Subject: Re: Mass. Bill to Block OpenDocument Format Standard
From: "Stephen J. Turnbull" <stephen@xemacs.org>
Date: Mon, 07 Nov 2005 12:58:46 +0900

>>>>> "Marshall" == Marshall W Van Alstyne <marshall@MIT.EDU> writes:

    Marshall> At 01:12 AM 11/5/2005, Stephen wrote:

    >> Analog HDTV was all about structural impediments to trade, not
    >> about technology per se.

    Marshall> But it was still about setting a standard, right?  And
    Marshall> to the extent the Japanese choice wasn't about the best
    Marshall> technology alone, perhaps it strengthens the point that
    Marshall> governments aren't always the best places to be choosing
    Marshall> those standards.

First, if you'd left it up to an industry consortium, the answer in
Japan would have been the same, it just would have taken longer and
wasted even more resources.

But the important point is that nothing of interest to Real People[tm]
is about technology alone.  Besides politics, there are things like
costs, social and environmental externalities, etc, and those things
cannot be reduced to a unidimensional criterion, let alone one with
the connotations that "better technology" has to people like us.
Profit-oriented vendor or altruistic government alike, the decision
will not be made solely on the basis of technical merit, and should
not be.

    Marshall> Content standards like HDTV are arguably examples of
    Marshall> "2-sided" network externalities that are only recently
    Marshall> being addressed in theoretical economics.

The Battle of the Sexes was solved in the Garden of Eden when Adam
conformed to the pure strategy equilibrium of doing it Eve's way.
There's really nothing new or hard here.  The point is that "network
externalities" are not a game-theoretic concept; they are a
decision-theoretic concept.  That is, they provide formulae for
utility functions.  If we dramatically oversimplify the dynamic form
to a static two-by-two matrix game between vendors, we get the Battle
of the Sexes, right?  And what we see in the real world looks like the
mixed strategy equilibrium outcome, right?  Vendors arguing over who
gets the "big win" vs. the "little win", while in fact both receive
"no win" far too frequently.[1]

It's true that sometimes these new, specific utility functions or more
dynamic formulations yield surprising results, but the consistent
thread over the last 45 years of game theory has been that new
concepts eventually produce the same results as the old ones do, given
the same kinds of restrictions.  We occasionally get useful insights
about what happens outside the restrictions, but mostly we just get a
much more scientific understanding of what those restrictions are.
This opens the door to better-engineered solutions, but does not
change the fundamental problem.

So let me respell that as "Any standard is better than no standard.
Decide quickly."  Classic example: AFAIK today everybody agrees that
SONY Beta was "better" than VHS, and (equally AFAIK) no Real Person
cares, and they haven't since 1980 or so.  They've been too busy
renting videos, eating popcorn, and making out on the sofa.[2]

It's very hard make an _economic_ mistake by choosing a standard, any
standard, _quickly_.  Just as long it's not _so_ bad that the majority
of users positively refuse to conform (the OSI (sp?)  network
standards, or the ISO 2022 Coded Character Set Extension Techniques
standard).

Cf. Stephen Jay Gould and the theories of "punctuated equilibrium" in
evolution.  You make the standard, you find out what your mistakes
were, in combination a couple years later you make a new standard, in
the meantime a truckload of popcorn gets eaten.  Or, if you prefer,
Schumpeterian "creative destruction" is the way to go.

    Marshall> These face a distinct chicken-and-egg adoption problem:
    Marshall> content providers don't want to provide HDTV content
    Marshall> until there is an installed base of HDTV boxes but no
    Marshall> one wants to buy the boxes until there is HDTV content.

IMO, it's not chicken-and-egg.  No one will buy boxes until there is
content, true.  But content providers don't need an installed base;
they need an "installed standard".  What the installed base does is it
substantially raises the bar to changing the installed standard.  But
suppose we can find another way to "encourage" conformance to a
specific standard?  It will have the same effect.

Don't forget that humans are endowed with modest intelligence and
imaginations.  They can see things that never were, and ask "why not?"
If the answer is No Reason, some fool^Wentrepreneur will try!

    Marshall> Stepping back, it seems that these 2-sided networks
    Marshall> (with content creation & content consumption) require
    Marshall> some kind of coordination mechanism to achieve a viable
    Marshall> standard.

I disagree.  The content provider-consumer coordination problem is
trivial.  The problem is rivalry among the technology vendors, leading
to deliberate obstruction[3] of coordination among the vendors.  That
rivalry is not going to be solved by coordination, only by compulsion.

    Marshall> The question is then who is in the best position to
    Marshall> coordinate that standard.

To me, the question is "who do you trust with enough compulsory
power?"  IMO, who will make the better choice of standard is a pretty
secondary matter compared to that.  YMMV, of course.

Footnotes: 
[1]  You shouldn't be confused by the war of attrition aspects of the
coordination failure case; it's basically a Battle of the Sexes
problem, because the symmetric equilibrium is worse for all than any
of the asymmetric coordination equilibria.

[2]  Cf. http://www.jwz.org/doc/groupware.html.

[3]  And we mustn't forget the complications that arise from FTC and
WTO rules.  But the war of attrition aspects of vendor choice of
standards are the main cost.

-- 
School of Systems and Information Engineering http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba                    Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
               Ask not how you can "do" free software business;
              ask what your business can "do for" free software.