Re: [Gnu-arch-users] documentation and licensing

[Ludovic Courtès]

Hi,

Thomas Lord <address@hidden> writes:

> 2) No, do not use TeXinfo or Docbook -- go for ASCII-art, Wiki-style

Perhaps Texinfo sucks (I'm not a big fan either), but it turns out to be
GNU's primary documentation format (`GNU' as in `GNU Arch'...), it
exists, and it will probably survive for a long time, whether you like
it or not.

Furthermore, for it is non-ambiguous, Texinfo markup can and will be
easily translatable in any other kind of markup, if need be[*].

I for one am interested in markup languages and document preparation.
I'm even rolling my own right now, but I wouldn't suggest its use for
the GNU Arch project.  About two years ago, you wrote the tutorial with
your home-made markup, part of Systas Scheme; Systas Scheme now no
longer exists and is not available anywhere as it seems, but GNU Arch
still exists.

Remember that currently GNU Arch has _no_ documentation since "arch
Meets Hello World" was removed in favor of the new, awiki-based,
placeholder.

> 3) Somebody make a plan.

Yeah, of course, that would be great.  But the point is (i) we have no
documentation whatsoever, and (ii) nobody is working on it.

For this very reason, I suggested to just revive the tutorial, making it
up-to-date.  That's not a very ambitious goal, but at least that's a
first step in the right direction.  We need to get some documentation
ready by the next release.

> b) A few people (not just me) integrating all the lessons we've learned
>    from Arch, git, Subversion, Darcs, monotone, etc. to make arch 2.0.

Out of curiosity: what new ideas brought by Subversion are you referring
to?

> Absent 2.0 I think it likely that Arch will die.

I agree with Andrew in that Arch 2.0 is very unlikely to exist any time
soon --- even just getting documentation for Arch 1.x is not granted, so
let's forget about implementing Arch 2.0 for the moment.

Furthermore, I persist to think that Arch 1 is a good tool, even
compared to the more fashionable thingies that have been brought into
existence recently.

Bazaar might have been a ``engineering disaster'' (I'm not able to judge
that), but at least it showed that there are still a number of ways that
the _implementation_ of Arch 1 could be improved, be it superficially
(e.g., `branch', partial undo, etc.) or more deeply (e.g., merge
algorithms, ancestry pre-computation --- that sort of things, admittedly
debatable).

Thanks,
Ludovic.

[*] See, for instance, `stexinfo':
    http://lists.gnu.org/archive/html/guile-user/2004-07/msg00047.html .