Google Base

From WhyNotWiki

First found out about: 2007-06-12 13:47

It looks like all it is is a way to associate structured metadata ("attributes"/"details") with a resource/document, so that users of Google search technology can search for it using that information.

That in itself is not a really novel idea, but of course Google is mixing it up a bit. It looks like its main draw will be that it has an easy-to-use interface that "normal people" (end-users) will be able to figure out quite easily.

Also, it handles all the publishing for you -- so it doesn't have to already be published on the Web. I don't yet know how well it works with information/resources that you've already published on the Web.

My main concern is that you'd have to go through their interface to create and edit this metadata. Hopefully, and probably, however, they will release an API that lets you edit it from other applications.

To me, it seems that metadata should be stored with the resource/document that it describes, rather than in Google's database. I don't really like how they're trying to turn this backwards and force (encourage) users to create metadata primarily through their site and then have the only way to retrieve this metadata be through Google.

I would rather if they encouraged people to publish their data as RDF documents on their own web sites (on the users's personal/organization's sites, not on Google's sites) and let Google index and scrape the structured data found there, sure, but not be the primary store of that information!

In particular, Google seems to be taking the exact opposite approach as Piggy Bank[1] and the DIG (Decentralized Information Group). Google wants to make all information centralized using their services/servers. DIG (and I), in contrast, wants to decentralize [the world's] data.