EBU • UER - Technical Review

No. 283 (June 2000)

Webcasting - one for the record book?

Since 1954, the EBU's Eurovision Song Contest has been broadcast live on TV and radio stations throughout Europe. This year’s event, which was held on 13 May in Stockholm, was remarkable because it was the first to be broadcast live over the Internet. Before the event, Microsoft's MSN issued a press release announcing their plans for this event to "achieve an on-line audience of up to 6 million visitors" and, hence, to win a place in the Guinness Book of Records – as the biggest ever “webcast”.

How well did it work in practice? Let me tell you about my personal experience …..

I logged on at the start of the contest at 2100 CET and was encouraged by the message below indicating that I was at the correct address, together with the message "Connecting to the server":

Unfortunately, I soon received the following "unhelpful" error message:

Success at last. After a few more attempts, audio plus video were received – even if the picture was slightly blurred and was refreshed only once every few seconds!

This success was short-lived.
After about 10 seconds, the sound stopped and the picture froze – followed by another error message ......

Trying again .....

- and again!

Those of you who know me personally will understand that I do not give up easily! I stayed on-line for one hour, repeatedly trying to access the Eurovision webcast. Throughout this period, the TV set in the adjacent room was delivering excellent quality video and audio!

My experience was that, in one hour, I received “video” for a total duration of 53 seconds. The longest video clip lasted 21 seconds before failing. Most of the time was spent watching "Connecting to server" messages, followed by innumerable error messages. Some people may have been successful, but our feedback suggests that most were unsuccessful. It seems unlikely that the webcast of the Eurovision Song Contest 2000 will enter the Guinness Book of Records.

So what is the moral of this sad story?

The Internet is able to do many wonderful things. It can already offer audio-on-demand services. When we have broadband connections (e.g. 1 - 2 Mbit/s) to the Internet, it will also be able to offer moderate-quality video-on-demand. But the current infrastructure of the Internet is not suited to simultaneous transmission of live events to millions of people (i.e. broadcasting). The problem is that a dedicated stream of data must sent to every single user. With millions of potential users, the network easily becomes congested by the millions of simultaneous streams of data. For several years, we have been told that the answer to this problem is "multicasting", whereby the data stream is distributed to many local servers which then "re-broadcast" the content to local users. However, most IP routers cannot support multicasting – and there seems to be no financial incentive for operators to introduce multicasting.

In essence, today's Internet suffers from two serious problems: the technology is immature and the economics are crazy! We know that the technological problems can be solved in due course – just as babies eventually become adults. It will probably be much harder to solve the economic difficulties.

Perhaps, we can also learn a lesson from the past. In the early days of telephones, telephone networks in cities were often used to distribute live audio from local opera houses and concert halls. In technical terms, these services were successful – apart from the fact that they impeded the use of the networks for two-way telephone traffic. Such telephone relays were quickly abandoned soon after the arrival of radio broadcasting, when it became obvious that the two-way capability of telephone networks should not be wasted by being used for uni-directional one-to-many services (broadcasting). Today, we seem to be repeating these early mistakes. Broadcasters should use the Intermet to extend their services, concentrating on the new and exciting opportunities offered by the Internet, rather than trying to use the Internet to deliver a poor imitation of broadcasting. The Internet is capable of much, much more.

Finally, do not believe all the hype you read or hear about the Internet!

Philip Laven
Director
EBU Technical Department