Re: Linked Data Mash-a-thon

I for one vote for
"John and Juan's Data Jamboree"


Less name choosing / bike shedding people! Let's pick some datasets to
mesh and start generating ideas...

I'll start:
1) http://wines.doconnor.user.dev.freebaseapps.com/index

Wines is a simple freebase application which looks for things typed
with Wine and does basic inference.
You have "Shiraz" in the name? You are probably made from Shiraz
grapes -> write it back.

This could easily be done with SPARQL instead - good example of a
dataset being used to make itself more complete.

2) Bestbuyers: Find friends from my FOAF who are located closest to a
Best Buy store which carries a product I am interested in (so I can
ask them for my birthday present!)

3) Oh-no Ohloh:
Mesh up bugs (baetle / other vocabs -
http://code.google.com/p/baetle/)  with Ohloh
(http://rdfohloh.wikier.org/) or DOAP to give me an idea of how buggy
a piece of software is before I use it.

4) AusGovBase - Mesh up Freebase RDF regarding organisations in
Australian government with AGLS data (there's a kind of horrible
serialization that predates RDFa in widespread use) which describes
what data is available by individual organisations.

Use case: tell me about the government entity I'm about to try to get
documents from, and what they've already got online.

Received on Thursday, 17 September 2009 08:20:03 UTC