Distributing incomplete information
Aug 10th 2006Wallabynerd alert
I’ve discovered, in recent months, the fun of contributing to Wikipedia (WP). The chemistry section (analytical chemistry in particular) is sadly incomplete — there wasn’t even an article on beakers until recently, for instance. And since I like writing, and I like chemistry, and I like digging for new stuff to learn about, it’s been quite a lot of fun to combine those three things into useful contributions.
The problem is that Wikipedia’s success makes adding to it more difficult. Suppose that there’s a very short article (a “stub”, to use the WP term) that I’d like to expand — say, “Molisch’s test” for the presence of carbohydrates. It’s something I know a little about, but not very much, so I start looking for more articles about it online. Google for “Molisch’s test” and what pops up? About 90% of the first 30 or so pages turn out to be … Wikipedia and sites that mirror Wikipedia. These sites are very well linked to, which means that Google thinks they’re useful and important… except that in this case, they’re actually pretty poor choices, compared to, for instance, someone’s online lecture notes from qualitative organic lab.
This isn’t even considering the nuisance value of sites that take Wikipedia content from a year ago and stuff it into a page full of banner ads, or useful-to-some-but-not-to-me sites that also ‘feature’ a ‘dictionary’ of unrelated stuff (cribbed, again, from WP) to boost their search engine results. (Laborlawtalk.com, I mean you.) All of this is also allowed, of course, under WP’s “free documentation” licensing, so long as the sites acknowledge the source and link back to WP. Which most of them do. But the broad distribution of *bad* information makes it increasingly difficult to find *good* information… which should come as no surprise at all.