Subject: Lack of imagination?
From: "Stephen J. Turnbull" <>
Date: Tue, 3 Jul 2001 13:11:35 +0900

>>>>> "Guido" == Guido van Rossum <> writes:

    Guido> Many open source projects look to me like they are trying
    Guido> to come up with an open implementation of something that's
    Guido> already available (or being developed) commercially.

Tactically, note that the commercial interests are aiming at
monopolies.  _Big_ monopolies.  Answering those threats is important.
Demonstrating that there is no technical need to integrate the browser
with the desktop, or the desktop with the authentication service, is

Strategically, that policy worked for Japan, 1940-1990.  Sure, "bubble
economy" RIP.  But very few of the open source programmers I know have
to live on earnings from open source.  If they're forced to take a day
job, they end up hacking anyway.  So open source is no bubble.  The
only thing I can see that can even slow it is dramatic expansion of
software and business method patents.  _Then_ innovation will be
essential.  :-(

And for many part-time hackers, working on a defined project with
visible short-term milestones is preferable to a blue-sky "innovative"
project that may or may not "pan out" in a year of Saturdays.  So
there will be a bias toward reimplementing known products, at least in
the volunteer OSS world.

    Guido> Where are the open source projects that *innovate*?

Funded by the government, possibly?

You're making the classic "architects are creative, implementers not"
error, I think, in ignoring implementation.  Many open source projects
innovate in their implementations.  Uli Drepper will tell you that
glibc is full of inventive approaches to implementation of the
standard functionality (aside from many important extensions); he
certainly points this out in every public speech.  Did you solve no
new problems in developing Python (aside from the innovations in the
_language_)?  Others will surely tell the same story.

And I think it's not just unwitting reinvention of the wheel.  It's
necessarily often _new_, since open source projects generally are more
decentralized and must be more modularized than proprietary projects.

But "implementation innovation" is not as visible as "product

And again look at the Japanese.  Until 1980, hardly any high-tech
thing they produced was _invented_ there.  But simply by producing
things cheaply enough to reach a mass market they _innovated_ them.
Transistor radio, 10kg 10-speed bicycle, video tape recorders.
Ie, "implementation innovation" generated "product innovation."

A relevant example.  I don't know enough facts to say it's probable,
but it's arguable that open source MP3 encoders created the MP3 player
industry.  (Can you _buy_ pre-recorded MP3s?  Can't in Japan.  If not,
where do they come from?  Napster, right?)  Wasn't _new_, but it was
an _innovation_.

Greasing up the wheel so that it goes past mere "rolling" and on to
"whirling", or even just making wheels available to all rather than
merely to the well-heeled, is something to be proud of.

Open source could do very well at creating enormous amounts of value
for people simply by forcing the commercial proprietary world to bust
their asses proving that it's with dealing with their silly rules.

Not to mention all the other stuff we do, as described in other recent
threads.  All that and product innovation, too?  Tall order!

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."