Freebase.com is the semantic web

I am not an expert in the semantic web but Freebase looks like it to me. I am surprised more hasn't been written about it. It is absolutely amazing and head-spinning. I will have a hard time describing accurately, hence the headline.

For me, it seems like Freebase does what the Semantic Web guys have been talking about forever.

In Freebase's own words: "Freebase.com is home to a global knowledge base: a structured, searchable, writeable and editable database built by a community of contributors, and open to everyone. It could be described as a data commons. Freebase.com is enabled by the technology of Metaweb, which is described at www.metaweb.com." (from Freebase FAQ)

A mental image for Freebase that works for me is that it is a lot like Wikipedia with the major difference that instead of free form text and images that you see in Wikipedia articles you see semi-structured layouts in Freebase.

Semi-structured, that is, the information is structured via types: this is a person, that is a computer, this other thing is a country, etc. So for example, in the entry for BlogBridge (type is Software Product) , you see a description of the software, just like in Wikipedia. But now, in the section that says what platforms the software runs on there is a list of entries of type computer operating system. So when I am editing the entry for BlogBridge, and I want to correct or add to the list of computer operating system it runs on, Freebase already knows the list of known operating systems.

Interactively on the web, it's quite interesting, but it really makes your head spin when you contemplate what it means as an internet API. I mean we have specialized APIs to look up books (Amazon) and CDs (CDDB) and many other things. But here's a unified and communally editable pool of API accessible information.

Check it out. It is worth studying.
Posted on November 23, 2007 and filed under Uncategorized.