see editors note below -- cook report advertisement

[David Farber <farber () central cis upenn edu>]

Normally I hestitate to send what I consider advertisement (and as the
below first note says what Cook considers advertisement also). I do believe
there is some socially redeaming value :-) in this one , so I am sending it
out. If such restrictions are imposed by Cook in the future, this may be
the last I distribute.


Dave




From: cook () path net (Gordon Cook)
Date: Fri, 29 Oct 1993 02:38:42 GMT
To: farber () central cis upenn edu
Subject:  cook report advertisement (reminder)


it will go up on com-priv momentarily.  It has some very interesting news.
However please post it complete and unedited to interesting people.  If you
feel it must be edited, please do not post it.


From: cook () path net (Gordon Cook)
Date: Fri, 29 Oct 1993 02:42:15 GMT
To: com-priv () psi com
Subject:  November COOK Report Examines Solicitation, ANS Office Closings, etc


The November issue of the COOK Report on Internet -> NREN is published tonight.
In this issue there is a detailed discussion of the ramifications of the current
VBNS solicitation.  An article on the closing by ANS of its regional sales
offices - (quoted in full in this posting).  Part 2 of Computer Networks and
Health Care - A survey of Four Growth Areas (3rd and final part to appear in
December); Jaimie Love's critique of the Clinton NII An Agenda for Action,
debate about Internic, a policy pronouncement from the FNC, and our October 11,
1993 "extra" questioning the legality of the vBNS solicitation.


RESHAPING THE LANDSCAPE
OF THE AMERICAN INTERNET --


AN EXAMINATION OF SOME OF THE FORCES AND UNRESOLVED QUESTIONS BEHIND THE MAY
1993 NSF SOLICITATION FOR A VBNS & OTHER SERVICES


is the lead article.  Several well placed sources went out of their way to share
information with us. As a result we have been able to piece together a rather
detailed analysis of the development and direction of the NSF solicitation for
the vBNS, NAPs, Routing Authority, and inter regional connectivity awards.  Our
7,000 word article analyzes the impact of a likely choice of ATM for the vBNS in
terms of its impact on the further development of ATM technology and on policy
decisions involving cost sharing and resale of backbone bandwidth.


We outline development of a would be legal challenge to the solicitation by
UUNET and PSI and unsuccessful efforts at the end of August to convince the CIX
to embrace the challenge.  (Section 102 (c) 8 of PL 102-194 mandates that, if
network services are available commercially, they must be purchased from the
commercial market.  In particular having the NSF pay for inter-regional
connectivity was challenged on these grounds.)


We look at the plans for vBNS in the context of the more recent announcement by
Sprint of a commercial high speed ATM service for the Minnesota Supercomputer
Center.  We conclude that the vBNS will in all likelihood be a testbed for
internetworking of IP over ATM transport -- something that is not available
commercially in the immediate future, and therefore something that could likely
withstand a legal challenge.


We examine the NAPs from the point of view of the CIX and the mid-levels and
identify some of the complicating questions of peering and policy routing in
bringing large numbers of IP nets into single interchange points.  We get a
better look at CoREN as perhaps the only cost effective mechanism for the
mid-levels to connect to the NAPs.  An irony is that CoREN might well end up
replacing the functionality of the ANS backbone.  We raise some questions about
how well NAPs may or may not scale.


We report on an hour long interview with ANS Vice President Joel Maloff on
October 13.


A few quotes:


> From an interested observer:


"Do you know the rather worn Murphy's Law of Government?  It essentially states:
Whatever government sets out to do, it accomplishes the opposite.  My thought at
this point is that indeed the NSF, in attempting to satisfy (or at least
partially appease or mollify) everyone, is deep in the process of making
everyone very profoundly unsatisfied and, even worse, making the proposition of
the global internetwork vastly less workable and yet more costly to all.


You raise the NAP issue and I'd say that this feels to me like a profoundly
fractious problem about to raise its ugly head.  Perhaps my technical innocence
is showing, but it seems that the pressure is on NSF to fund at least five or
six or even eight NAPs to avoid the political pain of having only three (East,
West, middle) attachment points.  Of course, the greater the number of NAPs the
more mammoth the routing problems, the greater the costs for the national
carriers, the higher the prices charged to regional and local service providers,
and so on.  But my point here is this, the momentous promise and potential
contained in the concept of a global internetwork may be unwittingly designed
out of practical existence on the drawing board of political expediency.  Of
course, having done all the R&D and spent much of the early money, the U.S.
could always "self-destruct" and wait for the Europeans and the Japanese and so
forth to develop this into a commercial undertaking."


__________________
Connecting directly to each NAP would likely be prohibitively expensive for a
mid-level.  And with over 30 mid-levels alone the complexities of peering all of
them at each NAP could become excessive.  The cost of trying to connect to a
single commercial services provider to reach all NAPs is uncertain.  When the
mid-levels realized what they were about to be hit with, it is understandable
that they blanched.  Thus, at the last minute, we had an Inter-regional
connectivity component added to the solicitation.


_______________
Will the kludging of all this together mean higher prices for ordinary network
users?  Will it mean measured usage charges for the ordinary grassroots user?
It seems quite possible.  In this unregulated brave new world we see no one
looking out for grassroots interests.  The administration talks a good game but
so far has yet to move from vision to policy.  The NSF carefully trumpets its
disengagement from paying for academic use of the ordinary network.  Yet in
total money spent it is getting ready to move from a sum of 12 million a year to
18 million - a 50% increase that is hard to equate with disengagement.  (It has
been pointed out that considering the growth of the over all network the
proportion accounted for by NSF spending is declining.)  The suspicion remains
that federal involvement is set to, under the guise of cost sharing, pick market
place winners.  It would be very surprising not to see MERIT and ANS get another
six month extension.  And the whole mess as usual is so complicated that, even
for one who follows it full time, it remains difficult to tell "who's on first."


_______________


To make sure nothing is seen as taken out of context we post the following in
full:


FACING CONTINUED FINANCIAL LOSSES ANS CO+RE CLOSES WASHINGTON, CALIFORNIA,
COLORADO AND MASSACHUSETTS SALES OFFICES


On October 13th, Joel Maloff told the COOK Report that the ANS $3,849,000 loss
for the fiscal year that ended on September 30 1992 was on target with the
projections of ANS's business plan.  We have since learned that another fiscal
year ended on September 30 1993 with sales of network connections failing to
meet even 50% of projections from the year before and apparently another
substantial monetary loss.  This time the loss was no longer "on target."  We
have been told by reliable sources that the Directors of ANS met at the end of
September and decided that cost cutting had to be done.  A decision was made to
close down all regional sales offices (Seattle, WA, Pleasanton, CA, Colorado
Springs CO, and Boston, MA).


The same sources say that ANS Directors of Client Services in the three Western
offices are being terminated.  Marketing leads are being transitioned to Reston
VA where a national telemarketing effort will be maintained.  Reston also houses
Interlock and other network security development work as well as a Director for
East Coast sales.  The ANS Office in Ann Arbor Michigan headed by Maloff and
Percey Foster will remain open.


On Monday October 25, we reached Peter O'Neil who declined to comment saying
that the regional offices did not fall under his authority.  We tried to reach
the other four offices and succeeded in getting through only to Jim Langer in
Pleasanton CA.  Langer said that ANS was "looking at a number of different
avenues by which to best focus our marketing" and acknowledged "there had been
some changes."  But he added that in the meantime "I am still here, same time,
same station."  He suggested we ask Joel Maloff for comment.  We did and we
print Joel's reply in full.


Gordon;


As usual, you have some elements of truth but have missed the real picture and
the truth.


For the record, ANS CO+RE has reorganized our Client Services department, but it
was not at the direction of the Board of Directors. As in any business, you must
evaluate what is effective and what isn't. As a result, I evaluated my own set
of responsibilities, made known my plans to our senior management team,
convinced them of the appropriateness of my actions, and implemented them.


This has included an expansion of our external field sales personnel as well as
an expansion of our Inside sales groups. As you might expect, I am not willing
to share numbers with you, as I consider that proprietary.


As the one responsible for sales, I must always search for ways to do better.
Please do not construe this as having done "bad". Our revenues, number of
clients, and number of client sites continues to grow at an advanced rate(again,
I give you no numbers - it is a competitive world). You must also understand
that - assuming we made only 75% of our sales objectives last fiscal year -
these were objectives or targets - not neccessities. And by the way, for
start-up's to make that percentage of objective means we did some things right
and might want to build on those successes.


I am very pleased with the portions of our organization that remain, and am
exceptionally excited about the new expansion plans that we have in place.


Joel


_______________________________________________________________
Gordon Cook, Editor Publisher:  COOK Report on Internet -> NREN
431 Greenway Ave, Ewing, NJ 08618
cook () path net                                   (609) 882-2572
_______________________________________________________________