Prototyping Weeknotes #81
We put our engineering house in order with the housekeeping work last week, and everyone's now back on project work. We kicked off some new projects this week, including our final development iteration on the and some research around a notifications framework. Laura, a design trainee, joined the team this week and will be with us until the end of October working across a number of our projects. The world also sadly said goodbye to , the co-creator of both the C Programming Language *and* of UNIX, who died on Wednesday.
It's Chris L's first week on the ABC-IP project; he's been getting in to Python and looking at an idea Yves had for segmenting audio based on the clustering of dbpedia tags assigned via the automatic transcription and tagging code he's written. Meanwhile, Yves has been writing up about automated tagging of speech audio, and getting some new interesting evaluation results, giving a clearer idea of what processing step has what impact on the results. Roderick has been working on how to decouple the speech recognition and tagging processes into a network of interconnected systems so we can efficiently and reliably process all 500TB on the cloud.
The most requested feature from users of the Programme List is some sort of alert - to remind them of programmes (but also to prod them to use the site). Sean's suggestion on hearing about this is to use it to build a notifications framework that is extensible and flexible enough to be able to be used across future projects. Theo, Pete and Laura have been doing some initial thinking around the UX and putting together a UX brief whilst Duncan and Dan investigate existing scheduling solutions and assess their suitability.
The News Linking work has been wrapping up and on Tuesday the team got together to do a retrospective. It was an interesting project - a good research question as a starting point and something that was data-driven, resulting in a research report and a prototype. And with a possible spin-off research project on the side. What worked well? Having an interested stakeholder all along, good feedback loops between data-driven prototyping and design, good team comms and having a flexible tech lead role based on skills and the stage of the project. What could we have done better? Presented more information about what we tried but didn't work (as well as what worked). Earlier in the week we also had some more good feedback from the editorial side of News who particularly liked the resulting prototype.
We're entering the wrap-up phase of the News Follow work and Theo and Kat have been ploughing through the transcriptions of the user testing. They're going to be pulling it all together and summarising the project next week so we can transfer our learning into future media.
Our final phase of development on the LIMO framework as part of P2P-next got under way this week. The project team (Chris N, Dominic, Andrew, Sean, and Theo) reviewed the work done to date and then prioritised and planned what we do next relating to HTML5 video, live streams, and event synchronisation. Related to this, Sean has been looking at the thorny issues regarding synchronization between multiple timelines which has come up in both the LIMO work we're doing and George's idea for RadioSYNC. Andrew, meanwhile, is readying himself to read all 85 pages of the HTML5 Video specification.
Steve B, our boss, along with Libby, Vicky and Chris from Audience Experience joined us for our team meeting on Thursday. Libby showed us the . Afterwards Barbara and Tristan got together with them to talk about generating portable profiles, user data, visualisation, good user recruitment agencies and EU projects.
Duncan has been working on a light weight github/gitweb sinatra app and watching proudly as interest in his 91Èȱ¬ QR codes grows; so far they've been used on 91Èȱ¬ London News, the Schools project and possibly on the new Radio 1 homepage. Kat has been doing lots of RadioDNS work. Sorting out slides about RadioTAG for TechCon at the end of the month, writing up a blog post about the RadioTAG trial and pulling people together to scope some possible next phases of work.
In other news, Yves and Nicholas have been hosting the second face-to-face meeting of the this week. It was a dual-site meeting with the MIT in Boston, MA. The group is working on updating and tidying the set of specifications known as RDF 2004 to reflect current usages and deployments. Olivier has been at talking about the 91Èȱ¬ and open linked data. George and I both had a week of administrivia so the less said about that the better and Tris ends his week at a workshop about the future of "social".
Interesting links
- : Earn qualitative achievements for the code you write, your skills, and your active community participation.
- : Why music ID resolution matters to every music fan on Facebook
- : A systematic approach to interactive visualisation.
- Channel your inner designer .
Comments Post your comment