Re: I resign as Guile maintainer

[Dirk Herrmann <dirk at ida dot ing dot tu-bs dot de>]

On 1 Dec 1999, Greg J. Badros wrote:

> I'll also briefly take this opportunity to state what I claim is the
> most significant problem with Guile:
> 
> Lack of reasonable documentation.
> 
> If I had to choose among extension languages right now, I'd be stunned
> if Guile were my choice.  Only after the documentation improves will
> people come to Guile to build a larger support infrastructure.  It's
> just too damn hard to get started with Guile the way things are now.

Another important point with guile -- in my opinion -- is, that everything
must be perfect before adopting it.  Most of the time this is a good
attitude, but there are some points where something that at least works
somehow would be better than nothing.  Documentation IMHO is one of these
points, since even a maybe-not-perfect documentation system encourages to
write documentation.  Compared to the effort of writing up documentation,
the changes that may at some time be necessary if the documentation system
will be improved/changed/completely restructured are of less importance.
If we sum up the time wasted due to missing documentation _and_ the time
lost for guile due to people that decided for a different language because
of the lack of documentation, in the same time we could probably have done
a couple of restructurings of a documentation system that was 
maybe-not-perfect in the very beginning.

Greg, you have quite a lot of experience with documentation issues.  The
SCWM documentation system is very good.  Wouldn't it be comparably easy to
adopt the SCWM system to guile?  You will know whether there are a few
modifications that might make sense since there are probably a couple of
lessons learned while using it.

Dirk Herrmann