Google's Pointy Stick

3:00 PM -- You can't swing a cat today without hitting someone who's ready and willing to offer an opinion about the implications of Google (Nasdaq: GOOG)'s big (OK, relatively small and experimental) decision to deliver 1-Gbit/s services to as many as 500,000 users over an open fiber-to-the-premises platform. (See Google Jumps Into Gigabit FTTH.)

So, Google's throwing its hat in the ring as the newest broadband overbuilder... and this one isn't one of those fly-by-night operations from the 1990s and early 2000s that tended to flame out when the money was gone and there were few subscribers to keep the financials stoked. Google, by comparison, has money to burn.

With that in mind, the cable guys most certainly will have to give Google some respect and at least acknowledge what they're doing, and begin to noodle what this might all mean down the road, should Google expand its gigabit play beyond a few small, municipal pockets of the world.

Not that Verizon Communications Inc. (NYSE: VZ) FiOS and other FTTP platforms aren't already causing enough competitive angst for MSOs, but Google's move might end up spurring along cable's slow and very targeted move to FTTP, or at least speed up development of Docsis technologies that can match the kind of speeds Google is spouting off about. After all, as a friend in the industry likes to tell me from time to time, "cable doesn't move quickly on anything unless it's being jabbed with a pointy stick."

Well, cable, it looks like Google's trying to stick you in the ribs a bit with this little stunt. How does it feel?

Cable's downplaying this potential threat at the moment. National Cable & Telecommunications Association (NCTA) spokesman Brian Dietz issued a statement that the Association is "looking forward to learning more about Google's broadband experiment in the handful of trial locations they are planning," noting that cable's already spent $161 billion in the last 13 years to build its national broadband footprint, and will be spending billions more to improve speeds.

So take that, you broadband wannabe! Cable's already got the goods. What you got? Nothing yet!

Granted, Google's experiment will be small at first (it expects only 50,000 to 500,000 users to be involved in the project), so it would be quite a reach to say that today's announcement will cause CableLabs and the industry's legion of brainy engineers to turn up the heat on whatever comes after Docsis 3.0, a platform that already enables MSOs to offer (shared) speeds in excess of 100 Mbit/s.

But there's already work going on that will allow MSOs to deliver the kinds of speeds that Google intends to offer. There's a distinct possibility that the next version of Docsis (4.0, or some other designation) would let MSOs apply their existing provisioning systems and "Docsis management plane" to the FTTP world.

And FTTP might not have to factor in, either. Last year, we caught wind that inquiries are being made on cable modems that can bond 32 downstream channels -- enough to do bursts of 1.2 Gbit/s. (See The Ultimate Cable Modem .)

Of course, this all assumes that cable operators would even have that many channels to spare. And that's not going to be bloody likely until operators can truly go all-digital, recapture their valuable analog spectrum, and get the hell away from the super-inefficient broadcast video distribution model. But, if you're on the cable side of the fence, it's probably comforting to know that there's still plenty of blood left to squeeze from the HFC stone.

And (Warning! Shameless plug alert!) what's next for Docsis and cable's access architectures will be the topic of the day on Thursday, Feb. 25 at The Cable Center in Denver when we kick off our third-annual Light Reading Live! event: Cable Next-Gen Broadband Strategies: Docsis 3.0, Wireless, Fiber & Beyond.

Hope to see you there. We'll have our bells on.

— Jeff Baumgartner, Site Editor, Cable Digital News

Jeff Baumgartner 12/5/2012 | 4:42:48 PM
re: Google's Pointy Stick

Google hasn't said squat about how much this little fiber experiment will cost, but Will Richmond suggests it could clock in at $750M or more, basing some of that math on some back of the envelope estimations that apply some of Verizon FiOS buildout numbers with Google's plan to go mad scientist with 1-gigabit systems reaching 50,000 to 500,000 homes.

That number factors in the network buildout and all the in-home wiring and work that would need to be done.

And then there's the fact that Google doesn’t really have all that much experience in this area and won't be in line for some nifty volume discounts, so Will reasons that its like-for-like buildout costs will likely be higher than Verizon's.

And then there's that whole consumer adoption and return on investment thing to consider…making a network available to 500K customers doesn't mean the network will have 500K paying customers right off the bat.

I suggested that Google's got money to burn to pull off something like this, even if it ends up just being for show…but some of these numbers make it look like the company may have a broadband bonfire in mind. JB

Sign In