[1600] in Commercialization & Privatization of the Internet
Understanding the NSF's Pricing Projections
daemon@ATHENA.MIT.EDU (Gordon Cook)
Tue Nov 26 21:12:00 1991
To: com-priv@psi.com
Date: 26 Nov 91 21:06:36 EST (Tue)
From: cook@tmn.com (Gordon Cook)
<<MESSAGE from>> Gordon Cook 26-NOV-91 21:06
cook@tmn
{_Kudos to Steve, FARnet, the NSF and the networking community for a
series of actions that certainly seem to give the nod to service for the
research and education community rather than to technology transfer to the
commercial sector - to the extent that such difficult choices must be made.
I have some questions - bet you'll hear this phrase a lot Steve!
I understand that the NSF has upped it's payment to MERIT from $2.9 to 10
million dollars per year to cover the cost of the T-3 backbone
Depending on whether you talk to ANS or the NSF this 10 million only
covers somewhere between 1/3 and 2/3 of the actual cost of delivery of
backbone services.
One MIGHT, given the objectives of NREN, expect backbone speeds to reach
155 mgbs in 1993 and perhaps aggregate speeds of 622 mgbs by 1996.
Presumably such increased speed would cost MORE to offer than T-3?
Certainly not more in direct proportion, but still SOMEWHAT more?
Yet the NSF's projection of what it expects to have to PAY for this
service DECREASES substantially in comparison to what it is now paying!?
Moreover it continues to decrease during the 3 year life of the agreement.
It *IS* true that if you look at the needs of the 4,000 largest
corporations in the world, as Syncordia is reputedly doing, we are talking
about a multi-billion dollar a year TCP/IP internetworking global
industry. So one might assume one could get a good deal for the taxpayer
by the assumption that service providers would find this agreement such an
excellent one to become a part of that they would bid it for a song as a
"loss leader" in order to gain the presteige and experience.
But, if this chain of reasoning is correct, aren't we looking at an
interesting contradiction in the effect of our telecommunications policy
on regulated and non regulated carriers? Am I correct if I assert that
AT&T and the RBOCs as }iproviders of the PSTN are regulated in such a way
as to POSSIBLY be prevented from bidding a cooperative agreement such as
this at a loss!?? I am not absolutely certain that I am correct here, but
think I am. If I am not I am sure someone will correct me.
So my question is how can the pricing projection you list be fair to those
regulated carriers that may not be able to go that low? If the answer is
that they are out of luck, doesn't that help to drive the NREN into
deployment as a PRIVATE networking solution?
Is whether the PSTN gets a significant share of benefits from data
networking (which will exceed voice traffic in total network use well
before the end of the agreement) a significant matter for national policy
debate? Modernize private networks and not the PSTN and don't we
eventually get a two tiered communications system that, if conditions got
bad enough, could even threaten universal access to POTS let alone NREN?
Sorry to get so sticky when all you want is just good old academic
networking. Unfortunately the ramifications of NREN don't stop at the
boundaries of the NSFnet. Life is plain bloody complicated... see what
15 months at the Office of Technology Assessment does to one's thinking!?
<SIGH!> {_xD}i{_}i{_~~fi}.'{_
