Subject: Re: Economics of software distribution
From: ghost@ALADDIN.COM (L. Peter Deutsch)
Date: Thu, 18 Mar 93 00:11:48 PDT

>    I tend to agree with Peter.  R&D has to be funded somehow and I'm not
>    convinced that there is a way for a company to recoup money spent on
>    R&D through selling "mass-market" free software.  It's ridiculous to
>    think that you can subsidize R&D by selling manuals or providing
>    support because if the market is lucrative enough you will have others
>    providing the same services without the liability of R&D expenses.
>    And it doesn't help matters that currently R&D is a major expense.
>    Russ's business works because R&D is being funded by companies that
>    are trying to sell hardware.  The mass-market software business
>    doesn't have that luxury.
> I think this posting shows very clearly the result of not completely
> adopting the paradigm of free software.  R&D, the way proprietary
> software companies do it today, does not work with the free software
> model; I am not surprised.  R&D can--and is--being done, but not in
> ways that many readers on this list are willing to recognize.  For
> example, I read today that some 30,000 programmers are looking for
> work on the east coast, after having been laid off by their DoD and
> DoE employers (either the gov't themselves, or contractors for the
> gov't).  30,000 programmers boggles the mind: that's probably 100
> times the number of people working on GNU in any capacity whatsoever.
> If these guys were properly cooperating and coordinating their
> efforts, I'm sure they'd all still be funded, but instead they are all
> duplicating (badly) the work of everybody else, and together they
> collectively cut themselves out of the market.

So what conclusion should we draw from this?  Free software is even less
cooperative and coordinated than commercially developed software.  Almost
all the free software I know about was done by individuals or tiny groups
that didn't need any coordination to speak of.  And if those programmers
didn't duplicate each other's work, but embraced the free software
approach, they'd still be without incomes.

>    On the otherhand, I hate the idea of hording software forever.  I like
>    Peter's model and there are other models which are nice compromises
>    such as copyrighting the software for a couple of years and then
>    releasing it into the free domain.
> The question is: do you like the free software model enough to let go
> of the proprietary paradigm?  As long as you prefer to discount the
> fact that hundreds (or thousands) of people at universities and
> research institutions are doing (or could be doing) "R&D" that
> directly feeds commercial free software projects, you will have to
> live with the reality that free R&D is not possible.
> Michael

"Letting go of the proprietary paradigm" means paying people like Cygnus a
lot of money to fix (or tell me how to work around) poorly user-adapted
software written by people who have little incentive to compete on quality
and no concept of how to write user-appropriate systems.  As a user, I say
thanks, but no thanks.

I really feel there is an air of unreality about this whole discussion --
most of the pro-free arguments seem to be coming from people who think the
Unix world is the zenith of software quality, and most of the anti-free
(or at least pro-diversity) arguments seem to be coming from people who
actually have some experience with the software market that matters (PCs
and Macs).  The economics and user expectations in the two markets are so
different that sometimes I wonder whether there is any point in trying to
discuss them in the same thread.

L. Peter Deutsch :: Aladdin Enterprises :: P.O. box 60264, Palo Alto, CA 94306, ...decwrl!aladdin!ghost ; voice 415-322-0103 ; fax 322-1734
	    "Implementation is the sincerest form of flattery."