Original URL: http://www.theregister.co.uk/2007/03/30/cable_in_stockholm/

Forget WiMax, Stockholm's got cable to spare

Red Ken, take notes

By Guy Kewney

Posted in Broadband, 30th March 2007 09:12 GMT

Column You could not think of a better place to demonstrate the wonders of WiMAX. A district, 84 kilometers across, divided into 24,000 plots of land with open, deep water between each plot. It's the Stockholm archipelago, East of Sweden, and almost every island is inhabited. Intel, Cisco and Stockholm Cable (Stokab) have turned this into the biggest showcase of WiMAX wireless you will find.

Ignore the WiMAX.

The really clever thing about Greater Stockholm isn't the deep water, it's the deep tunnels in the urban areas, carrying internet cable, not wireless. And the triumph isn't a technology triumph, it's a social management triumph. Oddly, nobody seems to want to admit it, because "social" sounds a bit too much like "socialist" for a lot of people's taste.

The problem - and its solution - should appeal to Red Ken in London. Ken Livingstone is faced with a serious problem - a future of traffic disruption. A problem Stockholm turned into an advantage a decade ago.

I was invited to see Stokab, an internet utility. When it was first proposed, it was greeted with absolute horror by the Swedish internet community because what it proposed was, quite simply, public ownership of the infrastructure cable.

It grew out of frustration by the city management over finding the streets of the city constantly closed off, with heavy earth-moving equipment making yet another hole in the road.

Exactly who first discovered that a high proportion of these holes were carrying internet cable is lost in history; but someone - clearly a genius - came up with the observation that the city would work a lot better if hole-digging was co-ordinated. If someone wanted to dig a hole, find out who else might benefit from having the traffic stopped and a trench opened up - and get them all along to put their pipes and cables and junctions and valves in at the same time.

And, this inspired administrator added: "While it's open, we'll stuff some high performance cable into it and gradually build up a high speed network!"

The question that mattered to the internet authorities, of course, was "who owns it?" and it was resolved unilaterally. A utility would be set up to run the internet cabling in exactly the same way a utility managed the sewage system. It would belong to Greater Stockholm.

Political dogma means that this experiment has been pretty thoroughly ignored by a lot of commentators. American politicians are sufficiently horrified by the thought of urban Wi-Fi "hotzone" development - all their well-trained reflexes, built up over years of anti-USSR and anti-Cuba propaganda tell them that "free enterprise is the solution...what's the problem?" and dissent from that basic axiom is difficult to express.

Dissent was vocal in the early days of Stokab, too. Internet providers pointed out the importance of competition. Prices, they pointed out, would never come down under a civic-owned authority - not the way they would if several private networks were competing to carry the bits.

Apparently, the dissent lasted about as long as it took to get Stokab up and running. As soon as it actually had an infrastructure, its charter forced it to provide those bits at cost. And the licence it had to put cable in any trench meant that its costs were a fraction of what independent ISPs had had to spend commissioning large scale civil engineering projects.

The simple fact is that most of the dreams of truly high-speed, high-def internet everywhere cannot be provided until high-speed fibre runs not just to the kerb, but into every home.

This isn't the place for a serious dissertation on interactive video, but anybody who has watched the "videophone" connections used by TV reporters these days will know that it's actually worse to have speech that's badly lip-synched than just plain audio without images.

All you have to accept is that the only really satisfactory way to have two-way video is to have pretty high definition cameras, with a minimum of data compression, a minimum of cache, and as little processor involvement as possible. Yes, it's possible to have video streaming work very satisfactorily, too. But as long as you're streaming, latency isn't a problem. Two way interaction, however, and suddenly latency is a real problem; and the internet itself generates latency at every stage.

The way around that is to have a raw data rate which is very high and a cost per bit which is very low.

In the UK, the cable experiment - imposed on BT by Margaret Thatcher, who hated state ownership or anything like it - meant that vast amounts of capital investment went into digging up huge urban areas, putting independent cable companies into business. Not one of those original cable companies is still in business. Their initial high investment costs and the low margins they were allowed to make, made them takeover targets - today, all we have is one cable company, Virgin Media. And there are plenty of districts where the cable pioneers never went and, probably, never will.

London's Mayor should have no inhibitions about imposing a city-owned utility on the capital. He is currently staring into a plan of infrastructure changes which will mean digging up the roads anyway.

This idea is not a challenge to BT or Virgin or any other cable operation. It's a co-operative, community-funded way of keeping costs down at a level where the cost per bit can get down to the level that true interactive, two-way video communications is feasible.

In the 24,000 islands off Stockholm, Cisco and Intel claim to be very pleased with their demo of how WiMAX can help cover areas where digging up the road simply isn't an option. The waters off the East coast of Sweden are very deep, and people have a habit of dropping anchors into them.

In London, who knows? Maybe a WiMAX experiment might even make sense, with a cheap basic infrastructure around which to build it? But, until global warming really does turn Greater London into a series of islands in the sea, probably not... ®