Subject: Re: "I've got more programmers than you"
From: David Fetter <david@fetter.org>
Date: Wed, 3 Oct 2001 10:27:14 -0700

On Wed, Oct 03, 2001 at 03:58:51PM +0200, Willem Konynenberg wrote:
> "Stephen J. Turnbull" wrote:
> > >>>>> "dblankley" == dblankley  <dblankley@iwon.com> writes:

> >     >> Free industrial-strength RDBMS?  Just forget it.  Nobody
> >     >> can afford do the kind of exhaustive QC such a beast needs
> >     >> for free.  Yes, I know about PostgreSQL and like it a lot,
> >     >> but it will never catch up with Oracle, Sybase, Informix,
> >     >> etc. 'cause the project doesn't (and a crisp $20 says it
> >     >> won't so long as it's Free) have those kind of resources
> >     >> available.

> >     dblankley> Couldn't the same argument can be made about OS,
> >     dblankley> desktop or web-server???

> > No, probably, no.

> Part of the error in the above argument is in the phrase "for free".

What error?

> I seem to remember that this sort of argument has been made in the
> past about just about every major piece of free software that we
> today take for granted.

> "A programmer's editor, sure, but a full set of UNIX utilities?"
> "some UNIX utilities, ok, but a C compiler?"
> "A compiler, ok, but a complete C library?"
> etc, etc.
> And each time, the next project came and showed the argument invalid.

That does not mean that it's wrong in this case.  While the Free
UNIX(alike)s and C's are impressive, they're not that impressive.  Why
would I say such a horrible thing?  It's like this:  UNIX and C were
pretty much fully designed, developed, tested, etc. by people with
extremely rare combinations of talents (not trivial) and on the tab of
some people with extremely deep pockets.  This cost a *huge* initial
outlay, and eventually, the vast majority of that stuff got to be
unencumbered enough that just about anybody could reimplement UNIX and
C.

No similar thing has happened with RDBMS's, and the kind of testing
and bullet-proofing such things take is inherently *much* more
expensive than that for an OS.  One could think of UNIX filesystems as
a kind of extremely crude and primitive database--hierarchical
databases are mostly found in database history books--and even those
are still in many ways in their infancy.

> Given the pattern sofar, I wouldn't exclude any class of software
> from being viable as a free software project.  I would only
> differentiate on the timing.  Some things may happen in 1 or 2
> years, others might take another decade or so.

This might have sounded a lot less naf 3 years ago.  Just now, it
sounds almost desparate.

Cheers,
D
-- 
David Fetter david@fetter.org http://fetter.org/~shackle/
phone +1 415 567 2690  fax: +1 415 567 2340