Freebasing the semantic web?

I’ve seen Freebase.com get some more attention lately as the Alpha invites continue to spread. O’Reilly posted a great walkthrough of an earlier version of Freebase.com here.

The discussion that follows in the comments is some insight into the various sides of the “semantic web” debate, and the challenges that come with organizing so much data. The holy grail is not just categorizing or labeling all of the information, but knowing the relationships between it all.

Will users find enough value in the Freebase system to want to actively contribute to it? Why not start at Wikipedia as a framework and add the relationship layer to the existing site? What about the existing social/inter-personal relationship layer of Facebook and adding other layers on top of that (the Parakey acquisition could be a step in this direction)?

This is the weakness in having individual sites try to be “the answer” to such inter-related problems. Until all of these pieces are decentralized and opened, I’m afraid we’re still stuck with a bunch of walled information gardens.

Unfortunately, the end goal of many of these efforts and the idea of the mythical “semantic web”, doesn’t exactly have a place for single-resource destination sites like Facebook, Wikipedia, or Freebase. Given that none of them want to relinquish control any time soon, we should continue to see these power struggles for a long time to come.

So who is in a good position to bring us toward a more semantic web? Out of the big guys, I think Facebook is in one of the best places. As far-fetched as it may be, if they were to open up a true Facebook API, opening their social network for use to the outside (not forcing people to play around inside), they could leverage their huge user-base and be the social network provider that’s plugged into every new service out there. Somebody will have to do it. People are dying for the “web 2.0 address book.”

On the smaller fronts, microformats continue to gain traction and OpenID is opening a few walled gardens (or at least creating consistent gates into them).

Lastly, I think web browsers are in perhaps the best position to take advantage of these evolutions. It may be no coincidence that the Firefox creators who started Parakey are now snatched up by Facebook. Look at what Greasemonkey and Firefox plugins have done to the way people view web pages. Look at what widgets, gadgets, feed readers, and the iPhone/Safari “platform” are doing to the way you consume/search/browse information from different sites and through different devices.

The semantic web might never arrive, but a semantic web may already be here.

One thought on “Freebasing the semantic web?

  1. Steven Forth

    Do you think Semantic MediaWiki is relevant in this context? At my organizatoin we are having great success in using Semantic MediaWiki to (i) build knowledge management and learning management systems and (ii) to generate organic, bottom up files that can be combined into larger RDF webs (I hesitate to call them ontologies) that we use for various information organization purposes and to explore emerging patterns of information connection. Semantic MediaWiki is designed to make it easy to add back into MediaWiki.

    Reply

Leave a Reply

Your email address will not be published.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>