From: Tim May <tcmay@got.net>
Date: Sun, 11 Jan 1998 03:54:52 +0800
To: John Young <cypherpunks@cyberpass.net
Subject: Re: Theory Behind USA v. Microsoft
In-Reply-To: <1.5.4.32.19980110191251.0101e398@pop.pipeline.com>
Message-ID: <v03102801b0dd7c58e771@[207.167.93.63]>
MIME-Version: 1.0
Content-Type: text/plain



At 11:12 AM -0800 1/10/98, John Young wrote:
>John Cassidy writes in the January 12 New Yorker mag
>of the controversial economic theory which undergirds DoJ's
>antitrust action against Microsoft.
>
>He cites a seminal 1984 paper by Brian Arthur, "Competing
>Technologies and Lock-in by Historical Small Events: The
>Dynamics of Choice Under Increasing Returns."

I've followed Brian Arthur's work for a decade or so, and find much in it
that seems quite accurate.

One of his main observations is that size _does_ matter, that larger
economic agents often tend to get larger. There are various plausible
reasons for this, but it does seem to match reality.

(I'm avoiding jumping to conclusions that Brian Arthur is anti-capitalist
or any such thing. Just discussing reality as it is.)

We see this in the "Intel-Cisco-Sun-Microsoft-Oracle" universe, where each
of these players has about an 80% or better market share in its respective
niche.

Now Brian Arthur doesn't claim that such dominant market shares will last
indefinitely--the dominant companies in 1900 are mostly no longer even in
existence in any recognizable form, and even the dominant companies in 1950
have mostly been completely replaced by "upstarts."

But size does matter, bringing economies of scale, the ability to set and
enforce standards, and the ability to withstand competitive onslaughts for
longer times (than smaller, less financially solid, companies).

>After years of disparagement the theory seems to have
>caught on, at least at Justice and with others who oppose the
>theory of free market determination of winners and losers.
>Arthur argues that market dominance by inferior products
>is possible, and cites MS-DOS as an example.

Sure, there are many, many "non-optimal" products. Much of society is
non-optimal, even in the infrastructure. Roads don't go where they
"should," the wrong kind of electrical sockets were adopted, and so on, for
examples I don't need to spend time listing.

A way of viewing this is of _inertia_ or _sticking friction_. Once certain
standars have been set, it is just not possible to roll back history and
proceed down another path.

For example, it might well be that the world would have been better off
using the Motorola 68000 family in places where the Intel x86 dominated
(for possibly accidental, local reasons, as anyone who has read the history
of IBM's adoption of the x86 knows). And ditto for adoption of a better OS
than MS-DOS was (same accidental decision).

But we are not in that world, and the installed base of PCs and x86 systems
and MS-DOS or Windows systems is so large that it is simply impossible to
"jump tracks."

Now eventually things will change. Some new paradigm will come along.

There is no guarantee that in 2025 the dominant players today will still be
dominant.

Let's not forget that two years ago many were saying Microsoft would be
wiped out by the advent of "Web browsers as operating systems and office
suites," with Netscape Navigator being the Swiss army knife of programs.
Recall that analysts were sagely predicting that Bill Gates had "missed
out" on the Internet.

Now we have the spectacle of Netscape demanding that the government give it
back its dominant Web position! (Maybe then the University of Illinois can
get the DOJ to sue Netscape to take away Netscape's dominant position!)


>
>Arthur is now a scientist at the Santa Fe Institute. He says
>his theory "stands a great deal of economics on its head."
>One critic said to Arthur, "If you are right, capitalism can't
>work."

Which is nonsense. All Brian Arthur has done is to analyze some of the
"physics of markets" (my name).

Schumpeter said much the same thing when he talked about the "creative
destructionism" of capitalism.


--Tim May

The Feds have shown their hand: they want a ban on domestic cryptography
---------:---------:---------:---------:---------:---------:---------:----
Timothy C. May              | Crypto Anarchy: encryption, digital money,
ComSec 3DES:   408-728-0152 | anonymous networks, digital pseudonyms, zero
W.A.S.T.E.: Corralitos, CA  | knowledge, reputations, information markets,
Higher Power: 2^2,976,221   | black markets, collapse of governments.
"National borders aren't even speed bumps on the information superhighway."