SUSTAINING TECHNOLOGY'S LAST STAND
May 6, 2000
SMART Letter #38 - May 6, 2000
Copyright 2000 by David S. Isenberg
isen.com -- "ILOVEU2 -- but I don't use Outlook"
email@example.com -- http://isen.com/ -- 1-888-isen-com
> NGN Ventures -- Sustaining Technology's Last Stand
> NGN Ventures -- Dr. McQuillan, Arno Penzias & Guy Cook
zero in on Optics, Crayons, Ethernet, Instant
Provisioning and Whatever Arno Is.
> A note on TIRKS, PICS, and DCPR by SMART Person Amos Joel
> Quotes of Note:
Arno Penzias on David S. Isenberg
Phil Agre on Competition and Network Effects
David Farber on Technology and the FCC
George Gilder on Market Timing and WDM
> Smart Remarks from SMART People
Arnim Littek on Venture Capital in New Zealand
James McKenna on Nasty Trends in BW & Content
Martin Elton on "Highly Regulated" Swedish Telecom
> Conferences on my Calendar, Copyright Notice, Administrivia
NGN VENTURES -- SUSTAINING TECHNOLOGY'S LAST STAND
Dr. John McQuillan's NGN Ventures conference (Burlingame CA,
April 25-27, 2000) provides the best look at the emerging edge
of networking you can get without signing an non-disclosure
agreement. The companies that present at NGN Ventures are
network infrastructure companies somewhere between their
second round of financing and their "liquidity event" --
usually an acquisition or an initial public offering on the
These companies were at a stage of development when it is
important to sell (i.e., to communicate) what they have and
what they're doing. But apparently the ground rules of the
conference specified that presentations were to be non-sales-
oriented. This set up a disconcerting tension -- often I was
not sure what some of the presenting companies did. Once, Dr.
McQuillan noted that the three companies we just heard "had
very different approaches to the same thing." Apparently he
spoke from inside knowledge (or maybe I'm dense) because I
couldn't tell the three companies apart. Another time, Dr.
McQuillan had to give explicit permission for the presenters
to discuss pricing.
Furthermore, it has been a long time since I've been to a
meeting where "disruptive technology" was such a minor meme.
Indeed the semantics of "Next Generation" (the NG in NGN)
imply orderly transition, as in, "Sonny-boy is a chip off the
old block (well, maybe he's a bit wild, but) he'll grow up,
settle down, get a job and raise a family, just like Dad."
Dr. John McQuillan is a serious, understated man. To him, the
recent stock market gyrations are merely "an inflection
point." Nevertheless, despite the low-hype-factor that
perfused his program, the most dramatic moments of the
conference came when the disruptive genie refused to stay
inside the lamp. For example, in the Metro Optical session
the audience broke into spontaneous applause when Dr.
McQuillan read an audience question that asked, "Why are we
still worrying about SONET when we can do straight Gigabit
Ethernet?" (Also, see the transcribed discussion with Dr.
McQuillan, Guy Cook and Arno Penzias below.)
If Dr. McQuillan weren't so comprehensively knowledgeable and
incredibly brilliant, he'd be boring. He introduced and
summed up every speaker in every session for three entire days
and hardly ever smiled. I kept expecting to be annoyed, but I
wasn't. His insight prevailed.
The "Innovative Communications Providers" session was my
personal conference highlight. Yipes -- where I am the entire
advisory board -- presented its simple, low-cost, scalable
Ethernet-based data service. So did the other two panelists:
Telseon and Broadband Office.
The panel's message was simple and clear; Ethernet's not just
for LANs anymore. As surely as steamships destroyed the great
houses of sail, carriers and equipment providers that do not
understand the ascendancy of Ethernet will be blown away.
Yipes! It is "ip" surrounded by "yes" -- a brilliant piece of
branding. Jerry Parrick, Yipes' CEO, laid out the numbers: IP
over Ethernet beats IP over ATM over SONET in operations by
about ten-to-one and in provisioning by about five-to-one.
Yipes! That's cheap! Yipes! It's scalable, too, especially
compared to SONET! Soon Yipes' customers who need more
bandwidth will click-to-provision, and then click to de-
provision when they're done. (This is a nice customer control
gimmick, but who'd ever want to de-provision bandwidth once
they get it? Especially at Yipes prices.)
The other two panelists were in pretty much the same services
space (about $2500 a month for 100-Mbit metropolitan service),
but their messages were not as well developed as Yipes', in my
entirely objective opinion. But just as you never learn the
finer points of sailing until you race, these companies, at
minimum, serve as trial horses for each other.
At this early stage of market development, the situation
reminds me of the one about how, "One laywer in town starves,
two lawyers in town and they both get rich." It is a rapidly
expanding pie. Yahoo! Yipes! Telseon?
There were several sessions that turned me off because they
ignored the headlong progress we're making, and treated some
of today's interstitial opportunities as if they're going to
last forever. I'm not in synch with that evolutionary thing.
(However, venture capitalists need to hedge their bets. As
Vinhod Khosla said, "I don't have a crystal ball. The way I
survive is by assuming that I don't have the answer.")
The DSL and Cable session was one such boring session. Its
subtitle was "Putting Voice Back into the Equation." It was
about all this special machinery over-engineered for
telephony. Yawn. Can't we get over it? Voice compression is
a PC application already. Session participant Gary Tauss, CEO
TollBridge Technologies, said no. He said, "We are a bet that
there is a phased transition. You can't take the ILECs and
retrain their whole workforce." OK, OK; there's a pretty big
installed base of telephones, but I can't get excited about
(Meanwhile, the problems with DSL -- and there are many, such
as distance, crosstalk, provisioning (especially competitive
provisioning) -- are being addressed by some pretty
interesting startups that were not represented at NGN
The Metro Optical session was another session where I kept
waiting for some action. It was aptly subtitled "SONET's Last
Stand." This is the one where the question about why don't we
go straight to GigE got applause. I can imagine General Sonet
standing in South Dakota asking Dr. McQuillan, "Are those
Ethernet feathers sticking over the crest of that hill?"
[For non-US SMART People, "X's Last Stand" recalls Custer's
Last Stand -- in 1876, General Custer and his troops were
wiped out by Native Americans who believed that white people
such as Custer wanted to subjugate them, destroy their culture
and take over their lands.]
On the whole, the conference was a superb learning experience.
It gave a sweeping context in which to position and compare a
wide variety of new technologies. At the end of the
conference, Telcordia Vice President (and SMART Person) Bob
Lucky summed up his impressions in three words: Provisioning,
Scale, Leadership. Indeed -- if I were venturing my capital,
these would be three excellent squares to put my money on.
FROM NGN VENTURES: "BOARDROOM PERSPECTIVES" BY DR. JOHN
McQUILLAN, GUY COOK AND ARNO PENZIAS.
[The best moment of NGN Ventures occurred in the "Boardroom
Perspective" discussion that followed the Optical Access
session. Right after the session I rushed out to the lobby to
buy the tape so I could share it with the SMART People! I
have edited the dialog to keep it short, readable and hard
hitting. -- David I]
Dr. John McQuillan: It is my pleasure to welcome Arno
Penzias, a venture partner at NEA. Arno, previous to this
position has had a very distinguished career. He headed Bell
Labs Research for many years in the 80's and through the 90's
and is also a Nobel Prize winner in physics. And secondly, Guy
Cook who is Vice President of Internet Services at Qwest . . .
Guy Cook: There is no doubt, from a carrier standpoint, that
we are going all-optical. We are going to get rid of as many
components as we possibly can. We are going to try to keep
things dumb in the core of the network and try to make things
more economic. I think there are some very serious challenges
. . . Qwest, by the way, is building fiber rings in 25 major
metro areas. We will be done in the next 18 months or so. We
will light up about 1000 buildings. We have a partnership with
Bell South and we're just about to take US West . . .
McQuillan: Take? Take??? Oh dear!
Guy: Right. [The U.S. West acquisition] opens a great deal of
opportunity and great deal of complexity . . .
McQuillan: You're building regional networks and metropolitan
networks and you're even building access?
Guy: The distinction between access and regional, or access
and metro gets blurred. When you're running fiber and lighting
up buildings, what's the difference?
McQuillan: Well, what is the difference Arno?
Arno Penzias: Well, the difference is how expensive and
complicated the boxes get. We seem to blame SONET for our own
McQuillan: No, it is *SONETs* fault! (laughter)
Arno: Oh. Good! So we didn't do anything stupid by running
a train in a circle where at each stop all the passengers get
off and have their tickets replaced by another ticket, and
then they all get back on and go to the next stop, and do it
all over again. I would love to give this thing over to the
Crayola Crayon people. You get a box with 100 crayons. Each
one is a different color and whoever is managing the network
just takes a crayon and colors in the network to one of your
1000 office buildings. No boxes, just crayons. A different
color every place. That would certainly save a lot of
McQuillan: Would you like to build that network?
Guy: I'd love to build that network. What you described,
Arno, sounded like simplicity.
Arno: Save software. That is the only thing that doesn't
work [better] by expending bandwidth. You asked me, John, to
explain the difference between hardware and software.
Software is the stuff that never works. It always needs to be
maintained and repaired.
McQuillan: And it gets worse over time.
Arno: And hardware is the rest. (Applause)
McQuillan: And I'd even go further. There is electronics
hardware, which seems to work until you find it doesn't. And
then there is optical hardware, which is just crayons.
Arno: Instead of TIRKS and PICS* and all this other stuff you
just take your crayons. You just color everybody.
McQuillan: [We just heard the previous panel] describing a
very lovely vision, where I could have a 20 or 30 megabit
[service agreement] on my Ethernet interface. I could go to a
website and click [and get] 45 or 65 megabits. It could
happen in 14 seconds instead of 14 weeks. Now, I think there
are boxes that can do that. Can your network and operational
support systems . . . ?
Guy: The answer, quite frankly, is no! This has been nirvana
for the entire communications industry for a long, long time.
We have to get there . . . with a mindset that if we have a
simple solution it is much more likely to work.
McQuillan: So Arno, what do you think about this dream of
being able to click and get more capacity?
Arno: I don't think it is going to be necessary.
McQuillan: Not necessary? You always surprise me! Not
necessary??? (Voice rising.) This is our dream!
Arno: No. It is not your dream. It is a limited dream. I
don't think you have enough imagination! (Laughter)
Seriously! In my lifetime, I've been to a step-by-step switch
-- not even a crossbar switch -- that was still working. And
I heard, "We can't change," and everyone had good reasons. I
always listen to the good reasons. But should we ever ration
bandwidth at all? That is all we are talking about here. On
the way to Guy's POPs you are going to get fast Ethernet, or a
one-gigabit pipe. Ethernet is a very forgiving technology. .
. . But just think in a different way, because . . . I have
been very serious until now.
McQuillan: Yeah. You're deadly serious . . .
Arno: There are four things in the world that are very much
alike -- two are them are money and sex and the other two are
storage and bandwidth. In every single case . . .
McQuillan: (Sarcastically) They are practically the same.
Arno: Yes, because in all four cases, only too much is
McQuillan: Thank you, Arno. Thank you. (Applause) Now we
know where Arno stands.
Arno: Don't ration it. Seriously!
McQuillan: (Sarcastically) I'm there! I think bandwidth
should be free! I think software should be free! Power to
Arno: The point is how you sell it. If you try to count every
bit, the overhead cost -- even in the Telephone Company of a
decade or two decades or ago, we spent much more money on
billing than we did on the entire facility.
Arno: We [didn't lose] that mentality [until one day] when
AT&T Wireless Services . . . said, "We are not going to worry
whether you call Omaha or whether you are calling across
town." . . . The savings and the simplicity made their network
easier and they saved a lot of software. [And, more
importantly, AT&T Wireless Services got a lot of new customers
who used to think that cellular telephony was expensive!
McQuillan: You know, I think we have already lived with this
for a decade with Ethernet in our offices.
McQuillan: In a sense Ethernet is the way God intended
networking to be. (Laughter) Metcalfe didn't invent it the
way it is now. The way it is now is the way God really meant
it to be, which is 100 megabit or gigabit, with structured
wiring and [non-blocking] switches, and it costs less than 100
Arno: And because it is a forgiving medium, it allows itself
to make mistakes.
Arno: And we don't attempt to ration it, we don't attempt to
monitor it and we put all the responsibility on the user at
the desktop. And the users, in return, are willing to take a
lower level of service. If we had to wait until we provided
cell phones with landline quality we wouldn't have anything
Guy: I am happy to buy that story but we have to get the
fundamental economics to make sense, not only in terms of the
cost of the boxes, but for implementation, maintenance, and so
on. Then we can provide those big pipes to everybody. That
would be perfect.
McQuillan: I'm going to close the session now. We would love
to continue this over cocktails tonight. Maybe Arno would
make even *more* sense after a drink or two. For me, I am
going to interpret what we're saying as some kind of cockeyed
endorsement [of] adopting Ethernet at the POPs.
Guy: Yeah, we are going to push as hard as we can to get
McQuillan: So, this is a great opportunity for technologists,
for business people, and for whatever Arno is. (Laughter).
Arno: I'm investing in technology. Right?
(Applause. End of session.)
*SMART Person and old-tyme switching guru Amos Joel sends Arno
"best regards" and writes:
"TIRKS = Trunks Integrated Record Keeping System,
PICS = Plug-in Inventory Control System, a part of
DCPR = Detailed Continuing Property Records."
[These are just three of the hundreds of Operations Support
Systems in the Intelligent Network. (Are these the systems
that helped the RBOCs lose US$18,600,000,000.00 worth of
hardware that they still bill us for? To find out, see
http://www.newnetworks.com/) -- David I]
QUOTE OF NOTE: Arno Penzias on the occasion of David S.
Isenberg's appointment as Bell Labs Distinguished Member of
"Your ability to present your thoughts clearly and
persuasively has solidified your company-wide reputation
as an innovative thinker and a leader in technology
[In the Bell Labs tradition, Arno asked me what he should
write, and since I can present my thoughts clearly and
persuasively . . . ]
QUOTE OF NOTE: Phil Agre, Notes and Recommendations,
Red Rock Eater News Service, April 30, 2000.
"When you have an industry like high technology . . .
where network effects and economies of scale
tend to produce monopolies, competition will
reward those who are good at a particularly fast
and violent form of land-grabbing in the early
stages, long before most anyone is aware that the
market even exists."
QUOTE OF NOTE: David Farber, Chief Technologist of the FCC,
keeper of the IP List, IP: Washington Diary #2, April 27, 2000
"I am, on the average, very impressed with the
staff at the FCC. They do what they believe is
best for the country, recognizing that they are
dealing with powerful industrial players who
often run to the Hill and the courts for
protection if they are pushed too hard. . . .
I was surprised [that] there are essentially no
technical people on the staff of the
Commissioners. . . . If I had one concrete
suggestion I would mandate a Personal Technical
Advisor to each Commissioner as they have a Legal
and Economic Advisor. How they can make informed
decisions in this rapidly evolving technology
business without that astonishes me."
QUOTE OF NOTE: George Gilder, Gilder Technology Report,
"We ourselves don't do price and timing, leaving
that to higher level thinkers. We prefer the
easy stuff, like the physics of WDM."
Smart Remarks from SMART People
From: "Arnim Littek" [firstname.lastname@example.org]
"In the last six months, there's been a drastic
turnaround, at least on the surface, of the
nature of investing in New Zealand. For a long
time, venture capital was something that happened
in the US. Kiwi investors were classified as too
risk-averse. But, of late, every Tom, Dick and
Harriet is announcing yet another venture capital
fund. They seem to be coming out of the woodwork
like someone's discovered that nothing ventured,
nothing gained. . . . Even in this
technological backwater, the shoals are shifting
with the global current."
[It must be a joy to be a VC in NZ, where the only dangerous
thing in the whole country is the Katipo Spider. -- David I]
From: James McKenna [email@example.com]
"I'm seeing the nasty trend to bundle bandwidth
and content in more places than the cable
industry. A few brief examples:
"Ebay -- Want an image on your Ebay auction? Got
to host it somewhere--but that somewhere is not
your local ISP. "Sure, we'll host your COMMERCIAL
content, if you pay for a COMMERCIAL hosting
package." Suddenly every transaction is all-caps
COMMERCIAL. I thought the Net was unbundling
commerce from big business, but it seems that's
not so. On this micro scale, the effect is that
millions of ordinary online humans are now out of
the running on Ebay. So much for the global
bazaar. You STILL need a vendor's license and a
box of capital just to show up.
"Local-access wireless (not) -- Know any wireless
companies who are making it easy for small fry to
send content to users' phones? I don't (not that
I expected to). Thus, WAP content is the same old
same old Yahoo, Mapquest, CNN, blah-blah.
"Universal broadband this year, I mean next year,
I mean . . . If there weren't an unholy marriage be-
tween content-distribution mega corporations and
infrastructure mega corporations, don't you think
this one would have been sorted out five years
ago? Instead cable access is stalled, DSL is
stalled, because the only thing they really care
about is protecting their content. . . . If
Napster is giving people hemorrhages when nearly
everyone who's on is on a dialup, is it any
wonder that universal broadband is always over
the horizon? . . . We're talking about the end
of the fable of Internet empowerment because
empowerment means the big checks will start
[James -- Don't assume a conspiracy when incompetence,
maladaptive corporate culture and The Innovator's Dilemma are
reasonable explanations. If there's an organized conspiracy
to hold back DSL and Cable, its a "Spy Who Couldn't Shoot
Straight" story. The content thing, on the other hand, is a
different kettle of fish. But don't blame the companies here
either. They just don't know any other way to (try to) do
things. -- David I]
From Martin Elton [firstname.lastname@example.org]
[Martin takes on the issue that Barbara Esbin and I dance
around (in SMART Letter #37) regarding whether Swedish telecom
is "highly regulated":]
Since the introduction of the telephone more than
a century ago, the Swedes have followed a
distinctive approach to regulation. Throughout
Europe most telecommunications companies had
their monopolies protected by law and regulation.
There was never such protection in Sweden. Quite
possibly, its light regulation contributed to the
(two-year) lead over the U.S. in cellular
Sweden seems to be off to a good start with DSL.
Currently a number of companies there are
offering two-way 2 Mbit/sec. service at a flat
fee of about $25 (US) per month, plus a one-time
charge of about $250.
In cable television, there has never been
regulation to protect the monopoly status of
cable television companies in Sweden. As a
result, there have been five or six competing
cable companies in Stockholm for a long time.
(So much for the argument that cable is a natural
It seems to me that the US cable industry wants
to have it both ways; to enjoy being a protected
monopoly, but to avoid the regulatory burden of
this status. It's time to stand up to the
industry's threat that it won't invest in
infrastructure if regulation interferes with its
CONFERENCES ON MY CALENDAR
May 7-12, 2000. Birmingham UK. World Telecommunications
Congress. I am an invited speaker for the session entitled,
"What's your network IQ?" Answer: Too high. For info, see
May 23-26, 2000. Laguna Niguel CA. VORTEX. It looks like
I'll be co-kibitzing the VC panel with Bob Metcalfe, then
talking about something (tbd) for a few minutes after that.
For more info, see http://vortex2000.com/
May 27, 2000. San Diego CA. Porter Stansberry's Pirate
Investor Conference. Avast me hearties! The bucko tar can
sail to windward. Don Luskin (CEO MetaMarkets) will be aboard
too! Sign the log at (www.pirateinvestor.com), or call
800-433-1528. Pre-registration $99, or register
at the gangplank for $149.
June 7-10, 2000. Toronto ON. TED CITY. My only role here is
as a paying member of the audience, but I think that Richard
Saul Wurman does a real job with his TED conferences -- every
one I have been to has had deep lasting impact.
June 26-27, 2000. New York City. Entertainment Internet
2000. I'll be on a panel with some of my favorite fiber and
bandwidth companies. The website is thin, but there's info
there -- http://www.imn.org/2000/a245 or call 212-336-6000.
September 13-15, 2000. Lake Tahoe CA. TELECOSM. Featuring
George Gilder, Clayton Christensen, yours truly, and a cast of
geniuses, troublemakers, and people who got rich by listening
to George. This thing sells out, folks -- a word to the SMART.
September 22-24, 2000. Woods Hole MA. An equinoctial weekend
of deliberation for SMART People here beside the rising tide.
Details soon -- but mark your calendars now.
COPYRIGHT NOTICE: Redistribution of this document, or any
part of it, is permitted for non-commercial purposes,
provided that the two lines below are reproduced with it:
Copyright 2000 by David S. Isenberg
email@example.com -- http://www.isen.com/ -- 1-888-isen-com
[to subscribe to the SMART Letter, please send a brief,
PERSONAL statement to firstname.lastname@example.org (put "SMART" in the
Subject field) saying who you are, what you do, maybe who
you work for, maybe how you see your work connecting to mine,
and why you are interested in joining the SMART List.]
[to unsubscribe to the SMART List, send a brief unsubscribe
message to email@example.com]
[for past SMART Letters, see
[Policy on reader contributions: Write to me. I won't quote
you without your explicitly stated permission. If you're
writing to me for inclusion in the SMART Letter, *please* say
so. I'll probably edit your writing for brevity and clarity.
If you ask for anonymity, you'll get it. ]
David S. Isenberg firstname.lastname@example.org
isen.com, inc. 888-isen-com
-- The brains behind the Stupid Network --