[OSGeo-Discuss] scale of FOSS projects

[jo at frot.org]

On Thu, May 08, 2008 at 05:14:40PM -0500, P Kishor wrote:
> On 5/8/08, Schuyler Erle <schuyler at nocat.net> wrote:
> >  is that the number-one sine-qua-non of *any* potentially successful
> >  software project is *shipping working code*.

> >  Until a developer does that, the discussion of whether or not his/her
> >  project needs or deserves institutional/organizational support

That is not what this discussion is about, though. (And the point
seems self-evident, given this is a discussion about open source
software projects, defined by having working code "in the wild")

> Steve Coast (OSM) echoed the same sentiment very elegantly -- "Real
> artists ship. For everyone else, there is wanking."
> After a short hesitation, I have really come to appreciate it. Yup,
> unless there is working code, everything else -- sponsorships,
> organization, standards, committees, mailing lists -- is pointless.

I really enjoyed the recent discussion here about non-developers
contributions to open source projects and communities. Writing
documentation and tutorials and maintaining translations, in
particular. That code-jockey primacy attitude is potentially alienating 
to people wanting to contribute this kind of hard work. 

For many it is easy to write software. There is a lot of code out there,
a lot of abandon-ware, projects that are "free" by a legal definition
but with none of the supporting infrastructure that helps them to get
used and to acquire a client base. 

At least Autodesk, for example, saw this and made bona fide effort to
"build community", rather than dropping millions of lines of
undocumented, hard-to-configure code onto the net, hoping an imaginary
"open source community" would sprinkle pixie dust onto it, as Sun did
at first - as if the time and goodwill of potential contributors were
inexhaustible.

There is this cultural pressure on "standards" to be marketing tools.
Because of the government and military context for GIS, this pressure
is particularly intense for us. It starts to loop back on itself
somewhat like this, http://frot.org/on_standards/statements.html 

This does have a countereffect on innovation in software and it also
probably does prevent "bona fide" standards developing in a natural way.
As well as creating this terrific and largely justified backlash
against some of the in-a-vacuum work done by OGC, ISO. (GeoDRM anyone)

However the process of working things out by rough consensus and running code 
takes longer, business process says, "first to market -> "natural monopoly|
de facto standard". 

It is unfortunate, because proper interoperability can be such a force for
good - cf MetaCRS, and the future time and hassle that is going to be saved
for many people, once the inevitable initial round of talking is done.

I know, this argument has gone round and round in the past, and many
are impatient with philosophising. I hope that philosophising can
sometimes provide energysaving insight, or i wouldnt engage in it. But
repeating "without code, you are nothing" grates on the nerves after a while.


jo
--