The Transforming Web

The carpet has been pulled out from under our feet and the new web is a very different place to what we were used to. On Friday morning at Webstock, eager conference goers were subjected to several hours of high level discussion from Simon Willison and Tom Coates on OpenID and decentralised social networks and The Transforming Web.

Simon's talk was fantastically eloquent and measured. From this, and from his Django talk, I was impressed at how well he managed to explain relatively technical and complex procedures to a room that wasn't totally full of techies. I'm neither an expert on OpenID, nor particularly interested in it, but I still gained a lot of useful stuff to think about. In the second part of his talk, Simon built on this foundation, discussing some of the problems of OpenID, leading in to an insightful exploration of the current generation of social networks. This is where things really started to get interesting. Given the current proliferation of online services that many of us subscribe to, how can we accumulate our identity data in a way that's easier to manage and allows us to build meaningful connections on the open web? I'm very interested in how we can better facilitate decentralized social networks - while I remain <a href=/closed-social">unconvinced about XFN</a>, I do think that the rel="me" relationship has an important part to play in this emerging ecosystem and Simon did a great job of explaining how this fits in with other existing standards. 2008 could be the year that we finally see open social networks breaking free from walled gardens like Facebook.

Tom took up where Simon left off, and dived straight into an entertaining and thought provoking journey through the landscape of data. Anywhere that the network can touch, data is being distributed. Thus your territory can be anywhere the network touches. We're close to a watershed moment in history, where this web of data is about to break through into physical, spatial contexts. The age of ubiquitous computing is here. Pages are dead. Long live the network.

Tom went on to explain how we need to shift our design focus towards recombination. Playing well together means not reinventing the wheel - taking existing services and platforms and building useful and meaningful new services on top of them. He asserted that every new service makes all existing services better - I'm not exactly sure how or why that's correct, but it's an interesting way to view things. Information is spilling out of so many everyday objects and contexts - we can capture this in streams of data and make it useful to people. The solution to the problem of how to navigate through too much data is to use more data. A good example of this is the way that Flickr allows searching and filtering photos by properties of the exposure itself. I spoke to several people afterwards who found this mind-blowing - they had no idea that most digital cameras encode this data in every image file they save.

As a little addendum, Tom dropped in a useful note about information architecture. On todays web, the traditional 'top navigation' has become a mere jumping off point and no longer defines the backbone and structure of the content organization. Hierarchic classification is a dead end - we're finally coming full circle, back to exploring the web using web-like navigational structures.

Putting all this together isn't a job for a bunch of Foo camp geniuses or large corporations. The open web is something we are all responsible for and we are all contributing to its future success or failure. You can laugh at me for being an idealist, but I'm excited about these possibilities.