Rob Curtis' Promise: We Will Get the Facts
Tuesday, 08 December 2009 21:13
FCC_Slide_41_last_1%"You might as well help us getting the information," Rob Curtis said to the companies at a broadband workshop. "We're going to get the facts we need." Bravo. Basic facts disclosed in almost every other country are held back as "confidential." Verizon even went to court to prevent Maryland finding out who in the state couldn't get broadband.

Rob's team at the broadband plan seem to be talking with everyone knowledgeable, and asking who else might have more to tell them. I'd bet the count of interviews and detailed emails is close to one thousand. They went beyond the DC lobbyists and lawyers directly to the people building networks, designing the equipment and doing the research. The CTOs of most of the carriers have come in. So has the guy responsible for most of DOCSIS 3.0 and an inventor of DSL. Folks who designed the Internet like Dave Clark and Vint Cerf helped out. So did a slew of other people who are out in the field day to day and close to the real problems.

The plan - much of which is written already - will be by far the most detailed information on broadband ever published.  But there remain some crucial gaps. (article to come)

The President set reaching the unserved as the first goal, but the needed data about that last 3-6% is very hard to come by. Randall Stephenson of AT&T visited Jules at the FCC and promised to help the work in every way practical. They shared some research that was helpful, I understand. I'm sure Verizon and Qwest would do similar if asked. They have the data and skills to get needed answers in days if they choose to. That's important, because 82% of the "unserved" homes are in the territory of the three Bells.

The September slides put forth figures of $20B and $35B for the remaining buildout, but the final is likely to be half that.

There are 3-7M "unserved" homes, many of which have cable TV but not data. Nearly half a million are in Charter alone, and would have been served years ago except Charter hit bankruptcy. Many of these can be reached for under $500 or under $1,000 (cable upgrades, DSL extenders, remote terminals for clusters). A goodly percentage of the remainder are in the low thousands, such as the $3K/home that Fairpoint was asking in the most difficult parts of Vermont.

The issue. as slide 41 of the September deck shows, is between .5% and 1.5% that are dramatically more remote. The curve rises gently to that point then climbs very rapidly. Many of these would cost more than $5K and much higher, which is why I assume the early estimates were so high.

Jonathan Adelstein at a Congressional hearing pointed to the role of satellite in the stimulus, probably in round two.  New satellites can go to 10 megabits with significantly improved latency. Not perfect, but not that bad either. Rob at the FCC workshop brought my attention to the improved satellite; I followed up and confirmed what the companies were saying for 2011 and 2012 is realistic.

There's something close to a consensus that improved satellite, with a subsidy if needed, is the right way to go for the last 1% or so. A maximum subsidy of, say, $5K/home, cuts the $20B in half  perhaps. I'm guessing without the data. Exactly where the point is reached is hard to know and makes a huge difference in the cost of the plan.

That's why the information is so crucial. The team has a respected professional outfit, Jim Stegeman's Costquest, working on the problem. Give him good data and we'll have clear choices in the plan.