Friday, February 12, 2010

Google's fibre ambitions

A very interesting announcement from Google this week:
"We're planning to build and test ultra high-speed broadband networks in a small number of trial locations across the United States. We'll deliver Internet speeds more than 100 times faster than what most Americans have access to today with 1 gigabit per second, fiber-to-the-home connections. We plan to offer service at a competitive price to at least 50,000 and potentially up to 500,000 people. Our goal is to experiment with new ways to help make Internet access better and faster for everyone."
What's particularly interesting  is that Google intends the network to be open access, giving users the choice of multiple service providers. Given how embedded existing complex commercial relationships are within current triple- and quadruple-play offerings, this will be a challenge, but if anyone can do it, it's Google? Building a network brings one set of challenges, building relationships with service providers to use you network brings a whole lot more.

At the NextGen09 conference in Leeds last November I asked the folks presenting the Digital Region project who they had lined up to run services over the new network. The answer? "Discussions are ongoing." Hmmm. The Digital Region site suggests they're still going, three months later. Which says to me that you need more than a philosophy of "if we build it, they will come" when embarking on such ventures? Field of dreams indeed...but, then again, Google have quite a few of their own services to run over a new fibre network, and I'm sure they have plenty of ideas about how to use all that extra bandwidth too.

The BBC's coverage flagged that in many ways this is just an extension of the fibre network Google has built to connect its data centres, which has been discussed many times previously, often in relation to the (ill-informed) net neutrality debate. This from VoIP News in May 2007:
 "Google head of special initiatives Eric Sacca has said the fibre helps Google avoid long-haul transport costs for traffic that needs to get to, say, peering points that connect to the AT&T network. Google also uses formerly dark fibre to interconnect its massive data centers and perform mundane tasks like replicating its search index to Google sites worldwide."
The article goes on to speculate what Google's other intentions might be, beyond the official line above, one of which was "a Google-branded telecom network" - good guess. But another analysis from Internet Evolution suggests Google's intentions may be very different, given the low return on fibre investments and the limited availability of "utility fibre" (=dark fibre? I guess the issue here is that what's there isn't where you need it?) in the ground:
"Given this, what’s behind the Google move?

Step back a bit. Ask yourself why Google pushed the notion of municipal WiMax for the Bay Area. Ask why they bid on a mobile spectrum auction. Google didn’t really get into any of these businesses. The goal was to manipulate others to build what Google wants. Even Android and Nexus One are manipulations of the mobile market. Now, Google is trying to start an arms race in broadband access.

Google wants an Internet that can grow to eat every other form of information and entertainment distribution. That gives Google, which is clearly the power of the Internet, a big ecosystem to support its growth. The part of the Internet that’s the hardest to grow isn’t the core, where traffic density and economy of scale apply, it’s the access connection. Nobody makes much of a return on broadband access, and Google is very afraid that will reduce access investment. A small Internet on-ramp means a small Internet.

Municipal governments, meaning taxpayers, can decide to spend money to create FTTH, and if they do, then Google will be glad to help out with ideas and tech support. But watch and see who carries the big cost. Google will provide proof-of-concept trials and will work to validate different technology approaches, but it's not going to pay the bills, because its ROI would be too low. When the dust clears on Google broadband, the same old players will have to be the ones to pull out their wallets."
The article does go on to temper this cynicism, by acknowledging that Google's offer could produce some notable benefits, for example by informing the FCC's National Broadband Plan (perhaps by encouraging the minimum bandwidth bar to be set a little higher?) and by driving further deployment and development by telcos. Another article on Internet Evolution highlights that "it takes more than equipment or cable to run a network", wondering if Google knows what it's getting itself into:
"...if Google really means to respond to members of the public and local governments with this FTTH plan, the company will have to be prepared for a lot more than a list of dark fibre and some major municipal glad-handing. The introduction of a super-fast FTTH fibre network anywhere increases issues of security, user support, and technological upkeep."
But if Google's plan is a success it could speed things up for everybody. Well, everybody in the USA at least...but then, who knows?

No comments:

Post a Comment