Egocentricity is for life, not just for Christmas
You’ll have to excuse the egocentric nature of this post - it’s a way for me to thank and respond to the various people who have been linking to me recently.
Austin Burbridge of Cinema Minima gets it. The Web is a cacophany of human voices, but a monoculture of machine readable voices.
Gareth Rushgrove and Ben Ward have different ideas as to how GetSemantic should proceed. Ben thinks it should do something new and unexpected, while Gareth thinks that RDFising existing formats is fine. Both and neither are what we’re shooting for. Sometimes the GetSemantic project will need to take existing formats and RDFise them. That doesn’t mean that we are going to take existing formats and change them - it means that we’ll use technologies like GRDDL to turn them in to RDF, and we’ll come up with tools and ontologies to make this easier.
When we are creating new formats, then we won’t necessarily be saying “there is only one way of doing it!” We’ll make a suggestion, based on what users have to say and whether we’ve had much luck implementing it in a particular way. But, we’ve basically got an open toolbox - eRDF, RDFa, microformats/semantic markup and GRDDL. We can construct whatever we want using those tools. And if none of those work how we want them, we come up with something new. The key idea is that so long as the eventual RDF is consumable in sane ways, the representation isn’t that important.
The other important part of GetSemantic is that we’re not going to be strictly tied down to the microformats “it’s gotta solve a problem” adage. Yes, it’s useful in advocacy to say “this solves problem x”. But, as Tom Hughes-Croucher says, a lot of the technology we’re building stuff out of today has little practical purpose. A lot of possibly interesting avenues for research and experimentation are closed off if you say “only practical research, please”.
With regards to GetSemantic, Andy Mitchell liked “bastardisation of the “long tail” buzz Microformats are the fat (common) parts of the tail; where as the Semantic Web is the thin end of the tail - for the many small cases where you need the additional flexibility RDF gives you”. It’s fun, but there’s a serious point - with eRDF and/or GRDDL, the barrier to entry for new format development is much lower than it has to be for microformats. Microformats, using semantic class names without namespaces, should be a conservative effort - avoiding confusion in the marketplace. We are not so tightly bound! This means we can play about with formats, and if they don’t work, no problem.
I do like Gareth’s description of me as being “card-carrying”. When they find me lying unconscious in the street, they’ll open up my wallet and find cards for the NRA, ACLU and the RDF. “;->”
On to Twitter. Ian at twopointouch has a write-up of my Twitter Tube experiments, and the question was asked “what’s the point?” Basically, Transport for London provide a similar service to the one I’ve built. True. What makes Twitter interesting from a developers point-of-view is that it’s basically a free SMS layer which you can build on. The Tube example may not be the most exciting, but it is useful.
The reason I chose the Tube is because of user need - I’ve tried the London Transport SMS service and it didn’t work for me. I tried two or three times to set it up and got nowhere. Eventually, I just gave up.
Twitter is far more user-friendly than the TfL service, in my experience. It is also far more ‘remixable’. I’d like to have it so that we can use user contributions to send data around. That’s more of a pipe dream until Twitter extend their API in certain ways.
Eventually, we’ll have a system that’s far more useful for Tube planning because we could have a script that could turn updates on and off for you based on where you are going. This is only a first step. Once developers get a chance to play with location and presence-aware mobile data neworks, all sorts of fun will start happening.
I am taking requests for more Twitter transport mashups. The more I look at the websites for metro services and so on, the more I react with disgust. So many of these sites could be improved! I’m not sure what I’m going to do next. Things with XML feeds come higher up in the list than services where I have to scrape some godawful website.
More thanks to Aral Balkan, Conor O’Neill and all the del.icio.us kids, among others, for the links and nice comments about Twitter Tube Tracker. I hope that other people find it useful. So long as I know when the blasted Circle line isn’t working, then it’s mission is accomplished.
Okay, self-indulgent ego-stroking over. Need to get back to Real Life.
