Metaweb‘s been working on a semantic search routine called “Freebase” that seeks to provide information from cross connected databases all over the world. NYT Article.
My understanding is that they want a simple, natural language search engine for people which will then access a huge network of data they have assembled from existing sources. Then users will be allowed to tag and add to that data, creating even more detail for the database.
Tim O’Reilly notes that Freebase:
… turns its users loose on not just adding more data items but making connections between them by filling out meta tags that categorize or otherwise connect the data items …
So, why is this better than Wikipedia or DMOZ? It’s got more data sources, will be easier to use, and hopefully won’t suffer from the many insufferable editors and participants that plague other user driven social media like DMOZ and DIGG. However I think you always need to be cautious assuming people will participate in these projects as intensively as appears to be needed here to make this grow and gain popularity. Maybe I was missing the point but I got bored with Del.icio.us after a short time and did not feel it was creating an infrastructure that would be all that helpful to me, though I certainly see how integrating tags into search will be helpful in the long run.
Isn’t everybody getting tired of working for big, for-profit projects by helping them categorize, rank, index, and detect spam? Where’s the project that lets me do whatever I want on my own terms (write, surf, learn) and then automatically integrates that activity into the indexing and distribution processes?
Pingback: FreebasePortal » Blog Archive » Here comes Metaweb’s Semantic Freebase
We are using Freebase’s data dump at our website to showcase our technology in making it easier to search for information in Wikipedia. Have a look at our about page (scroll down for the Quick Demo): http://goo.gl/f4wKm