> > > I agree with RTFM! as much the next guy, but it seems
> > > like users might be better off if we renamed
> > > the buildconf file to autogen.sh. Lots of
> > > other projects use a script named autogen.sh.
> > > I am not sure it is a "standard", but why
> > > be different for no reason?
[..]
> I guess I don't follow your logic there. So, since someone
> else does it incorrectly, the whole concept must be flawed
> from the get go?
You accused us of "[being] different for no reason".
All I'm saying is that there is no standard, thus we simply
cannot be different, because there is no objective standard
to measure us with.
> Why exactly is it "better" for apr to use a custom name for the
> bootstrap scrip?
Why exactly is it "better" for apr to use another custom name
for the bootstrap strip?
> I suggest you go download the source code to gcc and take
> a look at the toplevel configure.in file. Does it change
> your mind about using a configure.in in apr?
What relevance does that have here? While you bring up GCC,
it is really beyond me why they still have autoconf-generated
in CVS.
> Lets face it, people do not read the README.
That is their problem. I also don't complain to the GCC
or newlib maintainers that I cannot build their stuff from
CVS without reading the documentation.
Let's face it, developers who benefit Apache or APR are able
to read READMEs. The rest can download a tarball.
- Sascha
