For me, BarCamp Leeds started with Mark Sailes‘ session on TV 3.0. Mark’s session was one of my favourite sessions of the day and centred on his contention that TV needs to enter a third phase (1st phase: black & white, 2nd phase: colour). TV 3.0 is all about audience participation and analysis (this linked in with one of Tom Smith‘s irritations from his talk, that viewers can’t own and tailor their TV programming), resulting in rich metadata around programmes and films. Mark suggested one mechanism could be a collaborative highlighting and commenting method for
working playing with video streams, some mechanism like Flickr Notes to add comments, a way to tag themes like “lightsaber fight!”, “Vader”. You would then be able to choose different ways of consuming the content, “I want to watch all the light saber fights involving Vader” (Mark’s a Star Wars fan).
As you can imagine at a tech unconference, talk inevitably turned to the technical considerations of collecting this metadata. Copyright is an obvious hurdle, the MIAA and chums are not well known for their willingness to allow access to their material for audiences to play with. We discussed automating metadata collection for video streams, and both facial and voice recognition were mentioned, but how would these cope with accents and false beards? One potential opportunity was identified in TV Heaven from the National Media Museum, TV Heaven seem to operate a bit like the UK copyright libraries in that they attempt to collect all broadcast material… a metadata collection and search facility would be great for them.
The talk touched briefly on Tom Coates musings on URLs and addressability. I can see some definite challenges in devising a scheme of URLs to allow you to succcessfully address specific points or spans in video timelines, and to specific points or areas (maybe moving through time) in the image. If this could be overcome, and broadly accepted, then it’d be a major hurdle overcome in adding metadata of all kinds to video streams. I wonder if a universally accepted address could be a way for the media/copyright owners to derive business value from sharing their content in this way… their freely shared (yes I’m dreaming) control of the address within a universally used could allow them to extract commercial value from their content, in the same way that a company’s domain name gives people a universal point to access their services and/or products.
One of the reasons this was such a good session was Mark’s willingness to bring a half-formed idea that he was enthusiastic about, and get a good discussion fired up. As he said, everyone loves some kind of television, so it’s easy for people to see something they’d be motivated to work on in some way… a suggestion that Babylon 5 and Stargate would be spectacularly well documented, and EastEnders and Corrie less so was refuted by some of the women in the room, apparently women often ring each other and chat (who knew!?). This kind of conversation around TV content can be seen as metadata, and a rich source of information. Imran Ali mentioned that when he and Mark worked on this at a previous company, they’d had access to IM logs and would see definite spikes in activity around popular programmes, so maybe the male skewed shows have a similar phenomenon of chat online, rather than by phone. Overall the room agreed that geekage isn’t unique to sci-fi, there are Emmerdale geeks, horse racing geeks, news geeks, all of whom are just as passionate, the challenge would be making the interface accessible and attractive to these non-traditional geek audiences.
It’s great “up north”, I feel I’ve landed in fertile soil and that there’s a lot to be excited about… more on BarCamp Leeds soon, I’ve got to go and cook now.