Archives for November 2009
The BBC 'Archive' as managed by the Information and Archives dept is a massive and varied collection- it includes video and film, radio broadcasts and recordings from the very earliest days of the corporation, and written archives and sheet music from well before. Add to that hundreds of thousands of photographs, and you begin to get a picture of the vast collections we hold. Beyond that though is the 'metadata'! This is the crucial cataloguing information; the production notes, the cast lists, running orders, music notes, accounts even. Today every moment of broadcasting is planned, coordinated and tracked by using millions of pieces of information about the programmes, their content, and event the way we deliver it to you, the audiences.
R&D have a core commitment to support the Information & Archives dept in their mission to preserve this heritage into the digital age, and to help the work of Roly Keating and Tony Ageh in developing new ways to deliver access to the archive for all our audiences.
At Archak we tried to bring together all the people who make that metadata, and those archives, tick. And for seven straight hours we worked at inventing new ways to join these all together.
In the end we had fourteen presentations of various hacks that had been thought up, designed, attempted, and in a few cases actually built! The aim of the day was to try new stuff, so the occasional noble effort that falls short, but still shows the potential, was well worthwhile.
We had a hack that looked at recent broadcasts of Question Time and plotted them on a map- as a programme that's often heavily influenced by local concerns this could add a new and interesting dimension for browsing content of interest.
A number of efforts joined forces in a 'mashup of mashups' to give a second by second breakdown of television content, then allow viewers to 'bookmark' the show at key moments, and then presented a time line with illustrated bookmarks highlighting the most popular sections of the show, with those self same screen-grabs scaled for popularity.
R&D's own staff produced quite a few interesting proof of concepts, including '6-degrees of Bruce Forsyth' which visually linked entries in the programmes catalogue to the eponymous entertainer, and an audio hack which would allow user to mix their own sound for media within a browser. This latter technology could be invaluable in future services that seek to make programmes more accesible to audiences with diminished hearing. Some ideas were explored with ambition, and provided some great insights, without actually producing working demos- the 'Image Based Programme Entity Timeline' being one such noble effort.
What Happens Next?
The point of a day like Archak is not necessarily to produce the next user interface that you'll use to find the repeat of a show you enjoyed, or event to produce a prototype of a new tool that will help a programme editor find that perfect clip faster (though those would be nice bonuses). Rather, it's to get the creative juices flowing, and make the very best use of the collective intelligence of the BBC. On Tuesday we hacked, and on Wednesday we went back to our day jobs knowing about more data, more content, more tools and more people who can help us do our jobs better and more creatively- that's what Archak was for.
We had developers from London, Manchester, Cardiff and Bristol at Hackday, and at similar events we've had people from further afield. Later this month we're involved with BeeBCamp 3, a massive dual site event in London and Manchester, where we'll have a barnstorming conference to join up people from all over the BBC (plus a few guests from outside to stop us being too cosy), and hopefully have a similar effect across a broad spectrum of work areas.
So, watch this space- before too long we could be asking you to join us on one of these strange and magical days. And of course- we'd be delighted to come along to yours!
I'm John Fletcher, a Senior Engineer working in the Ingex team. Ingex is a PC-based audio/video recording system that can be used in multi-camera studios instead of video tape recorders. Tapeless recording systems, such as Ingex, provide a much faster and cheaper means of getting the recordings into the edit. And having the material in electronic form means you can do lots more with it, for example, putting it on a website for the production team to review. One of the aims of Ingex is to show that this kind of system can be put together using off-the-shelf IT components and open-source software.
The beginnings of Ingex were back in 2004 on the Children's BBC programme BAMZOOKi. The producer wanted to avoid all the re-typing of handwritten timing notes that happens when material is copied from video tape into the editing system. We put together a system that allowed the notes to be made on a PC in the studio gallery. The process is called logging and collects information such as: Scene 3, take 2 was a good one and is on tape between these times. With this we were able to automate the process of getting the right material into the edit (known as ingest) but it still involved a lot of video tapes.
We then worked on adding broadcast-quality video recording to the system and this was used for the next series of BAMZOOKi in 2005. We also provided new facilities for the programme makers such as DVDs of each day's recordings with a detailed menu allowing them to go straight to a particular scene/take/camera. The video below was produced in 2006 to showcase the Ingex systems use on BAMZOOKi (it can also be downloaded from http://downloads.bbc.co.uk/rd/pubs/whp/whp-pdf-files/whp133-downloads/BBC_RD_Production_Automation_2006.avi)
Since then, we've used Ingex on programmes including EastEnders and Dragons' Den, continually adding new features and refining the workflow. Right now we're using the system to help produce The Bottom Line, a business discussion programme with output on radio, TV and the web. The multi-platform aspect means a variety of file formats are needed and the topical nature of the programme means the edit has to be turned around very quickly. Add to that the need to rig the equipment in a cramped radio studio and Ingex is the ideal solution.
You might have seen that we recently won an Innovation Award from the Royal Television Society. The judges said Ingex had "without a doubt raised the bar in a dramatic style by expanding the scope of an already proven industry development that will now bring greater efficiency, flexibility and faster production turnaround times to many programme undertakings - and all at a cost that makes it a really practical proposition."
More technical detail on Ingex is available at the Ingex software site, where you can also download the open source licensed software.
My name's Steve Jolly and I'm a Research Engineer here at BBC R&D, working at our "North Lab" in Manchester. I'm one of around fifteen engineers who make up the "Distribution: Application Layer" section, a cluster of projects themed around improving some of the technologies which actually present broadcasters' content to the consumer.
Within the R&D department, the section has something of a heritage of its own: many of my colleagues here have worked together for a number of years on various interactive television projects, such as helping the BBC reduce the costs of producing interactive TV content. The three biggest digital television platforms in the UK (Freeview, Sky and Virgin Media) use incompatible formats for delivering "red button" interactive TV services, which can lead to inefficient use of developers' time if the same application has to be created separately for each one. The team spent a number of years looking at solutions to this problem, and ended up playing a major part in the development of the DVB Portable Content Format. This is a language for describing interactive TV applications that can be converted automatically into the several different formats required by different platforms. Since the conversion is automatic, each application only has to be written once.
The team also worked on extending MHEG-5, the interactive TV format used on Freeview and Freesat, to allow audio, video and interactive TV applications to be "pushed" by a broadcaster to digital video recorders (DVRs) in consumers' homes. The user could then select from this pre-recorded content in addition to the programmes that they had selected for recording themselves.
In 2006 and 2007 the team helped run a trial of this technology, in which 250 homes were provided with specially modified DVRs. These gave them access to a "catch-up" service, in which programmes selected by the BBC were automatically downloaded to a reserved area of the DVR's hard disk, and made available to the trialists on demand. The same trial also explored the possibilities of the technology for enhancing interactive applications with rich media components downloaded to the DVR well in advance. For example, an interactive quiz game was written incorporating video clips of the quizmaster asking the questions - something that sounds simple but which is surprisingly hard to do with conventional interactive TV technologies. Feedback from the trial was positive, with participants reporting, amongst other things, that the "BBC recommended" programmes introduced them to things they wouldn't normally have watched.
An interactive quiz incorporating video "pushed" to the DVR in advance using the MHEG-5 extensions developed by BBC R&D.
The work of the section today covers a much wider range of projects. For example, there's a small but enthusiastic team looking at how the BBC could make better use of mobile phones, and not just as devices to watch television programmes or listen to the radio on. I'm a part of this team, although my current project (described below) is a bit more general. My colleague Jerry Kramskoy wrote a post on the Internet Blog a little while back talking about some of our aspirations in this area.
A couple of us are also involved in the "Wombile" project: mostly in our own time we've written a basic platform for multiplayer mobile games, and a few games on top of it. My colleague Matt Hammond and I also wrote about that on the Internet Blog back in June.
There's another project in the section looking at the energy efficiency of broadcasting. Obviously there are environmental issues here that the BBC takes very seriously, but saving energy saves money, too. R&D engineers also need to be aware of our impact on consumers. We don't design the devices that people buy in the shops, but we do design some of the technologies and components inside them. Energy efficiency is therefore one of the things we bear in mind when we work on technologies that could find their way into people's homes. Not only does doing so reduce carbon emissions and save consumers money, it also helps the companies that make televisions and set-top boxes meet ever-stricter environmental regulations.
Unsurprisingly, I'm particularly excited by my own work: a colleague and I are designing a web API that would allow network-connected set-top boxes (and similar devices, such as internet radios) to be accessed and controlled by other devices on the home network. The implications of that go far beyond being able to control the telly from your PC or mobile phone, although that alone would be both cool and useful. The most important thing we want to enable is the development of user interfaces that are completely independent of the set-top box's built-in interface, optimised for use by people with specific disabilities or impairments.
Take blind people, for example. They're at a disadvantage when it comes to finding a conventional remote control in the first place, and most televisions and set-top boxes give you only visual feedback when you press the remote control's buttons, which clearly isn't a lot of use if you can't see it. It would be far better if a blind person's remote control was an app running on their mobile phone (which is likely to be within easy reach at all times), and if it could convey information about what the set-top box is doing by speaking to them.
A standard way to access set-top boxes via a local network would allow the development of this and many other innovative applications. Imagine if your mobile phone could find out what was recorded on your DVR, for example. It could then download mobile-size versions of the programmes you hadn't watched (or had only watched part of) for you to catch up with on your way to work, for example.
Over the coming months we'll be bringing you more detail on these projects as they reach important milestones, and we'd love to hear back from you the thoughts and ideas that they inspire.
[Ed: Our thanks to Steve Jolly for this post- the first of a series of round ups from the various research sections of R&D. Do let us know which areas you'd like to hear more about and if you have any questions about the subjects raised by Steve just pop them in the comments]
Phil de Nier, Matthew Marks, Phil Tudor and John Fletcher accept the award from Martha Lane Fox.
In the "Under the Bonnet" category our Distribution Core Technology team that developed DVB-T2 was awarded the prize. This is the technology that will enable Freeview HD (due to launch to audiences in 2010), and the work recognised with this prize has included some staggering breakthroughs in the deployment of advanced broadcast technology. The DVB-T2 team is headed up by Principal Research Engineer Nick Wells.
Justin Mitchell, Chris Nokes, Andrew Murphy and Martin Thorp receive their award from Martha Lane Fox
Our FM&T colleagues at BBC iPlayer won the prestigious Judges Award which was collected on behalf of the team by BBC Online Controller Seetha Kumar. This award was created by the RTS to celebrate the greatest vision in determining how media might develop in the future.
The BBC R&D High Frame Rate Television Experiment was also shortlisted in the Raising the Bar category but was pipped to the post by Ingex. High Frame Rate (or HFR as it inevitably gets called) is a little known and almost 'blue sky' area of research where we pioneered the use of very very high speed video cameras to produce pictures with breathtaking realism, and which may lead on to the technologies that follow after High Definition, as described in the relevant High Frame Rate white paper.
Open Innovation is a core principle of how we work and builds on the collaborative traditions the department has worked to throughout its history. We appreciate the best and most creative work comes when we share what we know, and draw on a wide community to build on that knowledge. Many of our projects are collaborative efforts, with research bodies, fellow broadcasters, universities, and companies from around the world. For several years we have been supporting the 'BBC Backstage' program, opening data and resources to independent developers and 'hackers' from across the UK.
There are a couple of existing blogs which are related to this area:
- Our Prototyping team has previously been blogging at RAD lab. From now on their posts will appear on this blog.
- KEP as a knowledge exchange network was set up earlier this year to encourage and develop knowledge exchange between the BBC and academia, specifically in the arts & humanities area. That effort is now ready to expand into a wider domain, so we felt it was a good opportunity to fit it into a wider R&D blog effort too.
The authors who will be posting here come from all across our department, and will tell you about the very latest developments we are moving into the mainstream, and some that lie further ahead. We're often appearing in public at conferences and conventions, so we'll let you know the one's we're heading for and share with you the highlights we pick up from around the world of media technology.
The coming months are an important time for BBC R&D- in February of 2010 we will be moving into our new southern home in the heart of the BBC's West London campus. As I'm sure you can appreciate, the effort of relocating over 100 staff and their facilities from our previous base of 60 years is substantial.
At the same time we are rapidly building up a BBC R&D laboratory in the North; the team has just moved into their interim accommodation in New Broadcasting House in Manchester. In 2012 that team will be on the move again, to take up a core role at the centre of the BBC's Media City base in Salford Quays. With each stage of these changes we'll keep you up to date, and share some of the trials and triumphs of these challenges.
This is a two way communication of course. If there's any particular area of our work that you'd like to know more about, a project, or a team, or perhaps some area of technology that you think we ought to know about, do let us know. Keep an eye on us- we'll be posting regularly and in some depth.
Thank you, I'm looking forward to the conversation,