[686] in Commercialization & Privatization of the Internet
NREN
daemon@ATHENA.MIT.EDU (Stephen Wolff)
Tue May 14 12:32:16 1991
To: adamg@world.std.com
Cc: com-priv@psi.com
Date: Tue, 14 May 91 12:08:45 EDT
From: Stephen Wolff <steve@cise.nsf.gov>
> If that is the case, then, wouldn't Gore's bill really just amount to a
> huge subsidy for whichever companies (IBM and MCI?) lucky enough to get
> the contract to run NREN?
The Interim NREN, at least, will look a lot like the current Internet, with
a lot of independent suppliers; there'll be no single contract for a single
company or consortium to get.
> If commercial clients are demanding high-speed
> data services, won't some company come along and provide it for them?
They already do: ANS, PSI, AlterNet,... There are substantial economies of
scale in the communications business; the taxpayers will benefit from the
presence of the paying commercial customers, since the (Interim) NREN can be
built to some degree on the margin.
> Given that such a network has some potentially lucrative bonuses (digital
> TV transmission, anyone?), why should taxpayers pay for this? After all,
> this isn't 1969, when nobody thought there was any future in packet
> networks.
cf. supra. Taxpayer money goes to support networking in the communities
targeted by the NREN program/legislation, but goes farther because of the
commercial customers.
In the immortal words of Bob Kahn: "You gotta have the flow to get the skim."
> Wolff said the government is going to wind up subsidizing any high-speed
> network no matter what, because the researchers at whom it will first be
> aimed would only include the costs of network access in future grant
> proposals.
The Federal government supports research at academic institutions. One
component of the cost is network access.
> Maybe, but shouldn't the companies that stand to benefit the
> most from this privatization pay more of the upfront costs?
I estimate that IBM and MCI have spent $60-70 million provisioning the
NSFNET Backbone **already** - with 17 months to go in NSF's Cooperative
Agreement with Merit. This kinda dwarfs the $28 million NSF will have spent
on that project for the whole five years, no? Then there's the $10 million
they put up as venture capital for ANS. How much more up front do you want?
