I only just stumbled across this the other day, but my alma mater Columbia University has its own unofficial wiki site that’s actually quite good. This genre of wiki site — university history and alumni wiki — has had decent success, as the community profile overlaps with Wikipedia’s quite well.
You can even see it in the design of the pages, as they effectively use MediaWiki (the software that drives Wikipedia) and more advanced Wikimarkup features such as categories, image galleries, tables and templates. Usually small/medium wikis find it hard to balance open registration and contribution, while resisting spam and vandalism. The following chart shows that university sites have had different levels of success with that problem.
Stats about university wiki sites
See: WikiCU
Where’d you get the graph (or the data for it)?
I should have linked it, but it comes from the WikiCU:About page. It’s not clear where they in turn got the stats though.
I don’t think that total pages/legitimate pages discrepancy has anything to do with vandalism – I’m almost certain the numbers came from each wiki’s “Special:Statistics” page, which lists both “total pages” and “legitimate content pages”. The latter number doesn’t include stub pages, redirects and talk pages, among others, which usually leads to a substantial undercounting.
Yaron, thanks for the info. It does seem the signal/noise ratio is mostly due to massive number of stub articles, especially on the Case Western wiki. Kudoes to you folks for letting WikiCU grow organically, and not filling everything in with one-liners.