[193] in Commercialization & Privatization of the Internet
Should the NREN be funded?
daemon@ATHENA.MIT.EDU (Marty Schoffstall)
Wed Feb 27 00:58:51 1991
Date: Wed, 27 Feb 91 00:35:13 -0500
From: schoff@psi.com (Marty Schoffstall)
To: com-priv@psi.com
Two weeks ago, an OTA workshop on the "Privatized NREN" was held
in DC. Attached is the uneditted PSI Presentation to the workshop,
I'd like to encourage CERFNet, ANS, ALTERNet, AT&T, ADAPSO, and the
RBOC's to post their presentations.
Your elected officials, and the unelected bureucrats may shortly
spend some of your taxes, appropriately or inappropriately.
Marty
----------------------
The Future of Networking
in the United States
or
"Why the NREN is Trouble"
a presentation to the
Office of Technology Assessment
Congress of the United States
Workshop: The Privatized NREN:
Marketplace and Public Policy Issues
Washington, DC
February 14, 1991
by
William L. Schrader
President & CEO
Performance Systems International, Inc.
11800 Sunrise Valley Dr., Suite 1100
Reston, VA 22091
Tel: 703-620-6651
Fax: 703-620-4586
email: wls@psi.com
Information presented here is my personal position, and may be
shared in any form, preferably in context. In bullet form, I have
attempted to establish a common ground of facts, then evolve to a
logical conclusion.
Overall Message:
The Internet is a prime example of successful technology
development and transfer by government. By investing early in
experimental networks, by continuously sponsoring and nurturing
enhancements, and by limiting their use to research and education
through acceptable use policies, government has encouraged a true
market to evolve.
If NREN funding is properly handled, government will:
- leverage its earlier investments,
- exploit economies of scale,
- position US firms to lead in global competition, and
- remove itself from an endless financial burden.
FACTS:
- The importance of global data traffic will grow (forever) and
is today well served by the Internet ("Regional Networks",
Merit, NSI, ESNet, DRI, UUNET, PSINet, etc.), and by VANs and
PDNs (Sprintnet, BT Tymnet, ATT's & IBM's networks, etc.).
- Internetworking (TCP/IP, OSI) technologies and services are
growing in acceptance and will continue to do so.
- Government sponsored and nurtured networks are decreasing
as a proportion of the Internet.
Government/Academic:
- NSF funding for the NSFNet Backbone terminates Nov 1992.
NSF funding will continue to Regionals and to "NREN", if
approved.
- DOE and NASA will never cooperate on shared resource
networks, due to their perceived mission requirements. Nor
will they outsource to PDNs.
- DARPA and NSF are the only agencies with the willingness
and wherewithal to manage an NREN.
- ANS enjoys a high profile, yet
- remains a nascent financial/lobbying/marketing organization,
- is working hard to deliver on promises,
- has convinced NSF and NREN advocates that ANS is currently
delivering the NSFNet Backbone and can build the NREN.
- DS3 services are not essential in 1991, will be marginally
used at less than 16 sites in America for many years, and
only for aggregated use on a "backbone".
- Individual operational network applications demanding DS3
or gigabits are years away, and probably limited to less than
several hundred users. The use of NREN's hundreds of
millions of dollars for only hundreds of people is inefficient
Big Science.
- Regional Networks recognize that commercial services
eventually will be of higher quality and lower cost than their
own. However, they do not want to lose control without
assurance that their missions will be addressed by emerging
systems.
- Acceptable Use Policies have encouraged the marketplace to
develop commercial networking. Private enterprise makes it
possible for government to get out of the costly business of
networking.
Commercial:
- PSI built the first national commercial Internet, is a
recognized industry leader, and provides service to over 10%
of the US Internet.
- Commercial Internetworking providers are increasing in
number (eg. UUNET, CERFnet) and their service offerings and
availability are growing and responding to market demand.
- Commercial service providers are properly motivated by
marketplace economics to integrate themselves with one
another fairly, reliably, and legally, without government
intervention or regulation.
- Economy of scale for T1-based services is reached by
commercial organizations at $3 to $5 million in revenue.
- Economy of scale for DS3-based services is reached at about
$15 million in revenue. (ANS's annual payment from NSF).
- Potential new commercial players in the Internet have, thus
far, been kept out of the market by government action,
principally through economic disincentives such as free
backbone service. Such subsidy ends in 1992.
- Since Congress plans to direct NREN funding to NSF, DARPA,
DOE, and NASA, which in turn will very likely redirect funds
primarily to existing contractors, there may not be fair
bidding.
- Legal action will be brought if NREN is funded this way, on
charges of restraint of trade and unfair government
procurements.
- BOC/RBOC and ATT interest is still growing, and will continue
to grow, and grow, but without any serious commitment, until
other commercial providers succeed in the market.
- If NREN funds are channeled to only one provider (such as
ANS), then the government is stepping backwards in the
marketplace, rebuilding an unwanted ATT monopoly analog.
Recommended Action:
- PASS the NREN bill, provide as much funding as taxpayers
will allow over the next 'n' years.
- Modify the bill to route ALL the funds through enlightened
government agencies which are committed to the concept of
shared resources.
- Modify the bill to require ALL operational funds to be
distributed DIRECTLY to America's post-secondary, secondary,
and elementary educational institutions and non-profit
research centers to be used to purchase internetworking
services.
- Modify the bill to minimize the amount of funds reserved for
network research, which will now be partially provided by
commercial firms in their own interest.
Expected Three to Five Year Results:
- a market segment will be seeded in K-12, and matured in
colleges,
- network value will be theirs to learn, feel, enjoy, and pay for,
- each site will be able to buy from their provider of choice,
- institutions make their own investment along with NREN funds,
- institutions chose their own timing,
- the general commercial market segments, which may be
ten times the size of NREN, are not destroyed before
they are mature,
- acceptable use limitations will be a non-issue,
- 10 to 20 (or more) commercial and non-commercial network
service providers could be at the DS3 scale of operation,
- hundreds or thousands of businesses will offer value-added
services,
- all players act on their own, minimizing government regulation,
- NREN funding could be stopped with no devastating effects,
- Gigabits services will evolve when they are needed by this
market,
- millions of dollars could be saved in the project.
Conclusion:
If NREN funding is provided directly to educational institutions,
then government will:
- leverage its earlier investments,
- exploit economies of scale,
- position US firms to lead in global competition, and
- remove itself from an endless financial burden.
Otherwise, do not pass the NREN bill.
