[1384] in Commercialization & Privatization of the Internet
Re: CSnet
daemon@ATHENA.MIT.EDU (Stephen Wolff)
Fri Sep 20 15:14:20 1991
To: Robert Ullmann <ARIEL@RELAY.Prime.COM>
Cc: ietf@isi.edu, com-priv@psi.com
Date: Fri, 20 Sep 91 14:13:25 EDT
From: Stephen Wolff <steve@ncri.cise.nsf.gov>
As many have already pointed out, CSnet was "the first" in many ways. E.g.,
* It was NSF's first entry into the networking business.
* It was the first NSF-initiated-and-funded network to achieve financial
self-sufficiency; it set a standard of aspiration that has been
inherited by the present generation of NSF mid-level networks.
But nobody has yet mentioned CSnet's most important "first".
DARPA (then ARPA) and the ARPANET demonstrated that packet-switching
technology permitted the sharing of expensive switch and transmission
facilities by different **users**.
With CSnet and ARPANET, DARPA and NSF demonstrated sharing by different
**sponsors**.
The extraordinary and foresighted agreement between the ARPANET and CSnet
principals to allow CSnet traffic to flow on ARPANET facilities, though
clearly consonant with OMB Circular A130, was nevertheless the "first" that
foreshadowed
- the (much) later initial instantiation of NSFNET (forty NSF-funded
additional nodes on ARPANET to reach the NSF community),
- the soon-to-expire "FRICC shared use" agreement in which DoE, NASA, NSF
and DARPA agree to carry each others' traffic if it meets **either**
network's acceptability criteria,
- the NSF agreement with ANS to permit commercial traffic on NSF-funded
facilities on a reimbursable-in-kind and non-interfering basis, and
- ultimately the whole (Federal) NREN idea itself.
I don't know for sure who were the ARPA/ARPANET and NSF/CSnet longthinkers
who made that original deal, but we owe them a lot.
