Subject: Re: Do We Need a New Evangelist
From: Brian Behlendorf <brian@hyperreal.org>
Date: Wed, 31 Mar 1999 09:10:47 -0800 (PST)

On 31 Mar 1999, Russell Nelson wrote:
> D. V. Henkel-Wallace writes:
>  > At 01:09 31-03-99 +0000, Russell Nelson wrote:
>  > >I'm sure it wasn't Apple's idea; I'm sure it was persistent
>  > >negotiation on the part of Apple's users.  Like all heads of state or
>  > >business, Jobs is taking credit for it.  But it's the people who drove
>  > >them to it.  Elites never give up power willingly.  They have to be
>  > >forced, through the efforts of 10, 10K, or 10M people to do it.
>  > 
>  > Sounds great, but I suspect exactly the opposite.
>  > 
>  > I don't want to end up with "open source winter" which is why I care.
> 
> Hehe.  The continuing part of the analogy is that if the decision was
> Apple's, and no Apple users care, then it will be a failure.  You see
> the same effect when the elites force a free market on the populace.

Right.  I think in so far as it's been OSI's role to evangelize Open
Source, it should also be its role to advise companies when an Open Source
approach may not be the best one to take.  I, too, fear an "open source
winter", by which I mean a failure of a series of high-profile experiments
that causes companies to either denounce publicly the approach, or
privately have enough distaste for it that they don't attempt new Open
Source projects.

I think there's one major aspect of OSI's approach that could use
reconsideration.  I don't think that it's the software companies who
should be lobbied to open-source their code.  I think it's their large
*customers* who should be approached, explained the merits of having
control over their software, etc etc.  Which company do you think, as of
today, has made more money, as a sum of (extra revenues and expenses
saved), from Open Source?  I would contend it's companies like Yahoo and
Hotmail and Geocities, who were able to build extremely complex sites on
commodity hardware and software, that without the tools produced by the
open source community, would have probably had to spend a great deal more 
in hardware, software, and engineering time.  Next on the list would be
ISP's, both small and large.  After that, probably academic institutions.

The real problem I'd like to head off is "too much code, too few
developers".  I think we can increase the developer pool by proving to
these non-software-companies that yes, there's a good reason why they
should hire someone to develop public software.  

	Brian