Subject: Re: ways of funding
From: "Stephen J. Turnbull" <>
Date: Wed, 10 Nov 1999 09:52:49 +0900 (JST)

>>>>> "shapj" == shapj  <> writes:

    >>> When was the last time in the course of _building an
    >>> application_ that you looked at libc code?  Or built a
    >>> substitute for an unusably buggy libc function?

    shapj> The situations are not entirely parallel.

I understand the difference between the processes that make libc a
"you can bet your job on it" reliable collection of functions, and
CPAN an unknown quantity spanning the gamut from utterly dysfunctional 
to absolutely bugfree (from the spec on up), possibly an interval open 
at one or both ends ;-)

But how about the processes _using_ these collections?

Do you agree/disagree with my contention that free software needs a
silver bullet (order of magnitude improvement) to overcome the revenue
deficiency over the whole market (my target is 1/3 market share in all
software or better)?

Do you agree/disagree that extremely high reliability in
1st-generation modules is necessary for them to not become obsolete as
components for the 3rd-generation modules?

    shapj> Finally, the observation that people spend money based on
    shapj> their *perceptions* of value, not based on objective value.
    shapj> While Stephen knows this, it's something that economists
    shapj> frequently overlook.

This is most definitely part of the reason why I think free software
advocates overestimate the amount of reuse that can be achieved with
current organization of the repositories.

Now that you point it out, it is obvious to me that _even though I
know my own bias_ my unjustifiedly (== small and biased sample) low
estimation of CPAN quality does make me less likely to use Perl than
objectively is warranted.

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 two straight lines for?  "Free software rules."