This story appeared on Network World at

http://www.networkworld.com/columnists/2007/073007bradner.html

FCC may be told to tell truth

'Net Insider

By Scott Bradner, Network World, 07/24/07

In the mid-1990s, Congress worried about the slow pace of broadband Internet deployment in the United States, so it included a requirement in the Telecommunications Act of 1996 that the FCC annually report on deployment status. Since then, the FCC has been living up to the letter of the law but producing almost useless information.

Congress may be about to order the FCC to change its ways and produce some information that actually makes sense.

I’ve written about the FCC’s deceptive reports in past columns: "Reading into the FCC’s ’Net access stats" and "Continuing deceptions". The reports are supposed to track the extent of the broadband deployment in the United States but are not addressing real broadband nor are they tracking its deployment.

Unlike the rest of the world, and unlike the telecom act requirement, the FCC does not even use the term “broadband." It did at one time, but has switched to using “high-speed." It is reasonable for the commission not to use broadband because the FCC is only talking about 200Kbps in one direction rather than the multimegabits per second that many countries define as broadband. The best way to describe the FCC’s criteria for high-speed networking is to say it is pathetic.

Equally pathetic has been the way the FCC shows the extent of deployment. The commission has been claiming that a single person subscribing to a high-speed Internet connection within a zip code means that everyone in that zip code could also subscribe to the same service.

Taking a quick look at the U.S. Census stats for zip codes shows the vast differences in area and population for zip codes. In addition, it is hard to imagine anyone not knowing someone who has been unable to get high-speed Internet service where they live, even if many others in the same area can.

Finally, the Senate Commerce Committee seems to have had enough of the FCC’s useless information posing as statistics. It just unanimously passed the Broadband Data Improvement Act. This bill would order the FCC to rethink its definition of broadband and collect information about the class of data service that is “not less than the data rate required to reliably transmit full-motion, high-definition video." Additionally, the bill requires the FCC report using the finer grain zip+4 codes. The bill does give the FCC an out: it would not have to use the zip+4 if the cost would be prohibitive according to the FCC’s own definition of prohibitive. So the FCC could still find a way to continue to generate useless info but at least it would have to explain why it decided to do so.

The bill would also order the FCC to do other things, such as develop a way that consumers could find out what performance they are getting from their “up-to X Mbps" service (see "Truth in speeds — broadband access") and to make a realistic comparison between broadband deployment in the United States and elsewhere in the world. The bill also establishes a grant program to help develop statewide programs to collect accurate data on broadband deployment.

The FCC seems to have been more interested in showing that the United States had good deployment of high-speed Internet services rather than finding out if it actually did. This bill would make it harder for the FCC to continue on that path, but sadly not impossible.

Disclaimer: As far as I know, proving results without determining the facts is not taught anywhere at Harvard. Thus, the above observation would not be one that Harvard would make, so I did.