Prototyping Weeknotes #52
It's my first time writing the team's weeknotes. Hello!
Monday's stand-up kicked off another productive week on Watch Later. Tris and Theo continued to work on user stories for Watch Later, distilling numerous potential use cases down to five representative journeys, and applying Ideo's user-centred design toolkit to the process, to see if it was a useful addition to how we work. Frameworks can be sometimes be restrictive, but they are also useful in letting us see what we've missed.
Monday was Mark's last day in the team, as he goes off to Centre House in West London to continue his industrial traineeship in another R&D team. Akua baked another of her amazing cakes to send him off, and we were sad to see him go. The cake helped us get through it though.
Unfortunately, some last-minute tidying on Audio Sync revealed reliability issues which have led us to put it on hold for now, but watch this space for an update. Akua bravely stepped out into the freezing cold to ask a few shoppers on Oxford Street about their television viewing habits, and came back with some good results for Watch Later.
On Wednesday, a few of us went to watch Mark, Chris N and Duncan present some of their recent work at an R&D Lightning Talks session, which allows presenters from around the department five minutes to describe projects, approaches or the technologies they're using. Mark showed Audio Sync, Chris N talked about his work on MapReduce, and Duncan gave a short introduction to Sinatra, our team's preferred Ruby web application framework.
Olivier continues to settle in, sorting out w3c admin and hijacking (his word) a few meetings to understand better what the team does. We've asked him to report back to the rest of us as soon as he knows.
George and I sat down on Wednesday to kick off a complex bit of work: figuring out the business processes required for running RadioDNS infrastructure for a number of partners. Chris L and Sean have been working together on the RadioTAG spec, looking at authentication and authorisation protocols and some possible OAuth-based replacements. They met with Andy Buckingham of Global Radio on Thursday, and are now working on proposals to modify and improve the spec.
We're also thinking about why and how audiences might want to tag radio in the first place: I'm digging out existing audience research, speaking to colleagues in Audio & Music, Sean has been thinking about an earlier, similar project called Bookmarker, and Chris L went to meet a producer at Radio 1 to see how programmes are segmented and recorded.
This project is fast becoming a radio corollary of Watch Later, and we discuss with Tris, Theo and Vicky how best we could test both prototypes with focus groups. The Watch Later team agree on Thursday that they should create a low-fidelity, but working, prototype as soon as possible. It feels like this is the kind of thing you have to live with, so we need something that works but isn't necessarily fully featured or ultra-reliable.
They get to work straight away: Theo working on some simple user interfaces for Duncan to implement in Sinatra and host on the cloud so we can see them working on different devices quickly, and Chris N looking at building a searchable database of programme schedule metadata, using Apache Solr.
Thank you for your comments so far on how you remember TV programmes. If you have any more thoughts on that or what you might expect from being able to "tag" radio, please leave them here!