| How do we measure, then? I agree that the one sub/ZIP is not a very good measurement, but I'm challenged to come up with a way that cable, telco, and wireless providers can send data back in a meaningful way to the FCC for processing.
If only the ILECs had to respond they could estimate the number of homes passed where they could provide broadband, and since ILECs have non-overlapping territory it's pretty easy to obtain an accurate picture. But what about when you add wireless and cable over top, how do you remove the duplicating coverage areas? Perhaps it will require some kind of GIS or sampling.
Frank |
|
 marigoldsGainfully employed, finallyPremium,MVM
join:2002-05-13
Saint Louis, MO
kudos:1 | Use the census defined TIGER/lines as a baseline for coverage. Throw on top of that the block household densities.
Have the providers supply data as address and number of subscribers. Geocode the addresses and match to TIGER lines.
Sum up the block densities assigned to each TIGER line segment and create a weighted penetration for each TIGER line based on the number of assigned subscribers then sum across the network. Or, alternatively, create a trend surface off percentage penetration and weighted by population density
Bingo. Number of subscribers per household, either by road network or by area. Areas with zero subscriber densities are areas likely to be unserved.
You can create a separate coverage network for each type of service, or sum subscribers across service if you want overall coverage by industry or for all broadband.
--
ISCABBS - the oldest and largest BBS on the Internet
telnet://whip.isca.uiowa.edu
Professional Geographer
Geographic Information Science researcher |
|
| Sounds like there is a way, we just needed geographer as opposed to a technologist to pipe in. I'm sure that service providers don't have TIGER-based groupings in their systems, but you're saying that all that needs to be done is supply an address?
Is it a reasonable reporting requirement to ask service providers (of any type) to supply address data? Perhaps I ask too quickly -- it's probably being done all the time by marketing/advertisers anyways. =)
Frank |
|
marigoldsGainfully employed, finallyPremium,MVM
join:2002-05-13
Saint Louis, MO
kudos:1 | E-911 compliant addresses fit into TIGER groupings (which are basically street segments with a to and from range of addresses on each side of the street plus an encoding of whether the range is odd, even, or both).
The geocoding process itself actually masks the data so that individual address information cannot be derived.
--
ISCABBS - the oldest and largest BBS on the Internet
telnet://whip.isca.uiowa.edu
Professional Geographer
Geographic Information Science researcher |
|
