Here’s the summary of the comments I filed with the FCC on its broadband deployment report to Congress. A lot of the ink in the mainstream media today echoes a Free Press talking point to the effect that the Commission proposes to reduce everyone’s broadband speed, but that’s not correct. Read the whole filing here.

1. Rather than setting an arbitrary standard for the definition of “advanced telecommunications capability” or broadband, the Commission should rely on an objective, empirical measure sensitive to application requirements. The standard should be devised by measuring the requirements of the world’s most popular applications. The FCC’s goal should be to ensure that America’s networks are good enough to ensure that these applications are not performance-limited by our networks in the typical household.
2. Wireless networks have become America’s primary Internet connections. We move more data over wireless networks and spend more time using wireless devices than we do with legacy devices. Consequently, the evaluation of broadband deployment should stress wireless over wired networks. Wired networks increasingly serve a single need: they enable us to use video streaming services without concern for exceeding data caps. With the advent of 5G, they will take on a supporting role in backhaul.
3. Wireline networks are still important, but they have a different role in the 5G era than they’re had in the past. Just as Internet backbones have become feeder networks to CDNs across the Internet generally, wired networks at the Internet’s residential edge are becoming feeders to Wi-Fi access points and small 5G cells. Permitting processes are not keeping pace with the needs of fiber backhaul networks.
4. A US broadband policy that seeks to diffuse the broadband technology deployed in the cities and states with the best capabilities across the nation as a whole effectively brings the entire nation to the top rank of international capability. The cost of deploying broadband across a nation is largely a function of two factors: 1) the average distance of customer premises equipment (CPE) from the carrier facility; and 2) the average distance of CPE from the nearest Internet Exchange Point. Consequently, it is unrealistic to expect the US to create networks that are both faster and cheaper than those in Singapore and Hong Kong.
5. The FCC’s current method of mapping broadband deployments to population uses the Form 477 survey of offerings by Census Blocks. It would be wise for the Commission to consider turning this system on its head and estimating population coverage from something like the coverage maps published by wireless carriers. This method would simply require the superimposition of coverage maps on population density maps. It should not be difficult for wired broadband providers to convert their data on coverage of units with political boundaries to the format used by wireless carriers.

One way or another, the nation will continue making progress toward our wireless future. For a discussion of how that’s going, see this Washington Bytes chat with Peter Rysavy, Jodi Beggs, Hal Singer, and yours truly: The Dawn Of 5G: Will Wireless Kill the Broadband Star?