Subject: Re: New ESR paper: The Magic Cauldron
From: "Stephen J. Turnbull" <>
Date: Sat, 26 Jun 1999 17:38:34 +0900 (JST)

>>>>> "esr" == Eric S Raymond <> writes:

    esr> Nevertheless, I hear objections that seem to issue from a
    esr> commons model very often.

Hmm.  Well, I don't.  But then, neoclassical economists who listen
long enough to get to that point will get the species of market
failure right ;-)

    esr> I have now added language as follows:

Good.  Answers my objection (except for the difference of opinion over 
"most people have an intuitive model", and I don't intend to do the
fieldwork, so I can't ask you to do it, either).

    esr> Good point.  My argument can be saved, but it requires a bit
    esr> more precision.  The sentence now reads:

    esr>     In analyzing this kind of cooperation, it is important
    esr> to note that while there is a free-rider problem (work may
    esr> be underprovided in the absence of money or money-equivalent
    esr> compensation) it is not one that scales with the number of
    esr> end-users.

I don't understand the "not scale" point.  I've looked at this in a
simple mathematical model, where "hackerishness" is the innate
propensity to hack, and is distributed randomly in proportion to
population.  In that case, the problem is homogeneous of degree one in
the number of users.  In order to get less than degree one, you need
to assume that in some sense the hackers are "first in" and as the
number of users rises the proportion of hackers falls.  This may be
true when you compare Linux against MS Windows, but it's not at all
clear to me that it's true in general, especially as you consider the
education effect of having source available.

If you take the "all bugs are shallow with enough eyeballs" argument
seriously, it leads to the conclusion that bugfixing and feature
provision should be increasing returns to scale in user population.
Now, it is true that with "easy" bugs and features, congestion and
coordination problems will quickly lead to decreasing returns (pace,
Steve McConnell).  But with "hard" bugs and features, the number of
discoveries should be sufficiently sparse that the normal OSS
procedure of "discuss on Usenet, code, and distribute" will be
sufficient coordination.  Presumably these are the highest
value-added, too.

So I have very little confidence in your statement.

    esr> In fact, my impression is that this kind of underprovision
    esr> is not a significant problem.  To my knowledge Red Hat's
    esr> product quality has never been constrained by its ability to
    esr> pull in revenues to buy developer time -- because most of
    esr> the work is being done by other people, and because Red Hat
    esr> has been able to attract capital through alliances.  I'll
    esr> have to think about how to address this in the paper.

I wouldn't bet my professional rep on the "Mentat computation," (q.v.) 
but I will say that the problem of underprovision is definitely _not_
an issue of improving the Red Hat Linux Distribution, or any of the
other individual free software projects out there.  It's an issue of
all the software that we can't imagine existing,[1] let alone being free,
under current institutional arrangements.  Then add to it the benefits
to freeing all the software that isn't free yet.  And on top of that,
improving all of it simultaneously.

I believe a large part of that to be possible if we can harness just a
small fraction of the effort that those currently free riding could
provide,[2] which is another reason why I disagree with you about the
free rider problem scaling.

For a completely different approach to the measurement that gives
comparably astronomical value-added, I'll quote `burley':

>>>>> "craig" == craig  <> writes:

    craig> My "design goals" for the OSS infrastructure that'll be
    craig> needed to carry us through the next 50-100 years include
    craig> accommodating a billion or so people using *and* hacking on
    craig> the same set of OSS at the same time, despite speaking
    craig> different languages, including many who are, for example,
    craig> blind.

Hear, hear!  A kindred spirit, of bold vision!

    esr> BTW, my wife reminds me there's a Stephen Turnbull who
    esr> writes rather good stuff on Japanese military history.  Are
    esr> you he by any chance?

I do not have that honor, but I do have the honor of virtual

    esr>         -- Thomas Jefferson, writing to his teenaged nephew.

I'm disappointed that old TJ so misunderstood the nature of boldness,
enterprise, and independence of mind as to think they are encouraged
by an overwhelming advantage in firepower.  All of the beneficial
effects I can think of would today be equally well served by a camera.

[1]  Yeah, yeah, I know, we're still "chasing taillights."  But that's 
not going to be true forever.  It's probably only 99% true today, and
the most exciting OSS is in the 1%.  And the truly novel share of OSS
will grow, as OSS grows as a share of the truly novel.

[2]  Of course some of it would be taken from other projects or other
fields entirely; but if the individual thinks it worth doing, and you
add to that all the external benefits from open source, surely the
aggregate net benefit to reallocating the effort is huge.

University of Tsukuba                Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Institute of Policy and Planning Sciences       Tel/fax: +81 (298) 53-5091
What are those two straight lines for?  "Free software rules."