Don's right on all counts. I've been itching to write this kind of
stuff myself for some time now, but I think we're not quite at the
point where we can do this effectively. *First* we need a
completed, stable first release. *Then* we need to evaluate the
UI to fix the major problems. (Some of the UI is pretty bad, but it
doesn't make sense to tackle this until the porting is done and
the basic architectural fixes are finished.) Once we have done
these two things, *then* we should kick the non-technical user
documentation into high gear.
UI testing and fixing needs to come first, since you won't need to
document nearly as much if you have a good UI to begin with.
We don't yet have a real process for doing this; I believe Talli is
leading the charge on this.