BBC R&D

Posted by Theo Jones on , last updated

These are weekly notes from the Internet Research & Future Services team in BBC R&D where we share what we do. We work in the open, using technology and design to make new things on the internet. You can follow us on Twitter at @bbcirfs.

Last week saw projects drawing to a close and new projects starting out on their journeys of research and discovery. Oh we were all watching the Eurovision Song Contest, right?

To demonstrate the flexibility of the ViSTA-TV Live Dashboard and to get into the spirit of Eurovision, Libby spent some of her precious Saturday tinkering with it to display real-time tweets containing '#eurovision'. The code was originally designed to show the numbers of people watching IPTV channels in real-time, but it's adaptable to other streams of data.

More from ViSTA-TV 

Chris Newell and Libby have submitted a paper to RecSys 2013, the annual Recommender Systems conference, about our client-side recommender engine and the two applications we've built with it: Sibyl and Infinite TrailersAndrew N and Chris L have been stamping out the bugs on the dashboard and open-sourced it to the world, while Libby made the prototypes work offline and drafted the end of year review presentation. Phew.

FI-Content phase 2

On Monday a bunch of us (Chris G, Chris N, Joanne, Penny, Sean, Theo, Libby) visited Lancaster University to kick off the collaboration on the second phase of the FI-Content project. We will be working with them to run a few experiments on the university's video streaming research platform, to study how user data can make features more personal and relevant. First off we'll be looking at 'viewing history' to see how the feature influences what people chose to watch.

In a different part of the FI-Content EU Partnership universe, Barbara has been in administration mode running telephone conferences, action planning, milestoning and reporting for the month.

ABC-IP

This week the ABC-IP project draws to a close. Yves and Dominic are preparing for the final TSB review next week and will summarise the project's achievements in a blog post. Yves has also written a paper for the International Semantic Web Conference about the project's last prototype - a visualisation of archive content related to current affairs. He also found time to submit a proposal for a poster at the next IBC. 

World Service Archive prototype 

Tristan and Michael S have been thinking about three keys areas of development for the World Service Archive. Firstly how to build up our community of active users; secondly thinking about the tools we'll need to make available for them to edit more of the programme data and lastly ways of measuring and displaying the activity in the archive and the improvement in the metadata from people editing and adding information.

Chris Lowis has re-enabled the editing features for programmes with unavailable audio, made numerous improvements to the search results pages and built a feature that removes programmes with rights restrictions. Pete and Andrew N have been working together to design and implement user interface improvements requested by our users and Yves has woven his magic in SPARQL and given a presentation on tagging the World Service Archive at the Media Management Seminar held at the Dutch Institute for Sound and VisionMichael S has been industriously visualising the co-occurance of tags with the d3.js javascript data visualisation library, take a look at his results.

Snippets 

Rob has been foraging for new datasets to include in our projects, like Snippets. He's been talking to Sports folk about how they use archives in their production process. Matt Haynes has been researching audio waveforms for the next work phase, while Andrew Wood and Joanne Moore have kicked off a need-finding exercise to define the user requirements for the interface.

Other projects

COMMA held its first catch-up with the project partners on Monday and Matt has been busy preparing infrastructure. On the Mood front, Denise has been evaluating automatic time-line mood classification for a number of different programmes, looking at the results for extreme moods. Matt helped out the News team with metadata ahead of the "News Festival" and is looking into metadata extraction from Wikipedia dumps as part of his 10% time. Barbara, Andrew N and Theo have been preparing for a workshop to prototype enhanced radio experiences. And finally... Andrew N's tweet owl prototype has (de)ceased to function after a soldering snafu.

Interesting Links

Data visualisations from this year's CHI conference in Paris - curated by NYU-Poly's Enrico Bertini

Strongbox made from DeadDrop - a service for secure anonymous communication is an ever more uncertain world

Get your linked data fix here - Linked Data at Globo (lots of JSON-LD!)

Something for archivist music lovers -  Ian MacKaye's Library of Congress Lecture

More fascinating things being modelled and represented on the web:

* News storyline ontology

* Ontology of Quranic Concepts - incredible piece of work

And back in the physical world:

* Review of the current state of 3D printing and forthcoming challenges

* "Made not Manufactured" photo series found on http://cornishworkshop.blogspot.co.uk/