[ih] Fw: History from 1960s to 2025 (role of NCAR and other labs)

Thu Dec 18 10:36:02 PST 2025

Hi Alex:

I don't know the full sequence as I joined the CSNET technical team after
BBN took over.  But the rough background, as I understand it is:

   - Initially, NSF awarded CSNET to someone (I'm guessing UCAR) to oversee
   two sites that created the initial CSNET - there were machines on both
   sides of the US (I think one at Rand  on the west -- don't recall if east
   coast was UDel or somewhere else -- Wisconsin was also involved?).
   - That divided management got CSNET up and going but proved
   operationally painful and so it was decided to consolidate the contract
   with one experienced corporate provider -- which ended up being BBN (and it
   may have been BBN had been considered in the earlier round, per your note
   -- I don't know).
   - The CSNET machines were shipped from the two sites to BBN, which then
   operated CSNET for the rest of its lifetime.

Craig

On Thu, Dec 18, 2025 at 11:22 AM Alexander McKenzie via Internet-history <
internet-history at elists.isoc.org> wrote:

> I think an interesting question is " *Why* did UCAR have the contract from
> NSF to operate (oversee) CSNET?" It is my understanding that NSF wanted to
> award a contract to BBN (because BBN was already operating ARPAnet) to
> operate CSNET, but that NSF rules did not allow giving (or made it
> extremely difficult to give) a contract to a commercial company to build or
> operate anything; NSF's charter was to sponsor research at "research
> institutions."  UCAR was a "research institution" that already was
> receiving NSF funding and was agreeable to do networking research for NSF
> by writing a contract to a commercial company after going through the
> official procurement process of soliciting and evaluating proposals. Being
> from BBN, I like to believe BBN's proposal to UCAR was the best on its
> merits, but it is possible selection pressure was exhibited by NSF; maybe
> someone closer to the process can explain more.  But I assume NSF had
> several institutions other than UCAR that could have been asked to work on
> this problem, so the above summary of the procedure doesn't answer the "Why
> UCAR?" question.
>
> Cheers,
> Alex
>
> On Thu, Dec 18, 2025 at 12:55 PM Alex McKenzie <amckenzie3 at yahoo.com>
> wrote:
>
> >
> >
> > ----- Forwarded Message -----
> > *From:* Craig Partridge via Internet-history <
> > internet-history at elists.isoc.org>
> > *To:* Guy Almes <galmes at tamu.edu>
> > *Cc:* "internet-history at elists.isoc.org" <
> internet-history at elists.isoc.org
> > >
> > *Sent:* Thursday, December 18, 2025 at 12:10:20 PM EST
> > *Subject:* Re: [ih] History from 1960s to 2025 (role of NCAR and other
> > labs)
> >
> > UCAR had the contract from NSF to operate CSNET (stated perhaps more
> > clearly, CSNET, predecessor /partial-prototype to NSF regionals, was
> > overseen by UCAR).
> > Craig
> >
> > On Thu, Dec 18, 2025 at 9:54 AM Guy Almes via Internet-history <
> > internet-history at elists.isoc.org> wrote:
> >
> > > Jack,
> > >    This is all so valuable.  It touches on the community to be
> > > connected, but aims primarily at the evolution of technical
> foundations.
> > >    I'm going to aim at the community part, based in part on a recent
> > > news story.
> > >
> > > On 12/17/25 5:17 PM, Jack Haverty via Internet-history wrote:
> > > > The January/February 2026 issue of Foreign Affairs contains an
> article
> > > > titled "How China Wins The Future".  ...
> > > >
> > > > Here's my thoughts -- based of course only on my personal experience.
> > > > I'd love to know what I got wrong or missed.
> > > >
> > > > - 1960s: Licklider creates his vision of Intergalactic Network; ARPA
> > > > creates the Information Processing Techniques Office (IPTO), which
> > > > initiates the creation of ARPANET.
> > > Note that it starts with a community to be connected.  For Licklider,
> it
> > > was ARPA and the community of scientists funded by ARPA.  This could be
> > > trivialized by focusing on the menagerie of computer terminals and
> > > modems that Lick had to log in to various computers, but it was surely
> > > deeper.  One historic healthy outcome of the postwar United States was
> > > the rise of systematic federal funding of science.  Whether researchers
> > > in various fields were aware of it, there was a rapidly emerging
> science
> > > community that shared federal funding, but also shared a growing need
> > > for effective collaboration among scientists in various disciplines and
> > > various localities.
> > >    Reflecting on the impact of this on the American and international
> > > university research communities, I'd sometimes playfully note that the
> > > US research university community was actually a very odd highly
> > > decentralized organization.
> > >    (At a meeting of networking leaders in Ireland circa 2001, a speaker
> > > noted that there were elements of this present in the medieval
> > > university world.  Patterns such as sending your best students to get
> > > their advanced degrees in other universities and such as everyone
> > > speaking a common second language (Latin then, English now), were
> > > significant.  But I'll return to the special energy behind the postwar
> > > situation.)
> > >    One special player in this is NCAR.
> > >    I am not an expert on NCAR and would very much like to hear from
> > > others who know more about the role it played and is playing.
> > >
> > >    In 1987, when I was at Rice University cobbling together one of the
> > > NSFnet-related "regional networks", we were suffering from a highly
> > > congested 56kb/s ARPAnet while waiting for the promised T1-based NSFnet
> > > Backbone.  So, to complement our ARPAnet connection, NSF kindly set up
> a
> > > 56kb/s connection of the prototype Fuzzball-based NSFnet backbone.  The
> > > specific connection was to the backbone node at NCAR in Boulder,
> > > Colorado.  I had not been aware of NCAR much, but as this intense
> period
> > > of Internet building continued, I became more and more aware of it.
> > >    Evidently, NCAR was a creature of UCAR, the University Corporation
> > > for Atmospheric Research.  UCAR, in turn, was a creature of atmospheric
> > > science departments from across the country.  I do not pretend to
> > > understand the bureaucratic details, but a few things were clear:
> > > <> these departments and researchers had a deep need to share data,
> > > computational resources, and other infrastructure, even acting on this
> > > in 1960, when computer networks were not on the horizon.
> > > <> NCAR has a neat campus "sort of" in Boulder, but on the top of a
> mesa
> > > and not even convenient to get at from Boulder, never mind any other
> > > university in the country.  Its physical situation almost cried out for
> > > a network.
> > > <> It served as a sort of NSF Supercomputer Center even before there
> was
> > > an NSF Supercomputer Center program, even anticipating the 1983 Lax
> > Report.
> > > <> It seemed to have deeply understood the strategic value of
> > > collaboration among atmospheric science departments at a large number
> of
> > > universities.
> > > <> Among other things, this meant that NCAR was a natural contributor
> > > and beneficiary of the use of computing and computer networking in
> > > connecting a nationwide and worldwide atmospheric science community.
> > > <> It also fostered the UCAR Unidata project, which used the Internet
> to
> > > share atmospheric data across universities and labs across the country.
> > > <> I suspect that the 1950s International Geophysical Year had a role
> in
> > > creating UCAR and NCAR.  Oh, and also Sputnik, but that's another
> story.
> > >
> > >    Let me stop there.
> > >    My narrow request is for a better understanding of how NCAR emerged
> > > and how it began to contribute to networking even before any modern
> > > computer networks existed.
> > >    My broader request is for other examples of how specific scientific
> > > communities with their need for effective collaboration and sharing of
> > > data and resources helped create the motivation for building the
> > > Internet, particularly in the 1980s and 1990s.  So community pull as a
> > > complement to our usual story of technology push.
> > >        -- Guy
> > > --
> > > Internet-history mailing list
> > > Internet-history at elists.isoc.org
> > > https://elists.isoc.org/mailman/listinfo/internet-history
> > > -
> > > Unsubscribe:
> > >
> >
> https://app.smartsheet.com/b/form/9b6ef0621638436ab0a9b23cb0668b0b?The%20list%20to%20be%20unsubscribed%20from=Internet-history
> > >
> >
> >
> > --
> > *****
> > Craig Partridge's email account for professional society activities and
> > mailing lists.
> >
> > --
> > Internet-history mailing list
> > Internet-history at elists.isoc.org
> > https://elists.isoc.org/mailman/listinfo/internet-history
> > -
> > Unsubscribe:
> >
> https://app.smartsheet.com/b/form/9b6ef0621638436ab0a9b23cb0668b0b?The%20list%20to%20be%20unsubscribed%20from=Internet-history
> >
> --
> Internet-history mailing list
> Internet-history at elists.isoc.org
> https://elists.isoc.org/mailman/listinfo/internet-history
> -
> Unsubscribe:
> https://app.smartsheet.com/b/form/9b6ef0621638436ab0a9b23cb0668b0b?The%20list%20to%20be%20unsubscribed%20from=Internet-history
>

-- 
*****
Craig Partridge's email account for professional society activities and
mailing lists.