Subject: authentication systems (.NET, .GNU): Its the desktop, dummy.
From: "Stephen J. Turnbull" <turnbull@sk.tsukuba.ac.jp>
Date: Thu, 13 Sep 2001 12:17:57 +0900

>>>>> "Tom" == Tom Lord <lord@regexps.com> writes:

    Tom> It hasn't helped when FSB's undertake to compete with one
    Tom> another on software products -- a tactic that makes much
    Tom> sense for proprietary software, but little sense for open
    Tom> source software.

Do you really think the XEmacs split, the GCC split, the SSH split,
the BSD splits, the Linux distros, etc are some kind of Free Software
Olympics, competition for the sake of rivalry?

People start projects, businesses or otherwise, because (a) they want
independence and (b) they think they are good at it.  Often enough,
(a) is independence _from_ an incumbent, for principle or personality,
and there are many reasons why (b) might be in the same line as
another project, often including the fact that the project is a fork.

Once you have parallel projects, it makes sense to make yours the
best.

Face it: forks and rivalry are a fact of life.  Not only that,
probably more so to FSBers than to most people, the independence that
a separate project gives is one of the highest positive values in
life.

We need to make competition work _for_ us, not bemoan its existence.
The decade-old Japanese malaise is strong evidence that in fact it
does so by its very nature.  Japan, Inc was a whole national economy
organized on the "Cathedral" principle.  Worked fine while they were
chasing taillights (sound familiar?), but now that there are no
taillights to be seen and lots of headlights in the mirror, they're
hosed.  This was going to be the "Japanese century", but it looks more
to me like it's going to be another "American" one.  China, maybe, but
they're hobbled by "Asian values" (== cooperation by patriarchy) just
as Japan is.

Why does this happen?  Because at the bleeding edge, it's not clear
which way is forward.  So be nondeterministic and parallel: try them
all.  It's hard to do that in the context of a single organization.
Each advocates his own idea, and tries to concentrate resources on it.
Politically, usually somebody wins, starving the others into
submission---or a fork.

Competition is the best way we know to measure which is more
successful.  It has its problems (among them the circularity of the
implied definition of "success"), but all the alternatives eventually
boil down to picking a wise leader and hoping he gets it right, which
has a terrible track record by any definition of "success."

We need to figure out more ways for developers to cooperate while
their projects compete.  We have some already, but they're not very
amenable to business organization.

-- 
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 straight lines for?  "XEmacs rules."