Image via WikipediaEver wondered what you might want to use a triple store for? Well, wonder no more. Toby Segaran has put one to good use to essentially capture his personal stream for the month of September. That stream includes phone records, email history, contacts, calendar and Facebook friends. Of course, being Toby, it’s not just about getting the data in there, but also doing something with it. At this time, it involves some clustering, and seeing the groupings of the connections.
Toby’s right. We have lots of applications which, rightly so, are optimized to do a certain task well, but few ways of capturing information across applications. In a world where you can output data streams in XML or JSON, or RDF, or whatever easily parsable format we choose, we need approaches that allow us to reconcile all that information. To some extent that is what something like Friendfeed enables, but not in the programmatic, data-mining ready approach that Toby has adopted. As he writes
I now have code to keep my triple-store synced with my friend network, my contacts, my phone records, my email and my calendar. I can construct queries across all of this (who did I forget to call on their birthday? Who have I seen recently who went to Stanford?). I’ll be sharing this code at some point, but I want to see how far I can take this. I’m also interested in hearing from anyone who has tried similar experiments and wants to collaborate.
I love when people do seemingly obvious and simple, but in reality non-trivial things like Toby just did. The great thing is, as our APIs get better at making information available, as we move towards more linked data, and platforms (like Talis, Freebase and the likes) that allow us to put data in, and then query across data streams, we’ll all be capable of doing what Toby did, and not just with our personal information streams, but other kinds of data streams, like data from multiple research projects, or assay types. Not trivial, but definitely doable given the right data types and information content.
Practical triple stores
Toby’s right. We have lots of applications which, rightly so, are optimized to do a certain task well, but few ways of capturing information across applications. In a world where you can output data streams in XML or JSON, or RDF, or whatever easily parsable format we choose, we need approaches that allow us to reconcile all that information. To some extent that is what something like Friendfeed enables, but not in the programmatic, data-mining ready approach that Toby has adopted. As he writes
I love when people do seemingly obvious and simple, but in reality non-trivial things like Toby just did. The great thing is, as our APIs get better at making information available, as we move towards more linked data, and platforms (like Talis, Freebase and the likes) that allow us to put data in, and then query across data streams, we’ll all be capable of doing what Toby did, and not just with our personal information streams, but other kinds of data streams, like data from multiple research projects, or assay types. Not trivial, but definitely doable given the right data types and information content.