Raw

Nova Spivack features on the latest podcast from Paul Miller of Talis, great stuff. I thought his ideas on groupthink and collaboration were particularly of note. When I read some of Nova's futuristic posts a very skeptical part of my brain kicks in. But when you listen to what he's worked on in the past, it puts things in a very different context - his speculation is far from idle. (It did amuse me that where Nova talks about $big companies in his intro, I mentioned our chicken shed). 

The good Mr. Beckett announces Raptor 1.4.15 with more GRDDL along with an online service which uses it : Triplr : " Stuff in, triples out"

A tool which uses the Python bindings of Raptor/Redland also has a new release - MOAP : "a swiss army knife for project maintainers and developers". 

In Nova's podcast he talks of superpeers on the Semantic Web, which sounds a very likely scenario (the forthcoming systems from Metaweb and Radar Networks being examples). It's my opinion that only a tiny proportion of the web needs to have rich Semantic Web capabilities for the web as a whole to gain significantly. A handful of superpeers could make a huge difference. But alongside sites and services that look much as they do today and the superpeers I imagine there'll be quite a lot of services and tools in between, many built to do just one or two fairly specific (often domain-independent) jobs. Can't think of a good word for this kind of peer, but Triplr is a good example, another is PingtheSemanticWeb, there are already quite a few more. These reflect the web as a distributed platform, allowing applications to be built as loosely-coupled composites of lots of smaller parts. The web as a whole is already an intertwingled mix of large and small (i.e. fractal), it seems reasonable for this to continue to be the case as it handles more data. 

While you've probably got to have a screw loose to find Semantic Web technologies compelling in themselves, applications built with them really ought to be. So its good to see Jon Udell's being drawn Like a moth to the Freebase flame. Another good sign is that he's asking questions that should be straightforwardly answerable:

I’d like to be able to answer questions like: "Who among the Helix collaborators is also working on .NET projects?"

...

I’d love to declare once that that these two blogs are related to me, then ask Technorati and a bunch of other services to refer to that relationship.

The first part should be expressable as a SPARQL query against FOAF/DOAP data.

I think the second thing should be expressable in RDF following the style of FOAF blogrolls. Not sure, but maybe this could be done with the hAtom microformat (which could be GRDDLed to FOAF), something probably needed for Technorati to pick up on it.

Wish I had a bit more time, it'd be interesting to see how well the current tools line up against Jon's requirements, from data creation through integration to presentation. I wouldn't be surprised if the whole lot wasn't already covered, although the presentation part might have to be covered by something generic like Tabulator. 

Anyhow, as Ivan Herman shows in his latest slide deck on the State of the Semantic Web, things are coming along nicely. 

@en