Campaign to Block new NSFNET Awards by Mike Roberts and Where does the Real Interest of Mike Roberts

[David Farber <>]

Never let it be said that interesting people steered away from
controversial issues. There are two notes below, the first  is a memo Mike
Roberts, VP of EDUCOM, sent out to his university members and the second is
a rebuttal by Gordon Cook. I will again accept and distribute arguments and
counter-arguments.


Dave


*******************************************************************************


March 15, 1994


To:     NTTF Members
From:   Mike Roberts


Subj:   Campaign to Block new NSFNET Awards


Since the action by the National Science Board on February 11 approving the
staff
recommendation for the first two of four awards in connection with deployment
of the next generation of NSFNET, a concerted effort is being made by several
vendors to block the awards, including Sprint, Alternet and PSI. Those of you
who follow the net traffic on this subject are familiar with the details. A cop
y
of the Foundation's press release on the awards appears below.


While the NSF external review process, in which many of you participate, is not
perfect, it has a long history of contributing to high quality science and
infra-
structure.  In the particular case of this series of awards, the Networking
Division of the Foundation went to considerable lengths to secure community
input, and the final version of the Solicitation document reflected that input.
The comments and suggestions of NTTF appear in a number of places in the
document. The review and selection process for these awards was unusually
thorough, as was the review by the Science Board itself.  I am confident that
the awards will be sustained.


However, the current campaign by a noisy few seeking to advance their own
economic
interests has the potential to delay the transition to the new network arrange-
ments, as well as to undermine the Foundation's ability to continue to use
the Cooperative Agreement law for purposes of forming public-private partner-
ships to advance networking research and development.


As the events of the past six years have amply demonstrated, the collaboration
of universities, computer firms, and telecommunications firms in developing
and using advanced networking has had a very tangible positive effect on
the competitive position of our electronics industry, and on the infrastructure
for research and education.  The use of the Cooperative Agreement Act by the
Foundation to form joint ventures in networking has had precisely the result
intended by Congress when it was passed.


It is not only in the national interest, it is in the interests of the
university
community to promote and defend the collaborative structure and process which
is contained in the Cooperative Agreement Act.  I hope that you will find
appropriate opportunities to do so with your campus executive structure and
with vendors with whom you come in contact.


- - Mike.






  ------------------------------- Message Contents ----------------------------
- -


  National Science Foundation                     February 14, 1994
                                                  NSF PR 94-8




  NSB GIVES GREEN LIGHT TO NSFNET'S NEXT STEPS:  NEW ARCHITECTURE AND A
  CUTTING-EDGE NETWORK


  Two major developments toward the information superhighway received green
  lights Friday from the National Science Board (NSB).


  At its meeting in Arlington, Va., the board gave its administrative approval
  for the National Science Foundation (NSF) to proceed with arrangements for a
  new architecture for the NSFNET computer network services, including an
  experimental expressway -- called a very high-speed Backbone Network Service
  (vBNS).  NSFNET currently links more than 1,000 U.S. universities and other
  research and educational institutions to the Internet.


  The NSB, a 24-member board that governs major policy decisions of the NSF,
  approved a proposal that the agency's networking division award the
  operation of the vBNS to MCI Communications Corp. of Washington, D.C., for a
  total not to exceed $50 million over five years.


  The NSB also approved a plan by the networking division to establish two
  five-year collaborative cooperative agreements to offer routing arbiter (RA)
  services for NSFNET and the Internet. Merit Network, Inc., a consortium of
  Michigan universities, will receive up to $11,099,743 to provide overall RA
  management and to take the lead responsibility for a registry database, an
  operations center, and software development. The University of Southern
  California's Information Sciences Institute, working with IBM, will receive
  up to $9,235,847 to develop route servers, advanced routing techniques, a
  testbed, and routing engineering.


  According to Stephen Wolff, director of NSF's Division of Networking and
  Communications Research and Infrastructure, "The board's action gives us the
  license and the framework to negotiate with the parties involved. The
  decisions to work with MCI, Merit, and USC followed an exhaustive review
  process."


  Today's developments, Wolff said, are among the first steps in what will be
  an orderly transition from the current NSFNET architecture to a new
  arrangement that, while advancing both, will distinguish more clearly between
  cutting-edge experimental efforts and widely used public "commodity" networks
.


  "The vBNS will be a physically separate network," said Wolff, "one to push
  the parameters of high-speed networking and to connect NSF's supercomputer
  centers, without affecting general network services."


  The routing arbiter will be one of three major components of the new, advance
d
  "general-use" network services. The other two components, which have not yet
  been awarded, are projects calling for network access point managers and for
  regional network providers.


  With transmission capability of 45 million bits of data per second,
  the current NSFNET backbone makes the connections over commercially leased
  lines by linking 19 sites-called nodes-throughout the U.S. At these sites,
  mid-level or regional networks are attached and thus interconnected. These
  regional networks, in turn, reach out to-and from-thousands of local
  networks at schools, universities, libraries, research laboratories,
  government facilities and supporting commercial organizations. This web of
  computer conduits allows its users to exchange electronic mail, avail
  themselves of massive computers, and search libraries and databases-all at
  distant sites almost instantaneously.


  Developed by NSF in 1985, the NSFNET program long ago outgrew its initial
  vision: to provide broadband access to NSF's five supercomputing centers for
  researchers with data- or algorithm-intensive projects. The program now
  supports not only the expanding backbone services, but also directory and
  information services, operations for regional networks, and connections for
  universities.


  The new architecture calls for the regional networks to secure backbone
  services from commercial network providers, with a five-year phasing out of
  NSF's backbone funding for the regionals.


  NSFNET Program Director Priscilla Huston said, "Our first priority as we
  make this transition is to ensure that the new infrastructure is robust and
  dependable before we shut down what has become a legendary national backbone
  service among the networking community. The new architecture will allow the
  research and education communities to take advantage of the excellent service
s
  available from commercial suppliers while it provides the interconnectivity
  that's essential for a vibrant national infrastructure."


  "The transition," said Wolff, "recognizes the changing nature of the
   networking  marketplace."


  "In 1987 you couldn't buy high-bandwidth Internet services," said Wolff.
  "There were neither suppliers nor market.  So we commissioned a
  high-bandwidth backbone; and, in the seven years since, a number of
  comparable commercial services have arisen-catalyzed in part by the market
  generated by the NSFNET backbone. So now we can give backbone funding to the
  regionals and let them buy that connectivity from the suppliers of their
  choice."


                                  -end-








********************************************************************************




From: cook () path net (Gordon Cook)
Date: Sun, 20 Mar 1994 22:13:50 GMT
To: farber () central cis upenn edu
Subject:  Where does the Real Interest of Mike Roberts Constituents Lie?


Dave Farber has shared Mike Roberts post to NTTF with me and asked me to
comment.  The timing couldn't be better since I have just finished the first
draft of a 6,000 word cover article for the April COOK Report on Internet ->
NREN comparing and contrasting, in great detail, the MCI protest of Sprint's
win of ESnet and Sprint's protest of MCI's win of the vBNS.  I make no secret
of the fact that I am sympathetic to Sprint.  Therefore I *regret* to report
that when I publish in about two weeks - based on the information I have now -
I will report that MCI's ESnet protest seems exceptionally strong and that
Sprint's vBNS protest seems exceptionally weak.  On the basis of the
information I now have sprint is simply not prepared to play political and
legal hardball of the requisite stature.


Now Mike's constituency is the Educom ivory tower academic elite and he is
perfectly entitled to lobby for them as he sees fit.  Only problem is how long
will it before some people on the hill begin to catch on to the self-serving
nature of what Mike is defending and to ask why large dollars are spent helping
companies like MCI, IBM, ANS and NT play technology catch-up?


If what the NSF, Mike Nelson and Al Gore are continuing to aide and abet were
to become widely understood, there would likely be a backlash against it.  If
there is a backlash Congress is likely to punish the academic and research
community that Mike Roberts represents. It seems to me that Mike is asking
the University community to defend the raid on federal treasury to support
companies like MCI


I suspect some of you saw my post on com-priv the other day about the
incestuous and secretive nature of the award process for the vBNS in
particular and the NSF 93-52 solicitation in general.  If Mike Roberts really
believes that this process was as open and above boards as he says, he is
decieving himself.  Several key players have told me off record that they
agree with my remarks.  That the cooperative agreement process that has indeed
served the academic community so hand$omely in the past cannot be expected to
continue to be used indefinitely into the future - given the increasing
strains on the federal budget and the social and economic problems we face on
all fronts.


Cooperative agreements, as we have seen with NSFnet, are not particularly cost
effective given contractual alternatives.  The concept of cost sharing as it
has been used to justify the federal give away to ANS, IBM and MCI up to this
point is especially specious.  To claim as Steve Wolff has that the intent of
the Boucher ammendment to NSFnet AUP was to further justify the kind of give
away represented by the grant to MCIof the right to sell commercial service to
the vBNS is an arrogant repeat of what caused the blow up documented by the
New York Times on December 20 1991.  I attended the March 12 1991 hearing out
of which that ammendent came and I will assert that this was NOT what Boucher
or the hearing intended.  Moreover several key people have said they agree.


Yes those in the Castle had better wake up.  The peasants *WILL*storm the
gates if they don't!


________________________________________________
Mike says: a concerted effort is being made by several
vendors to block the awards, including Sprint, Alternet and PSI.


Cook: What about MCI Mike?  Why do you leave their protest of ESnet out?


Roberts:  In the particular case of this series of awards, the Networking
Division of the Foundation went to considerable lengths to secure community
input, and the final version of the Solicitation document reflected that
input.


Cook:  And it then went behind closed doors to do its own thing...the rest of
us be damned.


Roberts:  The review and selection process for these awards was unusually
thorough, as was the review by the Science Board itself.


Cook:  Science Board Review was thorough???  In comparison to what?  They
spent 12 hours trying to understand the complexity of something that requires
full time involvement to keep up to speed on instead of the normal 6??


Robert:  However, the current campaign by a noisy few seeking to advance their
own
economic
interests has the potential to delay the transition to the new network
arrange-
ments, as well as to undermine the Foundation's ability to continue to use
the Cooperative Agreement law for purposes of forming public-private
partnerships to advance networking research and development.


Cook:  Considering the embarrassing outcome of the 1990 arrangements I hope
so.  *Financial* interests?  Gee Mike, I happen to think there is a PUBLIC
interest to be served by bringing these issues out into the open.  I can tell
you that compared to the tack that the strictly business oriented newsletters
are taking that although I doing quite ok I sure as hell am not getting rich.


Roberts:  As the events of the past six years have amply demonstrated, the
collaboration
of universities, computer firms, and telecommunications firms in developing
and using advanced networking has had a very tangible positive effect on
the competitive position of our electronics industry, and on the
infrastructure
for research and education.  The use of the Cooperative Agreement Act by the
Foundation to form joint ventures in networking has had precisely the result
intended by Congress when it was passed.


Cook:  to those who have intimate knowledge of the history of who was helped
and who was not, this argument simply will not withstand scrutiny.


Roberts:  It is not only in the national interest, it is in the interests of
the
university
community to promote and defend the collaborative structure and process which
is contained in the Cooperative Agreement Act.


Cook:  Rubbish!  If the university community doesn't wake up, this will blow
up in its face!


PS I will be off line shortly for three days at a washington DC meeting of
telestrategies.  Will be happy to deal with any fallout when I returnlate weds
evening march 23.


___________________________________________________________________
Gordon Cook, Editor Publisher:  COOK Report on Internet -> NREN
431 Greenway Ave, Ewing, NJ 08618
cook () path net                                   (609) 882-2572
Subscriptions: $500 corporate site license; $175 educational & non prof., $85
individ.
___________________________________________________________________