A report published this week by EDUCAUSE, a nonprofit association trying to promote higher education in the US, estimates that delivering a 100Mbps broadband connection to homes and businesses in the US using a fibre connection could cost up to $100 billion. That may sound like a large amount of money, but when put next to the £15 billion (approximately $30 billion at current exchange rates) estimated to fibre up the UK, it's comparatively a bargain.
Deploying fibre to the home (FTTH) to over 300 million people will be no easy task, particularly with the size of the country in hand. It is thus surprising that the cost per person in the UK ($500) is that much higher than the cost per person in the US ($333).
It is a pleasure to see the report is mindful of the future, considering what applications may be available today, but not forgetting that in 10 years things could be very different and much more bandwidth intensive. Currently, television is broaching on high definition (HDTV) with some content broadcast in this format. The future holds even higher resolution television such as Super Hi-Vision, being developed in Japan, which gives a pixel resolution of 7680 x 4320 (compared with the common 1280x720 for 720p/i), and comes in at 24Gbps uncompressed. Networks built now need to plan for the future when technology like this will be available to the masses rather than at the current, experimental stage.
100Mbps over a FTTH technology is just the beginning. The fibre being used can handle much faster speeds, and in the future, by upgrading the hardware at either end, could allow for necessary upgrades. This is where existing stop-gap technologies such as ADSL2+ being deployed here in the UK as part of BT's 21CN start to run into issues. The copper cable phone line being used are close to the limits of the bandwidth you can actually use them for at the distances being used. It is more cost efficient in the long term to deploy fibre sooner rather than investing in temporary technology. The problem is convincing the people who need to put forward the money.
For this, the report proposes the creation of a federal fund to put up money that would be matched by funds from individual states, the public and/or the private sector to pay for the costs. Spreading the burden of the cost is a wise plan, but the current US market, without a requirement for unbundling could see the creation of a monopoly on the network if regulation isn't imposed and private sector backing is solely employed. Whether this government aided approach would work in the UK is unknown. Previous public-private partnerships have been faced with a lot of opposition.
Unfortunately for the US, the Bush administration does not believe it should be intervening in pushing the market towards any particular technology, and this will no doubt mean that they won't be stumping up the cash to help get new networks deployed. Their view is in fact quite the opposite. Withdraw regulation that requires unbundling as they believe this is more likely to encourage local operators to upgrade their networks. This leaves little hope for a wide-spread FTTH roll-out within the US any time soon.
This is perhaps one aspect where the UK has been lucky. Stephen Timms has done a sterling job at promoting the need for high-speed broadband, and encouraging industry to work toward a solution, and we only hope his successor will strive to keep up his good work.
More details from the EDUCAUSE report, and about the US government's view can be read at ars technica.
Digg this
Add to del.icio.us
£15 billion here would be $15 billion there, not $30bn. The Lithuanian digging a trench here would get the minimum wage of £5.52 per hour and the Mexican doing it in the USA the Federal minimum wage of $5.85.
Then there's the planning issues and red tape around doing anything here to take into account.