"Stephen J. Turnbull" <stephen(a)xemacs.org> writes:
David Kastrup writes:
> You may convey a work based on the Program, or the modifications to
> produce it from the Program, in the form of source code under the
> terms of section 4, provided that you also meet all of these conditions:
>
> a) The work must carry prominent notices stating that you modified
> it, and giving a relevant date.
What's this all about, David? A copyright notice is normally taken to
be sufficient.
By whom? I am quoting the license conditions. It would appear that you
change the files around in order to fit XEmacs. That's certainly
permitted, but in the course of being able to properly identify bugs
reported to the AUCTeX team as a consequence of that, I am pointing out
that I have good cause that such changes are clearly marked as not being
the responsibility of upstream. I am aware that many terms of the GPL
are not usually taken very seriously, and am not usually trying to
bother people all too much.
In this particular case however, I would appreciate it if this
particular clause was heeded. Not because I am an unreasonable
troublemaker, but because the dissemination of incompatible versions is
worrisome for me. I can see where those changes appear technically
necessary for your version, but I want them clearly marked out. And it
happens that this is not too much to ask since the GPL conditions
support my desire.
If the LaTeX code were instead licensed under the LPPL (LaTeX project
public license), you would be required to change the file name before
changing the file, somewhat more of a hassle since that would require
changes in callers. So you see that this worry of incompatibility in
the context of LaTeX is not specific just to me.
The GPL, as opposed to the LPPL, does not reserve the file name space
for upstream. But it at least demands a prominent notice of the changes
done. I don't think that this is too much to ask, even if the license
did not spell it out.
In any case, adding kilobytes of comment and/or interactive notice
to
explain that this is not David Kastrup's version will presumably not
solve the issue. Perhaps you could hazard a guess as to what's going
wrong here?
CVS version tags are used for versioning the released files. Obviously,
the format of your version control system tags differs, as do the actual
versions. So when LaTeX tries parsing the version control tags, the
unforeseen format throws it off the track. The proposed changes (I have
not analyzed them thoroughly), would likely make the style file work.
However, I consider it likely that the resulting log file from the TeX
run (when using the \listfiles command, for example) would contain
absolutely no trace of this being a special version, but rather make it
appear like an upstream version with a strange version number.
That's not really satisfactory: typically, bug reports made on TeX
specific lists are made using the log output and including the output of
\listfiles. Version compatibility of LaTeX packages is usually assured
by demanding a file with at least a particular modification date. When
you use your own dates and CVS tags, this mechanism is broken.
It is important for diagnosing problems that the version is indicated as
not being the normal upstream version.
Why do I have to beg and rationalize to make you adhere to the license
conditions? I would consider it common courtesy to heed my request even
if you were not legally required to do so. What's the point in going
into confrontation mode about this?
As so often, I don't understand what you are trying to accomplish.
Fortunately, I am not dependent on my understanding in this case. Just
heed the license, mark the changes in the prescribed manner, and be
done. It is not like this would be onerous to do.
--
David Kastrup
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta