So the wires were ablaze yesterday with news of Broadband.com's new national bandwidth map.
Cool, a Google Maps mashup that promises to deliver a bandwidth map of the whole US. Slick.
Drill in a little closer and that map of bubbles comes into clears and more useful focus. The default map, for example, shows only the point locations of DSL central offices, a helpful start. One can select the estimated broadband availability footprint of those locations (green below). And one can look at the footprints of Ethernet over Copper (EoC), as shown in orange in the Metro DC region below.
And, as Stacey Higginbotham gushes at
GigaOm, this sort of effort has the potential to provide enormous insight into where, at what speed, and at what cost broadband is available.
As an example, the map shows that 7 out of the 10 cities with the most expensive rates per migabit of bandwidth are in North Carolina. As the NC legislature contemplates the "level playing field" bill that would effectively exclude municipal and other public networks, learning that industry is putatively failing tilts the rhetorical scales toward the munis.
But what is the method behind these claims? You'll look in vain for an explanation of how the map works, whence the data come, etc? According to Higginbotham:
The map borrows from the National Broadband Map – launched last month by the National Telecommunications and Information Administration — in the form of an API call that takes some of the data shown on the federally funded map, but it also adds true crowdsourcing and machine learning to deliver a greater variety of information.
Cool! Crowdsourcing! Hive mind! Awesome. But, uh, how? Saying we're tapping the hivemind is one thing. Explaining the method by which that is rendered as an estimate of broadband availability, cost, and speed is another. I'll hold off calling Broadband.com's map the greatest thing since sliced bread until I see more about their data and their methods. Until then it is an enlightening collage.
A sidenote:
as I've said many times before, bandwidth is an important issue. But it's not the only concern in the world of broadband policy. For starters, often claims and maps about bandwidth have a huge error term. Any number of discrepancies exist between the benchmark one gets at a terminal and what is actually "available". Nonetheless, bandwidth is
among the things that must be considered in a coherent national broadband strategy.
Cool, a Google Maps mashup that promises to deliver a bandwidth map of the whole US. Slick.
Drill in a little closer and that map of bubbles comes into clears and more useful focus. The default map, for example, shows only the point locations of DSL central offices, a helpful start. One can select the estimated broadband availability footprint of those locations (green below). And one can look at the footprints of Ethernet over Copper (EoC), as shown in orange in the Metro DC region below.
And, as Stacey Higginbotham gushes at GigaOm, this sort of effort has the potential to provide enormous insight into where, at what speed, and at what cost broadband is available.
As an example, the map shows that 7 out of the 10 cities with the most expensive rates per migabit of bandwidth are in North Carolina. As the NC legislature contemplates the "level playing field" bill that would effectively exclude municipal and other public networks, learning that industry is putatively failing tilts the rhetorical scales toward the munis.
But what is the method behind these claims? You'll look in vain for an explanation of how the map works, whence the data come, etc? According to Higginbotham:
No comments:
Post a Comment