[2027] in Commercialization & Privatization of the Internet
Do I Hear Some Agreement Here?
daemon@ATHENA.MIT.EDU (Steve Simmons)
Sat Jan 18 18:26:18 1992
From: scs@lokkur.dexter.mi.us (Steve Simmons)
To: com-priv@psi.com (Com-Priv Mailing List)
Date: Sat, 18 Jan 92 18:23:54 EST
It sounds like most of the parties to the discussion have some common
thoughts:
1. If the government wants to sponsor universities and research groups
access to the Internet, they should do so by some subsidy to those
groups, not by subsidizing one or more networks which then compete
directly with unsubsidized groups.
2. The CO+RE scheme proposed by ANS is opposed by all other groups.
3. The actions which gave birth to ANS and the T-3 backbone give the
appearance of impropriety. In particular there is strong feeling
that ANS has received an improper advantage by receipt of hardware
and connections intended for purely research/education subsidies.
If these statements are correct, we need to do a number of things.
Although there is some overlap between the issues, we are better served
to treat them separately. Clarity of focus and simple single-issue
presentation is of the essence if we are to get anything done about
them. Taking them case by case:
1. We need to establish a lobbying effort to effect this change. I
personally favor a voucher-type system wherein a university would
be allocated N dollars for networking and could use those dollars
in whatever manner they feel most appropriate -- connect to ANS,
PSI, EInet, Merit, whatever. The critical issue to to remove the
conflict between RE-only and CO-only networks by removing the
distinction. This lobbying effort should be supported by CIX but
probably not be a CIX function. This will let non-CIX members like
EInet and Infolan support it without having to deal simultaneously
with issues of joining/not joining CIX.
2. ANS people have said repeatedly on this list that CO+RE billing is
a proposal. It is clear that CO+RE is loathed by almost everyone
potentially affected by it. The proposal should be abandoned. ANS
and other commercial providers still have the problem of dealing with
commercial vs. research traffic. The way they should deal with it
is by getting behind suggestion 1. Given the political muscle the
existing NSFNet/ANS folks have, this would help ensure the success
of the lobbying effort.
3. ANS should be physically separated from NSFNet and NSFNet should be
privatized as quickly as possible. This privatization should take
one of two forms. Either establish a regulated common carrier which
sells only "freeway" services (data pipes); or make sure the resulting
company is owned by a group of relatively smaller parties. Special
consideration might be appropriate for MCI, IBM and any other purely
commercial agency which has contributed "above and beyond" to the
existing NSFNet, but the bulk of the network should be set up as a
commercial company and sold to the highest bidders. In either form,
care should be taken that the existing commercial networks be allowed
to buy in.
If we can accomplish these things, we will have a situation where:
o NREN money can be allocated in a rational fashion;
o We truly have an internetwork;
o Issues of impropriety in the foundation of ANS are not resolved
but their future effects are minimized;
o All players in the field are truly on the same level.
