Re: Towards a new LPPL draft
Jeremy Hankins writes:
> Jeff Licquia <firstname.lastname@example.org> writes:
> > OK. Now I'd like to hear the Debian side. Here are the conditions for
> > modification that are being proposed as I understand them:
> > - you must rename all modified files, or
> > - you must rename the whole of LaTeX in your modified copy AND
> > distribute a pristine copy of LaTeX as well.
> > Comments? Branden, Walter, Mark, and Jeremy, I'm especially interested
> > in your opinions, since you three are the current objectors.
> Yikes. I'd accept the former as free before the latter, personally.
> Giving users options is one thing, but option two seems to suggest
> that if Latex is forked for some reason we'll need to ferry around the
> original (from the date of the fork) version of latex whenever
> distributing the new version, forever. That's a far more onerous
> requirement than file renaming, imho.
let me first qualify the suggestion that Jeff made above
- the reason for it is to give the user the possibility to exchanges
documents with other using pristine LaTeX and obtain identical output
- it therefore quite pointless to carry around some old pristine LaTeX from
the day of the fork; if the above suggestion has to have value to the goals
we try to achieve with the LPPL license, then the suggestion would be to
keep a copy of current pristine LaTeX, though I doubt that could or should
be codified except as a suggestion.
> Re the renaming requirement, I disagree with the statement that the
> definition of file is a minor technical detail. Do the Latex folks
> really want to put this requirement on things like a readme.txt file?
> And what if Latex is ported to a filesystem that doesn't have
> filenames (or at least, not in the same sense)? Is the directory
> structure part of the filename?
the answer to the first is no and to the second "most likely no, except in
very unusual circumstances". To qualify that: I suggested at one point that
instead of requiring the filename rename it would be enough to achieve the
goals of the LaTeX community to require that from within a pristine LaTeX
kernel the original work and the derived work are distinguishable. In other
in a redraft of LPPL i wuold try to limit any renaming requirement to such
files that need to change their file names in order to make this distinction
(or use the requirement to distinguish as the requirement rather than the
file name and only remark that in most cases this might mean that certain
files need new names. -> readme.txt would not need to change.
from that it also follows that if on an OS with a different type of
filestructure (say MVS) that revised requirement would have other effects
(i've forgotten what the things got called on mvs, but all the classes lived a
single datastructures with members there, and so you wouldhave to chose a
different member name)
does this answer the question?
> I would be very interested in hearing someone from the Latex folks try
> to say precisely what should change in order to ensure than no one
> accidently use the wrong file from their document. The way a packages
> is called from a document (but that's potentially broken by the
> filename remapping -- perhaps the way it's called, assuming no
the answer (for me) would be
- if a file is not used by pristine LaTeX (that is a LaTeX kernel without the
remapping feature added) then there is no need for a renaming of any kind
- otherwise, if it is an object used at some point in the game by the
underlying macroprocessor (ie loaded) then the "name" used for loading
should be different --- that is not thefile name though in most
implementation it is a one to one mapping (which is why we suggested to use
the filename rename as an requirement.
> But that said, I'm not very certain in my own mind about the file
> naming issue. I'm uncomfortable with it, but there does seem to be a
> good reason for it (whether that reason is arguable is completely
> separate issue). But if we can't come up with a better way to make
> the restriction, I'm not going to say we should keep it out of Debian.
well thanks for saying that there seems to be a reason for it. I personally
would feel mcuh easier if that would be different in TeX and one could ensure
uniformity in a different way. but being as it is, I see no good alternative
as trademarking several hundred file names isn't (and wouldn't help Debian
either). This is why I thought and still think that a completely free hand
through the global remapping feature isactually giving everybody the best of
both worlds and is complient with DSFG
To UNSUBSCRIBE, email to email@example.com
with a subject of "unsubscribe". Trouble? Contact firstname.lastname@example.org