Mark Horton (mark@stargate.COM)
17 Oct 90 16:17:11 EDT (Wed)

vax135 was definitely an early player (there were very good relations
between the vax135 folks who put out UNIX/32V and the ucbvax folks who
got the first beta tape) and I could believe that the Netnews link was
duke<->vax135<->ucbvax. But the email link was definitely via research.

Somewhere I have on magtape a map or two from the spring of 1981,
back when there were only 30 or so machines on the net.

The oldest map I have (paper, 1/1/82) shows research as a leaf, vax135
as a minor hub (links to cornell, ihnss and harpo.) This map shows every
machine that got Netnews one one sheet and was only photo-reduced from
about 11x17.

I do have a 9/21/81 map made by Phil Karn, Mr. ihnss (forerunner of ihnp4),
which is not entirely accurate for the time. It shows chico, ihnss, and
mhtsa as the major hubs, with chico migrating into harpo.

The backbone is first mentioned in the
July 1983 map - it was not mentioned on the Jan 1983 map.
In July 1983 there were both logical maps (showing all hosts) and
geographic maps (showing the biggest host in each city on a US map.)

We did a geographic map only in 1/84, and in 6/84 the only map was a
machine-readable 4 page logical map done by Bill and Karen Shannon -
I had given up on trying to show it on paper, and did the last geographic
map in 6/84. Bill and Karen continued to put out their machine readable
one on the net periodically. Now, of course, Brian Reid puts out his
Postscript stuff.

The first links to Canada appeared on the 1/1/82 map (watmath, utzoo,
and 2 machines behind each.) In 1/83 philabs bridged the atlantic
to connect to mcvax, dutesta, and ukc. By 7/83 a second link from
ukc to vax135 (non-backbone) existed, and a link from sdchema in San
Diego to basservax in Australia (by magtape once a week!) appeared.

I have this stuff on paper, and can send it to one person who is the
keeper of all this.


This page last updated on: Jul 1 09:16