vast pool of editors to spot these things, and yet they still miss the tiny-but-important details, then how would a smaller law-wiki defend itself against subtle bias?

Finally, there’s the question of popularity. It’s trivial to prove that some articles on Wikipedia get a lot more attention than others: there’s even a game, “wikigroaning“, that makes light of this by comparing the lengths of a very geeky article and one that is much more mundane. (The link contains a few choice examples.) “Wikigroaning” only works because Wikipedia is done almost entirely by volunteers. Editors on the English WP are generally more interested in their own pet topics, so they devote more time and energy to those topics than they do to other, perhaps needier topics.

On a law-writing wiki, this phenomenon means that there will be much work on laws that relate to hot-button issues like abortion or war or taxes. That’s wonderful and everything, but most of a legislature’s in-session time is devoted to topics like infrastructure and school funding, which most people find excruciatingly boring but which are frighteningly important. Without additional eyes, though, editors with a vested interest may be able to get away with quite a bit.

+++

Now that I’ve said all that, I think that a wiki may work for writing laws, if it’s done right.

• Grant read-only access to unregistered users — otherwise you’ll have chaos in no time
• Recruit a small panel of editors — respected community leaders, some elected officials, and a few carefully-selected reps from interested groups
• Set up an extra-wiki feedback mechanism, so that the non-editing general public could voice their concerns
• Create an explicit and distinct mandate for the wiki — “write legislation that will achieve $GOAL” — and protect pages once that goal is achieved. (Leave article histories open, though, so people can see who made what changes when.)
• Ban discussion that isn’t germane to the topic at hand. The wiki is there to write laws; it isn’t there as a soapbox.
• Get the press involved, reporting heavily on the news from the wiki and encouraging people to check it out for themselves

Any other thoughts on this?

Related posts

• Upgrades suck. (0)
• Quick update (0)
• My cup runneth over. (0)
• I swear, there’s something very Brownian going on in my head. (0)
• Bulleted lists are good for the soul. (0)