[ih] History

Karl Auerbach karl at cavebear.com
Tue Sep 15 13:10:00 PDT 2020


Yes, it made it to the lists, but as an extension posted by Joe Touch on 
Sept 12.

I could send it again - but my copy has "the bad word" so I'd have to 
edit it. ;-)

The Interop saga is, to my thinking, quite important to the success of 
the Internet.

The show was an important place for two main reasons:

    - It was a gathering place where real tech folks could (and were 
required to) pound out flaws in specifications and implementations.  It 
replaced the old Bake Offs.  The early shows backed marketing with 
substance - there always were a large number of RFC authors in the 
background.

    - It served as a Procrustean metronome-of-doom that forced vendors 
to actually produce working products by specific dates.

There were lots of other reasons ranging from the fact that the show was 
a great place to learn a lot about how (at the time) large enterprise 
networks worked and failed to the fact that it was downright a lot of 
fun (at least to those of us who built and ran it.)  Those of in the 
core group got our hands on a prodigious amount of expensive gear from 
many vendors and we weren't shy about giving feedback, often negative, 
to the vendors.

And some tiny reasons grew - I remember a lunch at the Monterey 
conference where you and Craig Partridge were bemoaning the lack of 
means to monitor and manage the growing Internet.  That became fuel that 
powered the development of the various network management protocols - 
HEMS, CMIP/CMOT, and SNMP that we used in later shows.

The show was an early attractor of attacks.  I remember working with 
Carl Malamud to set up a bunch of NCD X terminals at one of the San Jose 
shows and being surprised when an remote attacker put up a 
username/password capture faux login screen.  Nothing surprising these 
days, but it was new back then.

And there was a lot of room to play - That's why we did the first 
Internet toasters (eventually with a robot to insert/remove the bread), 
juke box, weather station, talking bear, etherphones, walkabout TV 
station, USB/iSCSI/Wi-Fi RAID-5, ... and who can ever forget "Fluffy, 
the inflatable sheep" (there's more to that description but it is a bit 
"too colorful".)

That's not to mention all the small events, such as mbone shows and our 
impromptu rafting trips on the Youghiogheny river (that's where I met 
one of our now well known net security people by climbing through the 
window of her car) to some grand parties (including one three day long 
event at my house in Santa Cruz and nearby beaches.)  And of course 
there was that moment of fear and doubt when in Las Vegas we attached a 
remote location and saw traffic that wasn't ours - and we thought that 
we had accidentally tapped into one of the casino networks and we 
expected Guido from Baltimore to show up (turned out to be traffic from 
a show registration contractor.)   (A bunch - probably 15+ - of us did 
get mugged in front of the White House in DC - I can't imagine what kind 
of stupid mugger would go after a bunch of geeks with Motorola radios 
squawking away and within sight of the Secret Service.)

Some parts were unseen - like how we sometimes did our own version of TV 
station signoff - we all gathered in the NOC, sang the Star Spangled 
Banner, and turned off the power switches to the routers.

Walking into a giant convention center, like the Atlanta Congress 
center, knowing that we had a couple of days to install a huge network 
with thousands upon thousands of nodes, spanning many huge spaces and 
meeting rooms, crossing railroads, crossing roads and freeways into 
hotels, and with multiple external providers - that was scary, but fun.

And much of the hardest but invisible work was done during the weeks and 
moths prior to the show when we hot-staged everything, including the 
cable plant, in a warehouse.  We spent months planning everything from 
the physical layout to address deployment to DNS to complex redundant 
routing.  Then we built it.   It was there were we did some of the more 
stressful things, like doing power failure testing.  Then we packed it 
up and shipped it to the convention center.

We have a lot of photos.  I wish we had the raw beta tapes that Linda 
Feferman shot; only a tiny part of her material made it onto that 
video.  It would be nice if that raw material could go into the Internet 
Archive collection.

         --karl--





More information about the Internet-history mailing list