Getting blown away in the bandwidth explosion

Broadband is more available and faster than ever before. Guess who's ready to take full advantage - and guess who's missing the boat

I'm sitting at my desk at home with the computer of my dreams. It cost a few hundred pounds, and can do anything with sound, video and data I can wish for — if only I can get all that delivered. By early next year, a network gnome will have tweaked whatever it is that needs to be tweaked, my 2Mbps cable net connection will silently swell to 10Mbps, and my LAN will be the size of the planet. I think that will do the trick.

I didn't ask for this extra bandwidth nor will I be paying any more for it: Telewest will have quintupled my speed out of the goodness of its heart. Perhaps the profusion of double-digit broadband competition has helped this generosity: I could switch to more than ten times my original speed for less, but as so often inertia will keep me faithful

This profusion of bandwidth is more than matched elsewhere in Europe. For example, 20Mbps is the norm in France for around €20 a month — and France Telecom is talking about 100Mbps delivered over ordinary copper. An ISP ops guy explained it to me: "There's a glut of bandwidth. Everyone over-invested. So much dark fibre in the ground, and the lit stuff is nowhere near capacity."

File-swappers (you haven't gone away at all, you naughty people) will chortle at the extra pipeage, and the honest purveyors of online experiences can create even more exciting high-calorie content. It's great for media, of course: the Government is already seeing IPTV as the final pep pill necessary to get the nag of analogue switch-off over the line in 2012. Not within range of the transmitter? Never mind, have this set-top box plugged into the phone line — then we can sell off the analogue frequencies to the radio network people who still seem to think bandwidth is scarce and expensive.

The only people not rejoicing will be the Microsofties. It's not that they can't use the bandwidth — those quarter-gigabyte service packs don't distribute themselves, you know — nor that they didn't see it coming. Five years ago, the company invested billions in Telewest and NTL in exchange for a controlling interest in deciding what equipment the cablecos would provide. Remember Microsoft TV?

Unfortunately for Microsoft, the European Commission decided that this was anticompetitive. "Not that we mind the cablecos using MSTV," the Europeans said, "if that's what they want. But you have to give others the chance to bid as well in open competition." Microsoft slunk away from Telewest, selling its $2.6bn investment for $5m — while all this had been going on, the great dot-com bubble had burst and you couldn't give the stuff away. It didn't need the money. It must have been pique.

Microsoft has weathered such bad decisions before. It might have misread the market, the regulatory position and the viability of MSTV, but at least nobody else had taken advantage. Until Google. Showing as much perspicacity as Microsoft has clumsiness, the G has been slurping up that surplus dark fibre like a starving Sicilian at a spaghetti festival. It's not paying dot-com prices, it's not annoying any regulators, it's merely getting itself into the position where it can shift huge amounts of data wherever it likes and however it likes without being beholden to anyone else.

Microsoft's TV strategy relied on total control: the cableco, the set-top box, the data formats, the works. If one bit failed, the lot failed — the downside of control economies throughout the ages. Google may or may not have...

For more, click here...

...a TV strategy — ok, of course it does. It will be indexing, archiving and delivering the world's television before you can say "Murdoch who?" — but when it wants to play, it can just publish a client on its network and offer something at the other end too darn tempting to resist. Meanwhile, that network is busy moving databases, software services, telephone calls, IM, big maps of the world. The fatter the pipes we users have into the Internet, the better the Googlesphere looks. All its money comes down those pipes, all its money moves over its network.

Microsoft's latest reorganisation is much more buttoned down: applications software is different from operating systems is different from games. You might need to mix them up to build, say, a games computer that's connected to a media centre consuming online services, but you will have to buy your games computer and your media centre and your online services, and trust us to make it all work together. Software services, as it sees them, live entirely in that frame. Control resides.

Google thinks different. It knows that with smart-enough clients and fast-enough bandwidth, all you need to do is dress up digits, serve them hot and take your cut. Make it open — others will flood in and fill any gaps, mash the bits up, find ways of making money while making you money.

Take games. Why spend billions building an expensive, hard to control, obsolete-from-launch games console, discounting it, getting it out through the retail channels and waiting to get the money back on the games? You can build much more efficient dedicated game server farms and let smart clients suck it up through Ethernet-speed digital video feeds. All that business with DRM and trusted platforms, locking down the consoles so nobody can threaten your revenue stream only for a teenager called ZyberVole to get annoyed and crack it so you have to take them to court and get laughed at? Will never happen. Can never happen. Not an issue. Save your effort for the games.

Google doesn't have a gaming division. It may not even have a gaming strategy: then again, who are we kidding? You can bet your last battledroid that it has one and it's not dissimilar to the ideas above. The same ideas can deliver TV, the same ideas can deliver software services. Try it for free, pay as you go, heck, join in as a provider if you've got a good idea.

The only thing Microsoft has got going for its strategy is that it's worked in the past. The only thing that will stop Google is if there's a lack of global bandwidth, no way of getting it in the front door and no cheap, powerful, open clients.

I'm sitting at my desk at home with the computer of my dreams. It cost a few hundred pounds, and can do anything with sound, video and data I can wish for — if only I can get all that delivered. By early next year my 2Mbps cable net connection will have silently swelled to 10Mbps, and my LAN will be the size of the planet.

I think that will do the trick.