Re: [AUCTeX-devel] Release schedule.

[David Kastrup]

Ralf Angeli <address@hidden> writes:

> * David Kastrup (2005-05-12) writes:
>
>> a) document Windows startup using tex-mik.el or fptex.el,
>
> Are we going back to the scheme where you had to do both `(require
> 'tex-site)' and `(require 'tex-mik)' or do you have a way to place
> them into `load-path'.

I'd say that both are required.  I don't want to clutter site-lisp
with all sorts of compatibility files, and it is tex-site.el that
augments the load-path.

>> d) write up RELEASE and changes.texi.
>
> changes.texi is pretty much up-to-date except for mentioning the
> revamped installation procedures and the integration of preview-latex.
>
>> f) better document requirements, supported platforms, and software to
>>    download, in particular on the web page.
>>
>> I am sick of hearing the incessant "But what use is that to Windows
>> users?" whining.
>
> Have you observed something like this lately?

Yes.  It is the standard reaction.  Those that don't turn off in
disinterest immediately usually state something like "cute, but come
back when you have a solution for Windows users".  Try a bit of
proselytizing outside of comp.text.tex and de.comp.text.tex (where the
message either has probably registered by now).  I had a particularly
annoying mail exchange with some really unpleasant person that
simultaneously complained about preview-latex not being acceptable
because it was a security risk to unpack and install software from a
source archive cryptographically signed by somebody with less of a
name to lose than Donald Knuth, and whether it would not be possible
to have a Java-based installer which one could just click on.  I am
not kidding you.  Then he said that the pages contained no
testimonials (and one of Knuth, who is known never to use LaTeX, would
be desired.  Maybe we should make one up), and that I was an
untrustworthy source about any details or features of the package,
being its author.  And so forth and so on.  And that he'd prefer
software from authors which did not get cynical after a few loads of
this kind.

While this particular person was pretty much an obnoxious pest, it was
by this quality that I got any feedback at all.  Most people just shut
up silently.  And the one point that he was right about is that it is
close to impossible to guess from our web presence that Windows is a
supported platform.

> Personally I cannot remember hearing this at all.  Most people are
> aware that Emacs can be used on many different platforms and that
> this is also true for AUCTeX.

Nope.  Most people in our circles.

>> We really need to whack people with the information that all major
>> desktop operating systems are supported by now, and tell what one
>> needs to download on order to get from ground zero to a working
>> TeX/LaTeX/Emacs/AUCTeX/preview-latex/RefTeX installation depending
>> on your operating system.
>
> Hm, you mean something like a setup tutorial with a wider scope than
> the installation instructions in the manual, I guess.

At the moment I am not talking about optimal solutions.  At the moment
it is completely impossible to guess from our web presence:

a) can I use my operating system?
b) what will I need to download from here?
c) what will I need to download from other sites?
d) will my version of Emacs support this?
e) if I have not already chosen an Emacs or are not particularly fixed
   on one, what would the recommendations be?

All of this is, if present at all, scattered about the manuals, mostly
that of preview-latex, and some other files.

But people need to know this before they start downloading things and
thinking about "maybe I should try this out".

Of course, extensive tutorials for installation would also be nice to
have, but they are not essential right now.

> I am a bit concerned that there are so many combinations possible
> for setting up such a system.  There are different TeX systems on
> different operating systems, different package management systems
> offering different versions of Emacs, AUCTeX, and related software,
> and there is Emacs and XEmacs.  One cannot cover all these
> combinations and I am not sure maintaining a few selected ones is
> _that_ useful.

Believe me, it is. 10% of the available platforms cover 90% of the
prospective users.

> So while we are getting a lot of feedback from XEmacs users in
> general communication channels (during the last weeks and months it
> felt like relatively more than from Emacs users) there is nearly no
> feedback concerning AUCTeX development.  This is likely to result in
> problems after the release happened.

Not exactly new.  The day we have an AUCTeX release without specific
XEmacs problems is probably one to mark up in the calendar.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum