Subject: Re: economic efficiency of free software
From: "Stephen J. Turnbull" <>
Date: Tue, 10 Feb 2004 18:24:02 +0900

>>>>> "g" == DV Henkel-Wallace <> writes:

    g> I think the Innovator's Dilemma shows how GE, IBM, and PwC
    g> could actually develop an office suite for less than Microsoft
    g> charges _in the long term._ They can't do it for the
    g> incremental cost.  But I also don't see how they can do it
    g> without running afoul of the antitrust laws.

If they decided it was strategically necessary, they could "Embrace
Insanity" by simply starting with a few departments (eg, research
labs) that throw out much of their Microsoft Office stuff, and buy
the closest available replacements from a few vendors with the
understanding that (a) the vendors will get dropped if they don't
respond to requirements for certain features, which may include
compatibility with other vendors' complementary applications and (b)
that it be open source.  "We don't wanna use Powerpoint any more, and
we'll pay what it takes to get feature-compatibility."  Then expand to
other apps and departments as quality and scope get big enough.  Just
as GNU did with the OS services:  "Who needs a kernel?  We've got
Solaris/HPUX/Irix/AIX..., we just need a better/compatible/portable/
free grep/awk/find/ls/less...."

Sure, this gives them zero _control_, even in the medium term; that's
the "insanity" part.  But they do get some insurance against vendor
lockin, and the quality will be there---whatever canards you may
believe about the creativity of FLOSS hackers, they can copy and
kaizen as well as the Japanese (who suffer the same prejudice).  The
requirements cabal can't explicitly control setting of standards
without antitrust issues, but they certainly can all go to the same
vendors "'cause they've got the product we need," and sponsor open
standards bodies.

And in any case, they'll get it "for free" in the long run; I can't
imagine that hackers won't eventually scratch that itch too.  So they
don't have to pay full incremental cost; just add a little oomph at
strategic junctures.

    g> Yes, but most of the value is, as it should be, captured by the

It's not obvious that it "should be".  Even the vast fortunes amassed
by the likes of Bill Gates and Larry Ellison are tiny when amortized
over 50% of the economy or so.  Of course, a large proportion of "end
users" are businesses, many of which are not sufficiently competitive
to drive out this source of profit, so you end up with profit going to
a different set of stockholders, which is surely not a social win.
And remember that the "lump of value" whose maximization that makes
the (well-functioning) competitive market go does _not_ account for
the fixed costs that loom so large in software.

While I know you (gumby) understand all that, my point is that the
phrasing you're using is not a lot of help in "thinking outside the
box."  And (as Jean so tirelessly points out) that is precisely what
is needed here.  Except that the box that FSB needs to get out of is
not the "global" conventional wisdom that Jean keeps trying to
pigeonhole me in; we're mostly out of that already.  We have to get
out of the "FS conventional wisdom" box.  (Sorry, Tim, I guess that
was your line!  :-)

What we need to find is ways for FS suppliers to capture sufficient
amounts of value to cover those fixed costs while still allowing the
product to be FS.  Let's worry about what the "end user" deserves
later; to a first approximation software being free, by itself, should
be able to mostly take care of the end user.

I'm sorry that I can't be any more helpful in providing the "out of
the box" ideas themselves at this point, but I do think your phrasing
is a bit pessimistic.

    g> end user.  Does enough remain to fund the FS suppliers?  I
    g> think that still remains to be seen.

Depends on whether you count firms like Aladdin/Artifex as "FS"
suppliers.  I think (and hope) that such serial licensing modes will
become more common over time as businessmen and VCs, who aren't
susceptible to Free Software Fanaticism, also start to internalize the
lessons that folks like Varian and Shapiro taught about the
uselessness of trying to squeeze every penny out of intellectual
"property" rights.

Institute of Policy and Planning Sciences
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.