[ih] NO "settlements" as part of Internet History

[Guy Almes]

Geoff, Brian, Vint, et al.,
   Ah, settlements, ANS, the CIX, the NSF AUP, circa 1991.
   As Brian notes, I was a minor player since, in early 1991, I 
transitioned from the university world (where I'd run an NSFnet-related 
regional network in Texas) to working at ANS (the "joint venture" that 
John Markoff refers to in the NYT article below.

   As with the more interesting (to me) topic of transition from EGP to 
BGP, I'll try to emphasize the historical part and not attempt to be 
saying anything that might be part of a currently controversial topic.

   One preface: this was certainly an era of "exponential growth", both 
traffic and the number of sites connected and other measures were 
doubling every few months.
   And one property of a system experiencing rapid exponential growth is 
that, along with rapid quantitative changes come what seem like 
substantial qualitative changes.
   And this makes communication and cooperation and other things, in a 
word, hard.

   Anyway, one thing that I perceived in 1991 was the value of what was 
sometimes called the "NSFnet model" during the late 1980s -- with 
campuses connected to (usually) disjoint state/regional networks and 
with these all connected to a single very fast backbone.  Between summer 
1988 and the 1991 period of the Markwood article the advantages of that 
model were quite evident.  For example, the NSFnet backbone was using T1 
circuits in 1988 while most of the connections to campuses used 56kb/s 
leased lines.
   The factor of 24 increase in backbone capacity was key to the rapid 
exponential growth of the late 1980s.  And, by the way, it fueled 
innovation by raising the bar on scalable applications with growing 
end-to-end performance.
   As the regionals began to make greater use of T1 circuits, the need 
for the NSFnet backbone to grow to T3 was evident, but the thinking (and 
here I am repeating what I recall being said at the time -- I was not 
part of this planning) was that NSF could not pay for 100% of the cost 
of the needed T3 circuits, but they could pay a large part of that.  And 
this would especially work if some other source (e.g., non-agency and 
perhaps non-research/education-community) could pay for the balance.
   Similarly, ANS was formed, in part, to organize how the backbone 
could move from T1 circuits to T3 circuits.  The idea was that this 
would preserve the scalability and rapid increase in backbone capacity 
and thus spur growth and application innovation.  The new commercial 
networks could then connect to and benefit from this super-fast backbone 
and also benefit.
   For several different reasons, some of the pioneering commercial ISPs 
did not see things that way.  They were happy to build their own 
T1-based national networks and thus be "peer" national networks with the 
NSFnet/ANS backbone rather than build a set of regional commercial 
networks that would make use of the new T3 backbone.

   At the time, this seemed to me (I won't speak for my ANS colleagues 
or any others) as unfortunate and maybe irrational.  Unfortunate in that 
these new commercial networks would paying more for their multiple 
T1-based infrastructure and not be "contributing" in modest ways to the 
economies of scale of a shared super-fast (T3) backbone.

   The pioneer commercial networks, especially PSI and UUnet/Alternet, 
wanted to interconnect with "zero settlements" (i.e., paying nothing 
toward the costs of the T3 backbone).  This seemed unfair to us at ANS.
   And ANS wanted to establish a workable way for commercial traffic to 
contribute toward the shared 'backbone'.  This seemed unfair to the 
pioneer commercial ISPs.

   The term "settlements" was indeed used, though I don't recall anyone 
suggesting anything like the X.25/X.75 model of per-"call" money 
changing hands.

   This is all 30 years ago and so much has changed.
   Trying to be fair to all, I think it comes down to ANS believing in 
the value of a shared "backbone" layer, even with competition for 
commercial customers, and with figuring out how to amicably figure out 
how to arrange for the costs of the "backbone" to be shared between the 
research/education world and the commercial world.  It should have been 
a "win-win" but was a show-stopper.
   And the pioneer commercial networks believed in multiple independent 
national networks, peering with "zero settlements" at exchange points.

   In many respects, the pros and cons of backbone/hierarchy models and 
exchange point models remain to this day.  I won't say more here, mainly 
to keep focused on the history of the 1991 era.

   In hindsight, I think there were several hard lessons that we were 
all learning.
<> While the advantages of the backbone/hierarchy model remain, it was 
natural to overemphasize those advantages in 1991.  Over the multiple 
decades since the 1984ish deregulation of long distance (inter-LATA) 
telecommunication services, the relative cost of the "long distance" 
part of end-to-end communication costs has decreased.
<> The non-technical difficulties with asking the pioneer commercial 
networks to accept a notion of being upstream or downstream from a 
backbone provider, even apart from any technical or economic or 
operational issues, was great than I expected.
<> While the Internet of circa 1990 was mostly based on universities and 
research labs and the new commercial players (both providers and 
customers) wanted access to it, things were changing rapidly.  By 1995, 
with the recompetition for NSF funding for university-based regionals, 
the university world realized (if no before) than they were no longer 
the center of the Internet, but that they were, in a word, customers.

   So there was probably a mix of (a) companies striving for economic/ 
competitive advantage and (b) companies trying to navigate a set of 
paradigm shifts that few understood clearly.

   One closing comment; I vividly recall reading the John Markwood NYT 
article below when it was new.  One quote that makes me grimace / smile 
even now is Bill Shrader's [["It's like taking a Federal park and giving 
it to K Mart," Mr. Schrader said.]]  Given the origins of PSI with the 
NSFnet-based regional in New York state, this struck me as extreme 
chutzpah.  I recall wondering why it was bad with a Federal park, but 
not with a state park.
   But that was, as I say, 30 years ago.  At the time, it was not a 
pretty picture, but time heals all wounds?

   Cheers,
	-- Guy

On 9/10/20 8:21 PM, the keyboard of geoff goodfellow via 
Internet-history wrote:
> On Thu, Sep 10, 2020 at 10:42 AM *Jack Haverty via Internet-history
> <internet-history at elists.isoc.org <internet-history at elists.isoc.org>> wrote*
> :
> 
>> CCITT was working on X.25, and creating X.75, to interconnect their
>> networks.  It was a natural evolution of the PTT's prior interconnection
>> of their telephone networks.   Later, as DDN marched down the X.25 path,
>> the subsequent government Internet might have ended up based on
>> X.25/X.75.   If it worked.
>>
> 
> part and parcel of the X.25 and X.75 networking shibboleth was the notion
> of "settlements" -- just like with the PTT's interconnection of their
> telephone networks -- where The Fee for/of Transit was split along a
> "calls" participants/networks transited.
> 
> IIRC, Advanced Network and Services (ANS) tried to "imprint" the
> settlements model on the fledgling Internet -- with them In The Middle
> collecting a fee for transit (for commercial traffic) between networks as
> well as for interconnection with The Budding commercial Internet upstarts
> (PSI, ALTERNET, etc.) the resulting in the founding of the CIX, viz.:
> 
> 
> *Data Network Raises Monopoly Fear*
> By JOHN MARKOFF
> The New York Times
> December 19, 1991
> https://urldefense.com/v3/__http://www.nytimes.com/1991__;!!KwNVnqRv!RmH212ZDH6MdnCAAMUUO7OLJexwNzF_HWvEP5LJV-CL_IWJh4Qof2_qA5z-Yqw$
> /12/19/business/data-network-raises-monopoly-fear.html
> 
> Soon after President Bush signed legislation calling for the creation of a
> nationwide computer data "superhighway," a debate has erupted over whether
> the Government gave an unfair advantage to a joint venture of I.B.M. MCI
> that built and manages a key part of the network.
> 
> The venture, known as Advanced Network and Services, manages a network
> called NSFnet, which connects hundreds of research centers and
> universities. NSFnet also manages links to dozens of other countries. All
> these networks are collectively known as Internet.
> 
> Some private competitors say Advanced Network and Services uses its favored
> position to squeeze them out of the data-transmission market by
> establishing rules that make it difficult to connect to NSFnet.
> 
> *Traffic Has Doubled*
> 
> NSFnet was founded by the National Science Foundation, a Federal agency,
> and is composed of leased telephone lines that link special computers
> called routers, which transmit packages of data to three million users in
> 33 countries. Data traffic over the NSFnet backbone has doubled in the last
> year.
> 
> The Government wants to develop a national data highway for electronic
> commerce, digital video transmissions to homes and vast electronic
> libraries that could be drawn on by the nation's schools.
> 
> Advanced Network and Services, based in Elmsford, N.Y., was set up last
> year as a nonprofit corporation with $10 million from the International
> Business Machines Corporation and the MCI Communications Corporation.
> Earlier this year it set up a for-profit subsidiary, called ANS CO+RE
> (pronounced core), to sell computer network services. That led some
> competitors to complain that Advanced Network and Services would be able to
> compete unfairly because of its arrangement with the Government.
> 
> *Fear Loss of Innovation*
> 
> People involved in planning for a national data network say it is essential
> to provide for fair competition, which will lead rival companies to offer
> creative and entrepreneurial services in the hope of building market share.
> Without competiton, they say, the Government will have created a monopoly
> that has little incentive to innovate.
> 
> "This is the first major communication business to be born under the
> deregulation era," said David Farber, a computer scientist at the
> University of Pennsylvania and a pioneer in data networking. "This hasn't
> happened since the growth of the telephone industry. You want it to be a
> business that doesn't repeat the errors of the past."
> 
> In recent years, the National Science Foundation has tried to shift its
> operations and ownership of NSFnet to Advanced Network and Services. And it
> will try to establish competition through contracts for networks to compete
> with NSFnet next year.
> 
> But there is no level playing field, complained William L. Schrader,
> president of Performance Systems International Inc., a Reston, Va., company
> that provides commercial data connections to Internet. He made public two
> letters between officials of Advanced Network and Services and the National
> Science Foundation that he said gave the company unfair control over access
> to the network. The result, he added, was that the Government turned over
> valuable public property to a private company.
> 
> "It's like taking a Federal park and giving it to K Mart," Mr. Schrader
> said. "It's not right, and it isn't going to stand."
> 
> Performance Systems and several other companies have set up an alternative
> to NSFnet, known as a CIX. Mr. Schrader said his company and the venture of
> I.B.M. and MCI were competing for the same customers but unlike his rival
> he lacked a Federal subsidy. He said he might ask the Internal Revenue
> Service to look at the business relationship between Advanced Network's
> nonprofit and for-profit operations.
> 
> *'Very Competitive Environment'*
> 
> Allan Weis, the president of Advanced Network, disputed that his company
> had an unfair advantage. "It's a very competitive environment right now,"
> he said.
> 
> At the National Science Foundation, Stephen Wolff, director of its
> networking division, said I.B.M. and MCI had overbuilt the network and were
> selling commercial service based on the excess capacity that was available.
> 
> A number of organizations are working informally to settle the dispute.
> 
> "I think it's a mess," said Mitchell D. Kapor, the founder of the Lotus
> Development Corporation and now head of the Electronic Frontier Foundation,
> a public-interest group focusing on public policy issues surrounding data
> networks. "Nobody should have an unfair advantage."
> 
>