Open Tech 2005


Open Tech 2005 was last month, so maybe I should write my notes up. All the NTK-instigated events that I’ve been to have had very different styles, which works well. It still feels, in a good way, like people are making it up as they go along.

Danny O’Brien gave a great talk about privacy, fame and wealth. He didn’t touch on much from Public, Private and Secret, but there were some intriguing glimpses of the kinds of tools and techniques that professionals – and high-school clique leaders – are using to manage dealing with large numbers of people. My biggest fear is that some people, other people, are getting to see the finished, totally polished versions of Danny’s talks with all the conclusions highlighted and all the thinking already done for us.

I think you could be hugely successful just by, every five years, taking Ted Nelson’s (Wikipedia entry) all-encompassing view of data and implementing as much as is practical with current technology. Bi-directional links too hard? Drop ’em. Transclusion too complicated? Encourage everyone to publish under Creative Commons licences. His model of application development seemed, to me, to suggest projects like Apple’s Core Data and, everyone’s favourite all-encompassing metadata model, RDF, but the ZigZag idea of tying pairs of chains together, DNA-style isn’t something that I’ve seen pursued so ruthlessly. (There’s an introduction – with diagrams – and LtU hosts slightly more opinionated discussion.)

Francis Irving spoke, along with Chris Lightfoot, about MySociety’s projects. It was interesting to see people responsible for so many successful projects explain how hard it is to get new people involved with website projects – there’s no simple codebase for people to download and inspect, so it’s hard to foster a sense of potential involvement. (People sometimes regard the Feed Validator as a black box – maybe writing a simple HOWTO to get people running it on their own systems would help?)

Gavin Bell talked about social documents, topic maps and out-of-copyright literature. The idea of annotating books and linking, for example, names and places with the real world really does seem like something that enriches both sides of the equation.

Simon Willison gave an excellent product pitch for Greasemonkey, moving from subtle styling changes through to Book Burro. Looks like the important security fixes are almost into the shipping version, too. (Has anyone noticed that Google now turns all search result links into trackable, via-Google URLs in the ‘onmousedown’ method? There’s got to be a Greasemonkey script for that, right?) (Update: David Schlossberg noticed the same thing, and includes it with other annoyances and fixes for deliberately obfuscated HTML. Thanks to Joe Hall for letting me know. There’s also a Greasemonkey introduction, by Mark Pilgrim, up on O’Reilly.)

Also notable: Jeremy Zawodny starting his talk by taking photographs of the audience, who were in return taking photographs of him, most of which are now tagged and available on Flickr, which he then went on to talk about. It was like some creepy hyper-modern cross between autistic flirting and our very own home-made surveillance state, and I am partially ashamed of having participated.

At this rate, next year’s event will be a huge, invite-only corporate affair, arranged and held via wifi, in distributed fashion, exclusively from pubs. Looking forward to it.

(Music: New Order, “Crystal”)
(More from this year, or the front page? [K])