After I mentioned Freebase in a previous blog posting, I was contacted by the company and invited to check out their product. This isn't the first time I've been contacted by a company after blogging about them - the guys at Eyejot did as well - so someone out there is reading! Worth bearing in mind if you want to try out something that has invitation-only access.
Anyway, I've been busy and the invitation got put on the back burner. I've finally had a chance to check it out, and can now share my findings.
Basically the idea behind Freebase is that we can and should sort all the world's data semantically. Imagine all the Wikipedia data organised by type, (in fact Freebase's starting point for some topics is data pulled from Wikipedia), where that type and all the related data can be cross-linked with data of a similar nature. Imagine if everyone's tags were properly organised on every site on the web.
On the surface you get more powerful ways to search and mine data; for example, a search like finding all companies in London in the design industry with 100+ employees would become very straightforward. But the real kicker is when people start developing applications on top of this database. We've had mash-ups for a long time, but that's nothing compared to the potential of this resource.
Freebase has an API open to those who are taking part in the alpha phase, which means those developers can get their hands dirty with this stuff pretty quickly. I think you need to be logged in to see the stuff that's been built so far (and to be honest it's not that sensational as the database is still quite young), but the potential is there. Blog about the product and you might get a look in! Or ask me for a demo.
Jim O'Reilly has had a play with Freebase and has some screenshots on his blog posting. Could it become addictive? Perhaps in the same way Wikipedia compelled people to contribute. Whether it can become the starting pistol for Web 3.0....it's hard to tell. But in the meantime the excitement builds as we move towards beta and (presumably) a fully Open API which can be used by a much bigger community. I shall be watching with interest.