Subject: Re: Competition - What does it look like?
From: Rene Kienzle <rkienzle@bigpond.net.au>
Date: Thu, 28 Mar 2002 09:42:42 +1000

At 27/03/2002 02:56 PM -0800, Ian Lance Taylor wrote:
>Rene Kienzle <rkienzle@bigpond.net.au> writes:
>
> > What I have been running up against is the issue of
> > competition.
>
>Free software projects compete for mindshare.  They do this both in
>the sense of finding people to work on the project, and in the sense
>of finding people to use the project.

Yes, I mentioned this in the sense of inputs in my previous post. This 
would probably be called environmental munificence (abundance of resources 
in the environment) in other research. In this case the concept is perhaps 
a bit broader - usually the term is restricted to availability of financial 
capital.

>Anyhow, my point is, when free software projects compete directly
>against one another on functionality, there is generally some
>underlying historical reason for it.  It is not the natural state of
>affairs, and I suspect it is going to be hard to study using a
>statistical analysis of the form you propose.

I don't see any issues with statistical analyses. You can statistically 
analyse almost anything. The issue is about how meaningful the 
results.  For the rare instance where there is competition i still need to 
cater for it (measure it). Even if that competition is merely perceived. In 
this case, competition is again more broadly defined to include 
product(output) competition and competition for resources (human, social 
etc) that add to growth (performance)

>It would be more interesting to analyze the mindshare issues: why do
>some projects get it while other projects don't?

I hoping to answer that with the criterion being project growth. The 
mindshare issue is addressed by looking at a number of issues - the project 
leader/s and their approach to managing the project and their behaviours, 
the culture of the project (open, supportive, encourage innovation), the 
characteristics of the project (e.g. asset specificity, inputs/outputs to 
other projects), and other issues of reputation and value of various 
tangible/intangible resources, as well as controls for use of various 
sourceforge systems (bugtracking, forums etc).

At this stage I will only be surveying project leaders/founders and not 
actual participants. So that cuts out a lot of the harder measures of 
"human capital" that would be possible.

The idea is to build a picture that ties together project resources with 
performance.

Rene