Still mostly head-down writing a note and getting ready for Telco 2.0 next week. However, a couple of mega-uber value readers have written in with some interesting observations, which make me question just exactly what we're calculating when we talk about expectations of bandwidth requirements in future. A couple of weeks back Arthur D. Little issued a report on European broadband commissioned by Liberty Global, which made the following conclusion:
"As the majority of broadband customers are still experiencing bandwidths below 1Mbps, the migration to higher bandwidth is not expected to take place in the immediate future. It will become visible in a couple of years. Most markets, including Switzerland, France and Austria, will have the highest demands in the 1 - 6Mbps range in 2011. Still, 30 - 50% of broadband households will have 6 - 30Mbps in 2011; not more than 10% of households are expected to exceed 30Mbps in 2011. An analysis of up- and download bandwidth requirements of next generation broadband services suggests that 8Mbps upload and 50Mbps download rate will be sufficient by 2011..."
However, mega-uber value reader number one said that at IBC he heard the BBC and Sky saying that for a true HD experience, the channels need to "breathe" at 16Mbps, even in MPEG-4. Apparently, the BBC's HD trial over DTT has a bitrate of around 20Mbps (check out the lower section of this thread). Get a couple of HD streams running around the house, with the teenagers uploading HD v-logs and downloading some ripped HD content from the internet while playing World of Warcraft, and the family better hope it's not among the 90% of broadband "have-nots" which Little expects in 2011.
Moreover, I think the industry has become a tad myopic, assuming this is totally driven by video, when there are also other important things to be considered. My second mega-uber value reader sends in this piece from Trouw in the Netherlands, which deals with a grid computing community set to officially launch tomorrow in the Dutch city of Almere, itself an early adopter of fiber. Some grid members with fiber connections are already taking part in a study of bone aging being carried out by the Erasmus Medical Center in Rotterdam, which involves files of up to 5GB in size.
Who's to say what applications and innovations will flourish if we focus on unlocking abundance (wink, Martin) rather than counting the reasons why we don't need too much bandwidth?