Tuesday, June 23, 2009
Talking Net Politics in Silicon Valley
At EComm 2009, in Silicon Valley, at least one speaker from Washington DC failed utterly to understand the techno-libertarian Bay Area mindset. This person, a friend, was all Obama-Genichowski-Net-Neutrality-Democrats as if the audience were a bunch of liberal beltway thinktank staffers who, incidentally, happened to know how to wire breadboards and write code. The talk got resistance, pushback and hostile questions from the audience. The response to that pushback was ineffective, perhaps even damaging to open Internet initiatives.
When policy people go to Silicon Valley, they need to explain why politics is relevant to Silicon Valley jobs and companies. I just re-stumbled across a clip of my own talk from the previous EComm. I didn't make that mistake; the talk got across. Here's a substantial clip from it:
When policy people go to Silicon Valley, they need to explain why politics is relevant to Silicon Valley jobs and companies. I just re-stumbled across a clip of my own talk from the previous EComm. I didn't make that mistake; the talk got across. Here's a substantial clip from it:
Technorati Tags: FCC, LeeDryburgh, NetworkNeutrality, Politics, EComm
Comments:
David:
In your talk, you claim that there is little or no "last mile" competition in the United States. This is simply not true. As the map at
http://bennett.com/blog/2009/02/thought-you-had-no-alternatives-for-broadband/
clearly shows, there is in fact quite a lot of last mile competition. Where competition is lacking is in the "middle mile" -- the facilities that deliver wholesale bandwidth to last mile providers. Competitors would not only flourish but dominate the landscape if they could simply obtain wholesale access to the Internet backbone at reasonable prices. However, the market is broken, and ILECs and national backbone owners are profiteering or even refusing to deal. You should be focusing on this problem, not on the "last mile." And no one should be advocating regulation that, together with the broken "middle mile" market, would deal the coup de grace to competitive ISPs. Want to see lots of competition? Focus on the markets which have failed, not on those where there is lots of competition.
In your talk, you claim that there is little or no "last mile" competition in the United States. This is simply not true. As the map at
http://bennett.com/blog/2009/02/thought-you-had-no-alternatives-for-broadband/
clearly shows, there is in fact quite a lot of last mile competition. Where competition is lacking is in the "middle mile" -- the facilities that deliver wholesale bandwidth to last mile providers. Competitors would not only flourish but dominate the landscape if they could simply obtain wholesale access to the Internet backbone at reasonable prices. However, the market is broken, and ILECs and national backbone owners are profiteering or even refusing to deal. You should be focusing on this problem, not on the "last mile." And no one should be advocating regulation that, together with the broken "middle mile" market, would deal the coup de grace to competitive ISPs. Want to see lots of competition? Focus on the markets which have failed, not on those where there is lots of competition.
# posted by 
: 6/24/2009 2:15 PM
: 6/24/2009 2:15 PM
Brett Glass should be interested to know that there is an organization set up specifically to address his concerns about the Middle Mile. NoChokePoints.Org is a new organization formed specifically to address precisely the concerns Brett Glass has about the Middle Mile. [I say "should be interested" because he's already indicated otherwise to me in personal correspondence.]
The map that Brett Glass refers to does not "clearly show . . . quite a lot of last mile competition." Even taking at face value Brett's claim (incidental to explaining the map) that more than 750,000 square miles are covered by WISPS, that's only 21% of the area of the country. Then there's quality of coverage; Brett's own coverage is so thin that he needs to engage in controversial network management practices.
Even if the Middle Mile problem is fixed, does today's tech/reg framework permit delivery to over 300,000,000 broadband customers? If so, how broad? And how much controversial network management will be necessary to do so?
The Middle Mile has one set of problems. The first mile has another related but not identical set of problems. The backbone has several other sets of related but not identical problems. There are also problems with the physical layer, and related but not identical problems with the IP layer, and related but not identical problems with application related layers. I emphatically reject any suggestion that I "should" focus on one of these problems, while "no one should" focus on other problems.
I also note that "competition" by itself is not my goal. I would like competition IF and only if it led to low prices, quality products, lots of choice along parameters that I care about, better services for the underserved, cultural richness, low barriers to innovation in end-user services, et cetera. If there are other ways to achieve those goals, I favor those ways too.
The map that Brett Glass refers to does not "clearly show . . . quite a lot of last mile competition." Even taking at face value Brett's claim (incidental to explaining the map) that more than 750,000 square miles are covered by WISPS, that's only 21% of the area of the country. Then there's quality of coverage; Brett's own coverage is so thin that he needs to engage in controversial network management practices.
Even if the Middle Mile problem is fixed, does today's tech/reg framework permit delivery to over 300,000,000 broadband customers? If so, how broad? And how much controversial network management will be necessary to do so?
The Middle Mile has one set of problems. The first mile has another related but not identical set of problems. The backbone has several other sets of related but not identical problems. There are also problems with the physical layer, and related but not identical problems with the IP layer, and related but not identical problems with application related layers. I emphatically reject any suggestion that I "should" focus on one of these problems, while "no one should" focus on other problems.
I also note that "competition" by itself is not my goal. I would like competition IF and only if it led to low prices, quality products, lots of choice along parameters that I care about, better services for the underserved, cultural richness, low barriers to innovation in end-user services, et cetera. If there are other ways to achieve those goals, I favor those ways too.
David's posting above warrants several corrections. Firstly, I am expressly not interested in the "nochokepoints.org" coalition, because it appears to have been organized by Washington, DC lobbying groups who are lobbying nonstop for regulations that would destroy small, independent, rural, and competitive ISPs. If competition is eliminated in the last mile, it hardly matters whether the situation improves in the "middle mile."
The posting also falsely claims that my ISP's coverage is "thin." This is not true at all. With dozens of access points capable of supplying 45 Mbps each, and point-to-point radios capable of providing far more than that to individual high capacity customers, we are capable of supplying our customers with far more bandwidth than they could afford to pay for. (Remember, due to the "middle mile" problem, bandwidth costs $100 per Mbps at wholesale in our area.) The reason why we manage our bandwidth very closely is to keep service affordable for our customers despite these very high wholesale costs, as well as the high costs of operating in a rural area where population density is low. If we did not manage our network well, our quality of service would be poor and our business would not be sustainable.
Finally, while the coverage shown on the map is only about 21% of the country, one must remember that the map is not complete. It shows, I estimate, only about 1/3 of all WISPs, and it covers more than 95% of the US population. That's more than is covered by any other single type of terrestrial high speed Internet service.
Post a Comment
The posting also falsely claims that my ISP's coverage is "thin." This is not true at all. With dozens of access points capable of supplying 45 Mbps each, and point-to-point radios capable of providing far more than that to individual high capacity customers, we are capable of supplying our customers with far more bandwidth than they could afford to pay for. (Remember, due to the "middle mile" problem, bandwidth costs $100 per Mbps at wholesale in our area.) The reason why we manage our bandwidth very closely is to keep service affordable for our customers despite these very high wholesale costs, as well as the high costs of operating in a rural area where population density is low. If we did not manage our network well, our quality of service would be poor and our business would not be sustainable.
Finally, while the coverage shown on the map is only about 21% of the country, one must remember that the map is not complete. It shows, I estimate, only about 1/3 of all WISPs, and it covers more than 95% of the US population. That's more than is covered by any other single type of terrestrial high speed Internet service.
# posted by : 6/25/2009 3:38 PM
: 6/25/2009 3:38 PM
