When is it appropriate to use rdf?
Lucas Gonze brings up a good point about Anselm Hook’s recent efforts at creating a ’social engine’.
He points out that - even though this may be the one time where using rdf is entirely appropriate - as it’s designed to “pivot complex objects on shared features” - however it’s also true that in social software interfaces - it’s the HUMANS who do the pivoting.
My own feelings on this is that as much help as we can get from the machine - teh better - but don’t start drinking the koolaid and tell me about ‘automatic’ anything.
Social software is about humans using technology and no machine intervention can do anything but “juice the scenario” - making it more interesting - than not.
Writing a Social Content Engine with RDF
Anselm Hook is writing a generic RDF-oriented framework for things
like Delicious, Flickr and Webjay. He calls this type of software
“Social Content Engines,” a wording that works well. Here’s how he
describes the core functionality:
- Publish observations or ’stuff’ onto a website.
- Categorize it variety of ways.
- Pivot on yours or
others observations to discover other related topics or
persons.It’s not really software he’s building so much as a factoring or
simplification; a compact model that he can look at from any angle;
something like a simulator.A provocative aspect of his model is that he’s using RDF, which is
designed from the ground up to pivot complex objects on shared
features. This is one of the rare instances I’ve seen where RDF is
being used as RDF rather than verbose XML, and it might even be a
compelling-enough usecase to make the pain of building working
software with RDF worthwhile.My one misgiving about focusing on RDF is that most social content
engines have humans intimately involved in the pivoting, and it is the
social act of performing a pivot that provides the most compelling
content. It may be that the point is social pivoting, and better
automation as enabled by RDF is a distraction.[Lucas Gonze}
Another reason I’m real interested in what Anselm is doing - is that we’re trying to do something similar with the PeopleAggregator.
I just hate it when the same software gets written over and over again. It’s all a commodity nowadays - so why not just have some coolio open source code that we call can use?
I’d love to see where Anselm’s APIs - define the seam, entrance and exit from social networks.