Tuesday, May 16, 2006

Tech: The future of Broadband (CeBIT part 3)

This is the final episode in my matters arising from CeBIT Australia 2006. Part one is an overview of future trends; part two is on Business Intelligence misconceptions.

Paul Budde, a telecommunications industry analyst, gave a series of seminars on the future of Broadband. His company Buddecom spans the globe; its biggest customer is the US government. Budde came across as particularly credible, and was fearless in disagreeing with others on his vision of the future. The following comments are all his, but any mistakes in transcription are mine.

Budde considers 64 K to be acceptable for voice quality (everything here being measured in Megabits per second); 1 M provides excellent audio; 2M for videoconferencing; and 6 Mbps for good video. Yet what is referred to as broadband in Australia is only 256 or 512K. There is no standard definition, but most comparable countries refer to anything 1M and upwards as broadband.

“Telepresence allows us to redefine space, time and knowledge”
- Paul Budde
The future is definitely in Fibre to the Home, ie fibre optics all the way. Nothing else can sustain the bandwidth en masse. However, this requires heavy infrastructure investment, which means phased rollout, and a number of interim technologies will be used before we get there. FttH will be preceded by Fibre to the Node, according to Budde, with Telstra rolling out 20,000 nodes over the next 3 – 5 years. This would be enough to cover the major urban centres of Sydney, Melbourne and Brisbane (other areas to follow), effectively with a node – a mini-exchange - in every street. It will sustain a 50M bandwidth, compared with the current technologies at 1.5M for ADSL and maximum 8M for ADSL2. Budde predicts 70 – 80% of Australia will have access to FttH within 10 years. FttH will make sense in any builtup area with 1500 – 3000 homes.

All this will be specifically driven by video, which has highest bandwidth demands. For this, consumers will be at the front, rather than business – in particular, via IPTV: not broadcast, but video-on-demand.

Wireless will be one of those interim solutions, but Budde says it will eventually be relegated to urban fringes, where the lower population density can’t sustain local nodes yet can cope with the spectrum demands. Satellite can deliver to remote areas, but costs are high, and subsidies will be needed for this 1 – 2% of the population. (This link gives the latest on government plans for infrastructure spending – subsidies – which Budde fears being utilised less than optimally, due to pork barrelling.)

ISPs will roll out their own DSLMs (nodes); competition will be somewhat ahead of Telstra, but it will pay to be close to a node.

Meanwhile, Telstra is trying to prolong the life of its copper network. They claim to have no plans to roll out ADSL2+ (a protocol permitting 24M speed, but which tapers off to ADSL speed – 1.5M – after about 4 kms), however Budde believes they will make an announcement offering this service before the end of June. Currently, only one provider – Internode – offers ADSL2+. They currently have about seven nodes in Sydney, so coverage is far from complete. Of course, in all this talk it’s the higher density populations that will first have access to the new technologies.

In Australia, broadband use will reach 4 million by the end of the year, yet Wireless is currently only at 50,000 subscribers – it has not taken off significantly in Australia.

Throughout this, the regulatory environment – and Telstra in particular – looms large. Telstra will be battling to preserve and/or capture monopolies; However, there’s no guarantee government tenders for subsidised remote areas will go their way. Budde in fact anticipates Telstra’s structure to be rationalised by Structural Separation – making a clean divide between infrastructure and content provision. He expects this to happen within 5 years. (This makes eminent sense as a vision, yet I think it will be hampered by the government’s plan to sell off its remaining stake by the end of this year. That makes SS harder; it would have to be accomplished by legislating the breakup of a private company. Budde expects the politics to nobble the roadmap somewhat, and it’s hard to disagree with that.)

Meanwhile, Budde expect to see transformations in existing providers from dumb pipe operators to smart pipe operators, providing services such as outsourcing, data centres, and content hosting.

As I said, Budde came across as credible and knowledgeable. His seminars were always standing room only, which I saw at none of the other three presentation sites.

No comments: