wikipedia and the laptops
by Christopher Blizzard
Andy Carvin is talking about the fact that we’re going to try to include some content from Wikipedia on the laptops when they ship. It’s pretty cool that this is public now.
The trick becomes – what content? What set of knowledge do we include? Some subset, different on each laptop? That’s hard from a manufacturing standpoint. But it would be great if we didn’t just have access to wikipedia in its current form, but could actually use the local laptops as independent storage for a subset of articles. Remember, a lot of the laptops are going to be put into the world in bandwith-limited places and they only have a few hundred meg for storage. So using a cooprerative caching mechanism might be interesting. Don’t have a page in your local cache? Ask a laptop around you. Then you go to the interweb to find more content.
[...] Vía: Christopher Blizzard, Andy Carvin En: Wikipedia, olpc — Agosto 4, 2006 [...]
Presumably if the objective is to get masses of Wikipedia entries onto a small hard disk, you could forgo a lot of the markup and styles.
After doing this, and using some kind of gzip compression (or alternatively zip/rar/etc), at the very least I’d expect that tens of thousands of documents could be bundled.
Of course, you’re still right that it’s difficult to make choices, but presumably the fundamentals of education would be a great start, followed by anything relevant or useful to living in the target area.
p2p! p2p! :)
Luis is on the right path, I think. The wonderful thing about the laptops is that we can make an assumption about what software is on the laptop and they can act cooperatively in order to make a better experience for any individual user.
Would it be possible to set up some sharing network?
When one person queries for a wikipedia article, if first check his box if that is there.
If it isn’t, the request is forwarded to the next box, and so on and so forth, until the request is served?
This is interesting… “searching” becomes a state of being, rather than just an action that begins and completes. E.g. a kid might say “I’m searching for articles on sharks” and this fact would be known about him as he moves through various OLPC networks. People can then offer him the articles he seeks (presumably through some automated system, but possibly manually… they might just contact him and explain something).