Subject: Re: open source definition
From: "Jonathan S. Shapiro" <>
Date: Wed, 22 Apr 1998 14:08:21 -0400

> Keith Bostic writes:
>  > When this argument was floating a month or so ago, I challenged anyone to
>  > name a company that was writing software and then giving it away -- a
>  > company that sold software for a living, that is, not a company that sells
>  > support and packaging.  Nobody answered that challenge.  To my knowledge,
>  > there is not one single software company that plays the free software game.
> Nonexistance is not disproof, of course.  Just because someone hasn't
> had the courage or ability to attempt something doesn't mean that it
> can't be done.  If there's a road on the front of the mountain, why
> climb up the back of it, particularly when the goal is just to get to
> the top?


Some of us on this list have started more than one commercial
organization, and even more than one *successful* commercial
organization.  As I'm sure you know, the challenge is always finding
the money.

In some lucky circumstances, you can get funded by your customers.
This generally works when the required initial investment is quite
small (the common case), or when the customer perceives advantage in
the result being non-proprietary (damn rare).  Packet drivers are an
example of something with a small initial investment.

In really rare cases, you can do something as academic research.

Developing a new word processor or operating system or command and
control platform, however, doesn't generally fall into either
category.  The days when DARPA will sizably fund new OS work are

To get money from other sources for such large projects you need to be
able to show that there will be a revenue stream from which the
investment will be recouped.  To do that, it helps a lot to show that
the users of the product will pay to use it.

There are certainly a lot of products where such investments are not
necessary.  For the most part, I think such projects can successfully
be pursued under "open source".  It's very hard to sustain a company
that way.  Cygnus, for example, has been forced to shift into new
markets twice after tapping out an existing market for "paid
development services."

For large-scale new work, though, I do not see how to get the work
funded.  Where a "seed version" of a perl or a web server or even a
netnews can be built and then leveraged in Eric's bazaar approach in
the course of a year's spare time by 2 or 3 people, this is not true
of larger systems.

Perhaps people are content with saying "this model doesn't work for
something large and new, and that's okay with me."  I'm not.

You want people like me to give our software away.  Hell, at some
level *I* want to give my software away.  I cannot rationally do that
until I exhaust the profit-making options; there is just too much at