[ih] Internet Talk Radio
Karl Auerbach
karl at iwl.com
Wed Jul 22 18:58:56 PDT 2026
As one of the principal members of the Interop show net team from
roughly 1988 through the early 2000's I can attest that we did not
approach the show net as a marketing vehicle.
For us it was very much a "can we make this stuff work" event.
We began in 1988 (or 1987 if you count the Wollongong event in Santa
Clara) with a mad dash to get into the convention center, get it wired,
get out of the way of the teamsters, attach the machines, and turn it
all on - in a span of about 8 hours.
We quickly learned to rent a warehouse and pre-stage the entire network
infrastructure (everything except the stuff in the vendor's booths). We
typically spent several months prepping for a show. I remember that for
one show (in Atlanta) that it took 46 18-wheel truck/trailers to haul
our infrastructure gear. That was just our stuff - no vendor booth stuff.
Many vendors did not interoperate. We let them know they were not
interoperating. We rarely disconnected a vendor - although I do
remember cutting off D-link when they had a router that was spewing
garbage and harming a lot of others. And I remember HP cutting
themselves off by adopting IEEE 802 Ethernet SNAP framing rather than
DIX. We also found problems in various standards - such as when a flaw
was discovered in the FDDI specifications and vendors spent a pre-show
night re-blowing PROMs with new firmware.
We pushed things hard. And that is from the very physical sense (Alex
Latzko and I once used sledges to pound the beejeebers out of a cabinet
to enlarge a cable passage, to clambering through the unseen innards of
building, hotels, parking lots, and railroad tunnels) to pushing things
like twisted pair ethernet to its physical limits or cooking equipment
in a 105F convention center (the centers do not turn on the air
conditioning until the trucks are gone from the floor.)
I remember once when Carl M. and I were setting a bank of public X11
terminals in San Jose when someone from the outside discovered them and
began projecting windows onto them. I think that was my first
experience with an Internet carried attack.
We also pushed in other directions - we had early VoIP gear that
surprised many a telco when we placed an outgoing call tags.
Circa 1997 I created a mobile MBONE camera/microphone setup on a hardhat
from which I participated from the show floor in one of the conferences
with NTIA about the yet-to-be created ICANN - for many in DC it was the
first time they had participated in an Internet hosted conference call.
My setup also concerned some people - with the battery pack I kinda
looked like a mad bomber walking the show floor - and because the camera
was on my hard hat my wife was able to see every booth babe my eyes
landed on.)
We actually tied to make ISO/OSI work and we had IPv6 from rather early on.
It helped a lot that a lot of the iETF leaders who knew specific
protocols and implementations inside and out were there on the show floor.
One of our innovations that has rarely been discussed is how we managed
that stuff. We built an entirely physically distinct network, using
steerable fiber optics (mirrors on piezo-electric steering crystals),
and two-way fiber taps (people did not believe such things existed) so
we could tune into any segment of the entire convention center to
monitor traffic or inject traffic. I built the first Internet "Butt
Set" so we could get working on problems within seconds.
We also quite stressed external packet carriers - we were always dual
homed with our IPv4 45.x.x.x/8 network. When we were coming up we
bounced up and down a lot - so we really stressed out the route flap
damping of our external providers (many of them did not know how to
reset their gear to deal with this - but they got a quick lesson.) And
we moved that 45/8 block around the world often with a span of a couple
of weeks.
I think we also engendered a new world view among the union electricians
at the various venues - many of them were quite unfamiliar with network
wiring (once one person folded our yellow hose ethernet and another cut
it and tried to solder it back together) and learned that fiber optics
did not carry electrical signals. (We eventually brought a couple of
San Francisco based union electricians into the team - that helped
smooth the waters in other cities.)
We also played a lot - I built my machine to inject words into VoIP
streams (my ultimate goal was to elide the words 'no" and "not" from
VoIP conversations). And we built a RAID array out of USB flash drives
controlled by I-SCSI over Wi-Fi. We pushed IP multicast quite hard (and
got it to work acceptably well.) We also set the carpet in the lobby of
the Las Vegas convention center on fire - not our fault.
I claim that we were the world's best commando network team - about 20
to 40 of us all told - we got it in, got it running, kept it running,
and got it out - within hours. (We all worked pretty much 24x7 - the
Interop company had to pour food and coffee into us - and we survived on
short crashes of sleep, alcohol, late night movies in unwitting vendor
booths, and other kinds of social interactions not to be described in
polite company.)
It was hard work. But it was also fun. And perhaps most importantly, we
showed the world that Internet style networking worked and could be used
by people and companies. In that sense I feel that we were an important
part of building acceptance for the then nascent Internet.
--karl--
On 7/22/26 4:44 PM, Jack Haverty via Internet-history wrote:
> Interesting, another part of Internet History. I didn't know about
> the Paris show, but I did later help with a show (probably Oracle's
> "User Group" annual event) at Place de la Defense. A bunch of us
> almost got lost trying to take public transportation from there to
> visit the Louvre, and with only high-school French in our group we
> came very close to ending up in Louvre, France, which is a town quite
> a distance from the Museum.
>
> I've always thought of Interop as the "marketing department" of The
> Internet. I knew Dan from his previous career when he ran the
> datacenter at USC, where lots of ARPA work was carried out. At one
> point I recall that Interop started using a catchphrase of "I know it
> works; I saw it at Interop!" Then at the shows I saw that working in
> real time on the show floor. A horde of people would descend on a
> booth, consisting of the VPs (in suits) and the techies (in jeans and
> sweatshirts). The Suits would listen to the pitch, and turn to their
> techies to get confirmation that whatever was being sold was a good
> idea for their company. The Suits had an environment being deluged
> with personal computers, workstations, and LANs, and still had their
> mainframes too. The Internet tied it all together, and could be
> observed actually working on the Interop floor. Everyone was tired
> of single-vendor designs and waiting for OSI to make it all
> wonderful. TCP worked, you could see it, you could buy it, and all
> your new hires likely had experience with it already. Interop brought
> all that together and connected the techies to the suits to the
> vendors - i.e., the Marketing Department.
>
> After I left BBN and joined Oracle in 1990, I was local to Silicon
> Valley, and Dan recruited me to be on some Interop committee ("Program
> Committee?" can't remember). The committee got involved in figuring
> out what sessions and tutorials would be selected, and our vetting
> focussed on presenters that were techies rather than marketers. I
> think that made a difference in the attendance, encouraging techies to
> come to learn, and bring their suits as well to "see it working at
> Interop". I recall going with Dan to meet the people who ran Moscone
> and being disappointed when they said Interop wasn't big enough to
> book Moscone. Several years later I recall another meeting with the
> Moscone management to inform them that the next Interop would be too
> big for Moscone and would be held in Las Vegas instead. The Internet
> was growing explosively and so was Interop.
>
> Fun times. The history of Interop is an important part of the
> History of The Internet.
>
> /Jack Haverty
>
>
> On 7/22/26 15:48, Carl Malamud wrote:
>> The Paris show was scheduled to be announced and the marketing people
>> decided they had to have a video so they could show the people of
>> france what this was all about. Ole hired a camera crew in San
>> Francisco and we stalked the halls with the video camera looking for
>> people like you. :) I got to play a role in the subsequent show,
>> which was held at La Defence (near the arch!). We were, as was the
>> usually case in those days, the first network ever held in that
>> convention center (we did the same thing at Moscone center). French
>> was my maternal language and I also spoke geek fluently, so I
>> remember helping defuse a big showdown between NOC team members and
>> the convention center security team. I wasn't Interop staff or
>> anything official, wasn't even a member of the noc team, was actually
>> just a freelance writer at the time but there weren't a lot of other
>> press following this, so I had great access.
>>
>> On Wed, Jul 22, 2026 at 2:44 PM Jack Haverty via Internet-history
>> <internet-history at elists.isoc.org> wrote:
>>
>> Wow, Thanks Carl! Just out of curiousity, I found the "Interop
>> Movie"
>> that was filmed at Interop in San Francisco in 1992, but I guess was
>> shown later at Interop France. I had never seen that movie until
>> just
>> now, and was surprised to find that I'm in it!
>>
>> That was the year that Oracle took the MazeWars game to Interop
>> and it
>> seems at some point someone stuck a video camera in my face to hear
>> about what we were doing at Oracle. I don't remember that, and
>> didn't
>> even recognize myself at first -- until my name appeared on the
>> screen.
>> The memories of that show are a bit fuzzy. But I seemed to be
>> fairly
>> articulate, even after spending my first and only 96-hour period
>> with no
>> sleep, working before the show opened on getting everything to work
>> (along with a bunch of other sleep-deprived Oracle engineers).
>>
>> A true historical resource at archive.org <http://archive.org>!
>>
>> /Jack Haverty
>>
>>
>> On 7/22/26 14:11, Carl Malamud via Internet-history wrote:
>> > :) thanks Brian. Here’s a pointer to the full collection:
>> >
>> >
>> https://archive.org/details/RT-FM?sort=date&and%5B%5D=subject%3A%22Publications%22
>> <https://archive.org/details/RT-FM?sort=date&and%5B%5D=subject%3A%22Publications%22>
>> >
>> > On Wed, Jul 22, 2026 at 1:55 PM Brian E Carpenter via
>> Internet-history <
>> > internet-history at elists.isoc.org> wrote:
>> >
>> >> I just realised what a fantastic historical resource Carl
>> Malamud created:
>> >>
>> >> -------- Forwarded Message --------
>> >> Subject: Re: the dangers of April Fool's RFCs....
>> >> Date: Thu, 23 Jul 2026 08:53:28 +1200
>> >> From: Brian E Carpenter <brian.e.carpenter at gmail.com>
>> >> To: Craig Partridge <craig at tereschau.net>, IETF discussion list <
>> >> ietf at ietf.org>
>> >>
>> >> Craig,
>> >>
>> >> BTW, it was probably about a year after April 1, 1992 that Carl
>> Malamud
>> >> started Internet Talk Radio and Geek of the Week. Milo Medin
>> was the first
>> >> geek, interview published May 7, 1993. Carl interviewed quite a
>> number of
>> >> geeks during IETF 27 (Amsterdam, July 1983).
>> >>
>> >> See
>> https://archive.org/search?query=creator%3A%22Internet+Talk+Radio%22
>> >>
>> >> The point is: things move fast on the Internet.
>> >>
>> >> Regards/Ngā mihi
>> >> Brian Carpenter
>> >>
>> >> On 23-Jul-26 07:23, Craig Partridge wrote:
>> >>> Dan Harkin wrote:
>> >>> > You know, I wrote an April Fools I-D suggesting this sort
>> of thing
>> >>> to ensure diversity and I got in trouble!
>> >>>
>> >>> And I wrote an April Fools RFC (1313) that posited the notion
>> of a radio
>> >> station broadcasting audio over the Internet, an outlandish
>> idea we now
>> >> call podcasting!
>> >>> We should, perhaps, compose an (?April Fools?) RFC on the
>> hazards of
>> >> April Fools RFCs becoming reality.
>> >>> Craig
>> >>>
>> >>>
>> >>>
>> >>> --
>> >>> *****
>> >>> 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
>>
>
>
More information about the Internet-history
mailing list