Yup - agree with all this. Question was not the component networks, but
the "Internet". In 1987 we still had the core routing system - EGP - and
that was run centrally by the DOD. It wasn't until around '89-90 (later?)
with the inception of the CIX and the NAPs that we got away from that model.
Sorry for the imprecision in my previous post - but in '87, unless you had
a government sponsor you didn't get on the Internet and the actual
'control' and approval point for new connections was the DOD DCA. The
inception of the NSFNet changed this somewhat but the original connections
to the NSFNet were agreed to by ARPA who then directed DCA to make the
At 08:38 AM 1/24/2002 -0500, vint cerf wrote:
actually DCA had only responsibility for the ARPANET
and MILNET, officially. The rest were the responsibility
of the network operators - usually schools and research labs.
In 1981 the CSNET project brought up non-DoD components
including PhoneNet and the X.25 extension of Internet
developed by Univ Wisconsin. This was an NSF-funded initiative.
The first NSFNET backbone came up in 1986 (Dave Mills' fuzzballs).
At 10:24 PM 1/23/2002 -0800, Michael StJohns wrote:
>Umm... ok - 15 years ago. US DOD, Defense Communications Agency under
an agreement with ARPA ran the Internet (all 20-50 networks of it) and
its "core routing system". In fact the internet was actually called the
"DOD Internet". It wasn't until around '87 that a
non-government sponsored system (e.g. a system that wasn't sponsored or
under contract to DOD, NSF, DOE...) was connected to the ARPANet/MILNET core.
>At 08:57 PM 1/23/2002 -0800, Ari Ollikainen wrote:
>> Could you name the entity (and network) that you claim
>> centrally controlled "the Internet" a decade ago?