A STRUGGLE has begun for the future of television transmission over the internet. Its outcome will dictate whether the net becomes an outlet for programmes and films that cater to a diverse variety of tastes, or merely another source of bland reruns and repackaged shows from mainstream TV.
TV companies are already experimenting with internet broadcasts. Services range from rebroadcasts of old programmes, such as Warner Brothers and AOL’s In2TV, launching in January, to the Apple iTunes Music Store’s sale of new episodes from US series such as Lost and Desperate Housewives. With each network hoping to gain a foothold in this potentially lucrative market, expect to see the number of similar commercial ventures soar.
But while the big names of broadcasting prepare to battle it out, a handful of nascent small-scale networks are also getting in on the act. One prototype network, temporarily called DTV, is using the same free-to-download open-source approach to its software that spawned the Linux operating system and the Firefox web browser. Its authors have banded together under the banner of a fledgling non-profit organisation called the Participatory Culture Foundation (PCF), based in Worcester, Massachusetts. DTV, which is so far available only for the Macintosh OS X and Linux operating systems, integrates a channel guide, video downloader and viewer into one simple, free software package. The source code for the software has also been published, allowing anyone to adapt and improve it.
Advertisement
PCF has also developed an open-source video publishing tool called Broadcast Machine, which anyone can use to start up a DTV channel of their own. Upload the Broadcast Machine software to your web server, and establishing your own DTV channel and making programmes available to DTV viewers around the world becomes as easy as adding attachments to an email, or so the organisation boasts. “Someone whose day-to-day experience of using the internet is limited to email or browsing web pages would still be able to figure it out and get some videos and watch them and come back for more,” says Holmes Wilson, co-founder of PCF.
The organisation’s vision has drawn financial support from philanthropists Andy and Deborah Rappaport and from Mitch Kapor, the developer of the Lotus 1-2-3 spreadsheet – the application that helped launch the PC revolution in the early 1980s. “The people who did this are not first and foremost hard-core technologists,” Kapor says. “They’re designers, users and activists, and I think that’s given them a certain freedom to work at a higher, more user-oriented design level.”
DTV uses a simplified version of BitTorrent, a peer-to-peer distribution tool. It is a “swarming” system: every user who downloads a file also passes around parts of that same file to other downloaders. This means that if a file becomes popular, the burden of uploading it is passed from the website distributing it to the downloaders themselves, which avoids the need for the originating website to provide the huge bandwidth that would otherwise be required. Much to movie and television companies’ frustration, BitTorrent has no central clearing house and thus no place to police copyright restrictions.
Another fledgling public-access internet TV network is attempting to bridge the gap between the peer-to-peer, open-source vision and the centrally managed and distributed frameworks of today’s network television. In April, the Open Media Network (OMN) of Sunnyvale, California, released a prototype version of its free-to-download internet TV browser and broadcaster. OMN, the brainchild of former Netscape executive Mike Homer, is slicker than DTV and boasts more professional content, with shows from US public television stations such as KQED in San Francisco and KTCA in Minneapolis/St Paul. Flagship DTV channels, by comparison, tend to be outside-the-mainstream feeds likely to appeal to open-source purists, such as the alternative news programme Democracy Now, and former US vice-president Al Gore’s youth-oriented cable network Current.
In place of BitTorrent, OMN uses a commercial swarming file-sharing platform developed by Homer and called Kontiki. Unlike BitTorrent, Kontiki allows copyright holders to put expiry dates on their videos and to control downloads via a central server. An independent film-maker using the OMN/Kontiki system would initially upload his or her film onto the Kontiki server. Anyone downloading the film to watch it would retain fragments on their own computer, from where other users would be able to download it, thus relieving the burden on the server’s bandwidth. However, the server would retain control over the film, and if a commercial distributor spotted it the film-maker could take it off OMN and distribute it through conventional, profitable channels.
“Enhanced exposure, either in addition to or instead of [film festivals], is going to happen more and more on the internet. I think OMN is a great place for that, because we can handle very large files,” says Wayne Dyer, the network’s director of development and content.
DVD or even HDTV-quality broadcasts can now be viewed on OMN. And when the next version of the browser is made available early next year, users will be able to download video files to Apple’s Video iPod, a PC running Windows Media Player or a TiVo hard-disc recording device, for viewing on the move or on a traditional TV set.
Danny O’Brien of the Electronic Frontier Foundation, a group also founded by Kapor, says it remains unclear exactly how internet TV systems will evolve. It is unlikely that it will consist exclusively of straightforward adaptations of today’s TV programming, and OMN and DTV offer a glimpse of one direction it may go. “Watching Desperate Housewives on your Video iPod is OK,” O’Brien says. “But being able to video something and send it to somebody’s phone, or edit something, or subscribe to a channel that is composed entirely of crochet-loving fans of German industrial music, that’s new, and that’s transformative. And it will feed into and change everything to do with how TV is made.”
Squeezing in the bits
One of the vexed questions surrounding the design of the framework for internet television is what method to use to compress the video and audio signal. There is no prospect any time soon of internet bandwidth being sufficient to carry high-resolution signals that have not been compressed by some kind of digital coder-decoder (“codec”).
The most popular codec is MP3 – or MPEG Audio Layer 3, to give it its proper name. Apple’s Quicktime is another. But these and most other existing codecs depend on patented algorithms, so using them requires a licence from the patent holders. The price of every DVD player or decoder card, for instance, includes a $2.50 fee paid to a conglomerate of electronics corporations for their patent on the MPEG-2 codec, and built into the price of every DVD disc is a 4-cent MPEG-2 licence fee. To avoid the need for such payments, Linux programmers prefer to use an open-source royalty-free audio codec called Ogg Vorbis.
As a result of the need to pay patent holders, patent licensing “has a braking effect on the uptake of new technology”, says Tim Borer of the BBC’s research and development labs in Tadworth, Surrey.
So Borer is heading up an effort to do for video codecs what Ogg Vorbis did for audio. The BBC’s Dirac video codec project is able to compress standard-definition television signals down to the same data rates as Apple’s Quicktime and Microsoft’s .wmv family of codecs. This equates to reducing footage at 720-by-576-pixel resolution to a 1-megabit-per-second stream. It cannot yet compress high-resolution video signals in real time.
Dirac was developed with its specifications and source code free and open to all, and can be used by anyone without payment. It will be submitted to an international standards body for approval next year.
