Matt’s asked me to join in here as I’m helping to build some of the new services on bbc.co.uk around ‘socialness’ which will be appearing in 2009 and 2010.

We’re all massive fans of linked data as I’m sure you all are – and the benefits as summarised above are clear.

Where is gets really exiting, is where you begin to allow user to interact with that data in cool ways: becoming fans of artists, listening to tracks, attending events. And sharing that activity with other people across the LOD cloud. Thanks to semantic web technologies like FOAF and SIOC, we can expose that data in linked-open-data ways too.

However, with personal activity data, and sharing such data on the web – we have to be really careful. We’re strong believers in the principal that the User owns their data. They should be able to see it, share it and remove it. They should also control who they allow to see it – for example, people you trust should probably be able to see more than people you don’t.

One of our challenges with allowing users to expose their activities in LOD-style ways is to explain to them the power of the consequences of their actions. If you publish that you’re a fan of something, watched something or read something – or allowed it to be automatically published – then you might have a hard time removing that data once its been crawled by any RDF client.

Publishing cool data like the guys in bbc.co.uk/music and bbc.co.uk/programmes do is just the first step – but how we deal with publishing user-activity data in these ways too is a significant challenge – one we’re working hard on, and one we’ll crack… with the communities help!

S

this makes BBC and its Linked Data Initiative even more attractive! I would like to ask you for more information on this fundamental principle, because I think that your use case should become kind of a best practice for the LOD World. Thanks again!

The short answer is that bbc.co.uk doesn’t carry advertising, so this territory is basically off limits to us. The slightly longer answer, that would relate to the controversies around Phorm, is that all the ideas the BBC’s currently exploring around personalisation and recommendation are informed by the fundamental principle that users should own their own data and be comprehensively informed about what’s being done with it, with the ability to opt out if necessary.

barna: in answer to your question, we didn’t really see these technologies as immature. Quite the contrary – the fact that the likes of Flickr and YouTube had already built category-killing offerings out of linked data were something of a wake-up call to us in the BBC. We were also pushed in this direction by the complexity of what we do in the music space. In order to pull together all of our disparate activity as a broadcaster, we have to have a system that was dynamic and flexible. It’s also part of our public service remit as the BBC to be open to the web, both in order to support the wider ecology, and to provide useful routes into good external content for our users, and this was frankly the only sensible way to do it. Finally, we had a critical mass of visionary technologists and open-minded managers that enabled us to take on a new approach.

http://www.estc2008.com/index.php/program/program-list/48-norheim-and-engels