SMART Letter #10
August 12, 1998



            SMART Letter #10 - August 12, 1998

        For Friends and Enemies of the Stupid Network

            Copyright 1998 by David S. Isenberg

      This document may be redistributed provided that  

      the 11 lines containing this notice accompany it. -- -- 1-888-isen-com

      It takes SMART people to design a Stupid Network



ANNOUNCEMENT: America's Network,

will be carrying my monthly column entitled, "Intelligence at 

the Edge,"  beginning September 1.  Despite the magazine's name, 

it is a pretty good rag, with a healthily skeptical attitude 

befitting the journalistic ethic.  Since I don't have *that* many 

good ideas, much of the material will be redundant with the SMART

Letter, but SMART people may find something new, or even some

value in the *rest* of the mag, and most should qualify for the 

free subscription.  



"The Innovator's Dilemma: When New Technologies Cause 

Great Firms to Fail" by Clayton Christensen.  

225 pp.  Boston MA USA.  

Harvard Business School Press, 1997.

Incumbent companies, even well managed ones, rarely bring 

new, discontinuous technology to market because . . . they 

are good at listening to their customers.  This is the 

grist of "The Innovator's Dilemma."  The book's spine is the 

distinction between *sustaining* technologies and 

*disruptive* ones.  Sustaining technologies give us 

better-faster-cheaper in an established marketplace.  

Disruptive technologies open new markets, and deflate or

even destroy old ones.  Christensen's book develops a strong 

body of evidence that incumbent companies almost always lead 

sustaining technology to market, but virtually never lead 

in markets opened by disruptive technologies.  

It works like this:  Suppose you owned a desktop PC in the 

years before laptops were very good or popular.  Now suppose 

a salesman said to you, "I have this fantastic new disk 

drive.  It holds half as much data, it costs twice as much, 

and it is SIX OUNCES LIGHTER!"  

Desktop PC users don't carry their machines around.  They 

don't need a six-ounce lighter computer.  The weight 

reduction isn't "value-added" to them, and they certainly 

won't pay more for it.  They have a different set of 

problems.  They want MORE storage, even if it makes their 

machine heavier. The six-ounce advantage is useful in 

another, discontinuous market space - laptops.

Discontinuous markets start small.  They aren't cash cows 

in their first years.  More importantly, their initial 

customers don't share key values of established-market 

customers.  Portability does not matter in the value matrix 

defined by the established desktop PC market.

An incumbent firm, playing in an established market, must 

listen to its established customers.  At every budget 

cycle, it needs to decide whether to allocate resources to 

the mainstream business (and profit stream) or to a 

marginal, often money-losing "long shot."  Usually, it's a 


New, disruptive technologies become dangerous to incumbents 

when their performance expands into the value matrix 

occupied by the old market.  For example, Christensen 

explains how the 3.5 inch disk, which was a critical 

component of the mid-1980s portable market, gained storage 

capacity to become valued by - and successfully compete in 

- the older, incumbent desktop market.  And how it was the 

death knell for the 5.25 inch drive.  And how incumbent 

market leaders like Seagate and Miniscribe were replaced by 

upstarts like Conner and Rodime.  (Seagate actually 

started, and then killed a 3.5 inch drive effort.  The 

market for 5.25 was robust.  Why change?)

Christensen maps disruptive technologies, and the markets 

they created and served, and the companies that they built 

and destroyed, through several transitions.  The disk drive 

market is his forte.  He maps the disruptive change from 

14-inch disk to 8, from 8 to 5.25, from 5.25 to 3.5-inch 

technology, and from 2.5 to 1.8.  (Curiously, the 

transition from 3.5 to 2.5 was NOT disruptive.  This 

exception actually adds explanatory power to the theory - 

2.5 inch technology, he explains, was fully compatible with 

existing market values.)  To generalize his observations, 

he digs into the mechanical excavator market, mapping the 

transition from cable controls to hydraulic.  And he casts 

the discontinuous history of the steel industry in terms of 

the emergence of mini-mills, and the subsequent death of 

giants like Bethlehem and U.S. Steel.  And he shops for 

supporting evidence at Sears, examining the disruptive 

emergence of discount retailers.

He debunks the notion that incumbents fail because they are 

bloated, bureaucratic, and lazy. The emergence of thin-film 

disk coatings, a sustaining technology, was a heroic 8-

year, US$50M effort, led by incumbents like IBM and DEC.  

Incumbents were also the prime beneficiaries, because the 

new technology was sustaining to the old value matrix.

He also explains the inadequacy of so-called S-Curve 

theory, which sprang from Joseph Schumpeter's notion of 

creative destruction, to explain the emergence of new 

disruptive technology based markets.  The problem is that 

the Y-axis, the measure of value in the space in which the 

"S" is drawn, is simply different on the other side of a 

disruption.  For example, the Y-axis could represent 

storage capacity per dollar or storage capacity per ounce.  

The first would map desktop value, while the second would 

be more relevant to laptops.  

It's hard to tell what will be valued in the next market 

discontinuity. In 1992, the year before Mosaic appeared in 

public, the demand for web browsers was undetectable.  Yet, 

8 years later, they are indispensable to millions of users.  

The book takes a few swags at what might be coming next - 

electric cars to replace internal-combustion based 

vehicles, flash memory to replace disks - but every book 

has its limits.  Not every revolutionary technology defines 

a new market space (e.g., video telephony).  And it is just 

not possible to predict with accuracy what a nonexistent 

marketplace will value.  

For disruptive technologies that have already established a 

foothold, the book provides a powerful analysis.  Take the 

Internet.  The modal phone customer values very different 

properties than the modal Internet user.  Phone companies 

didn't "get" the Internet until Internet Telephony.  But 

since Internet Telephony, IP is hot in the telephone 

industry, especially the equipment sector.  Why?  Perhaps 

because they see Internet Telephony as *sustaining* their 

old value matrix.  In some respects it is cheaper.  Might 

better and faster be just around the corner?

Meanwhile, the initially disruptive Internet is expanding 

into all kinds of established value spaces.  In addition to 

telephony, these include retail, news, investment, postal 

service, advertising, gambling, radio, et cetera.  

Initially, incumbents in each sector will (a) listen to 

their mainstream customers, and (b) treat the new 

technology as if it were sustaining.  My prediction, after 

reading "The Innovator's Dilemma," is that they will miss 

most of the new value that the Internet creates, and that 

new companies will emerge to dominate these new markets.  

In the case of two-way, real-time voice, formerly known as 

telephony, I believe that the value will extend far beyond 

telephony, until voice is just a seamless piece of a 

communications mix that does not resemble telephony as we 

know it today.

What should an incumbent to do when challenged by 

disruptive technology?  Depends on which horn of the 

dilemma it chooses to ride.  Christensen addresses this, 

but this reviewer will let his readers discover these 

delights on their own.  "The Innovator's Dilemma" twisted 

my neck to a new angle. It also threw a few last straws on 

the back of my 12 year AT&T career.  It remains useful as I 

think about strategy with clients.  I believe that its most 

provocative point - that there are structural reasons why 

incumbent companies don't lead technology revolutions - will 

stand the test of time.  



Two articles, one in ACMnetWorker and one in Communications 

Week International, attack the Stupid Network concept!  

In the first, AT&T Executive Vice President Dado Vrsalovic 

(ACM netWorker 2.2, April/May 1998, p. 44-47.) treats the 

networking business as if all customer needs are known and 

all technological improvements are "sustaining."

He keeps coming back to "the kind of network we *really* want"

-- you know, secure, reliable, (yawn) monthly bills.  He says,

"Intelligence, to me, denotes an ability to react and adapt."

How about the ability to discover and create?  He never addresses

the undiscovered need, the innovative process, the new business 

idea, the disruptive effects of technologies upon markets.  OK,

so maybe "react and adapt" is the kind of intelligence they

want in an AT&T EVP.

The second article, "Why Stupid Networks need a little 

intelligence," by Ericsson VP Per Jomer (CWI 208, 20 July 1998, 

p. 9 or

is a briefer, but more satisfying discussion.  He gets that the

separation of "the communication layer" from the user and "the 

service layer" will facilitate innovation.  He believes that 

Ericsson will be able to provide significant value at "the 

service layer." Maybe he's right, maybe there is some 

"sustaining" juice to be squeezed from the combination of 

IN and SN, but if Christensen is right, they'll miss the big 


Jomer opens by calling "The Stupid Network" a buzzword.  

I chuckle at how what was once a modest essay with a goofy title,

became a "cult classic," a new career, and an industry buzzword.

Long and strange, indeed.


YEAR 2000 STORY: A few weeks ago I was talking to a ship's

captain on a major, passenger-carrying boatline.  He is a 

computer hobbyist.  In fact, he is the first person who ever 

showed me Microsoft Flight Simulator.  So I was

surprised to hear him opine that the Y2K Problem could be

"cleared up with 16 lines of assembly code."  I asked him

how many microprocessors were on the ship he captained.

He said, "Oh, fourteen, maybe sixteen."  I asked if that

included the engine room, and he said that there were

"a lot more down there."  I asked who made them and he

said, "Uh . . . Motorola . . . .and . . . . "  I am happy 

to report that he "got it" and I can't wait to hear how

he's following up!


YEAR 2000 SCENARIOS: The Year 2000 Scenarios that I 

generated last fall and wrote up in SMART Letter #7 

( still seem robust!

Despite the scenarist's efforts to project today's driving 

forces out into vastly divergent futures, events can rule out 

one or more scenarios surprisingly quickly, often in a matter 

of weeks, or sometimes even hours (e.g., the final report on 

a Mexican scenario project was sent to the client the day 

before the Colosio assassination).  Yet all four of my Y2K 

scenarios still stand, even at this late date. (T-minus-500-days

is August 19.) Where are the definitive leading indicators?

This lack of knowledge is itself a property of the Year 2000 




 + August 17-20, 1998 -- INTEGRATED BROADBAND NETWORKS '98 

   Hyatt Regency, O'Hare, Chicago 212-983-3500.

 + September 1-3, 1998: Mondo mongo Telecom Business 

   at the Javits Convention Center, NYC. I'll be talking 

   at 4PM on Sept 2.

 + September 14-17, 1998, Washington DC:  Jeff Pulver's 

   Voice on the Net.  THE event on the IP-PSTN synergy scene.

   I'm talking at VON on Monday then 

   boarding the evening flight to Reno for . . .

 + September 15-17, 1998, Lake Tahoe NV: George Gilder's 

   TELECOSM.  Where disruptive people discover disruptive 

   technologies.  A new addition to my "Former AT&T People 

   With Itchy Feet" panel is Bob Bailey, formerly of AT&T 

   Microelectronics, now the CEO of PMC-Sierra, which is 

   shaping up to be the Intel of communications chips.

 + October 14-15, 1998, Toronto ON:  IP Telephony and 

   Voice/Data Convergence.  A distinctly Canadian view.  In 

   many respects Canada is showing the way to the rest of us 

   (e.g., CANARIE, the Canadian national optical network 

   initiative). or 416-927-7936.

 + October 26-29, 1998, Cannes, France: ISP Forum! IIR, 

   Communications Week International and Total Telecom are 

   organizing this most awesome event.


[to subscribe to the SMART list, please send a brief, 

personal statement to (put "SMART" in the 

Subject field) saying who you are, what you do, and why 

you are interested in joining the SMART List.]

[to unsubscribe to the SMART List, send a brief 

unsubscribe message to]

[for past SMART Letters, see]


Date last modified: 10 Oct 98