Yep, this is a neat idea. I first saw it as part of Patrick Collison's wikipedia-iphone project (2007-ish), which could pull Wikipedia articles out of a compre

Yep, this is a neat idea. I first saw it as part of Patrick Collison's wikipedia-iphone project (2007-ish), which could pull Wikipedia articles out of a compressed bz2 archive, by indexing from article name into bzip2 block and then only decompressing the needed blocks for the article you asked for. We adapted his code at One Laptop Per Child to preload Wikipedia snapshots on every laptop: http://blog.printf.net/articles/2008/06/02/wikipedia-on-xo/ http://blog.printf.net/articles/2008/06/02/wikipedia-on-xo/ http://blog.printf.net/articles/2010/05/09/peru-olpc-and-wikipedia/ http://blog.printf.net/articles/2010/05/09/peru-olpc-and-wik...