Re: [Texi2html-bug] Re: FSF assignment?

[Derek Price]

Hi Karl,

Karl Berry wrote:

> Would it?  I'm not so sure.  It seems similar to combining a GPLv2+ work
> with a GPLv3+ work to me.  The GPLv2+ work does not have to explicitly
> change its license in that case.

Well, no, a package with a "use Texi2HTML;" line wouldn't need to
explicitly change its license, but not to do so would be misleading
since that package and any packages which included it would now be
implicitly GPL'd, with the Artistic half of their licenses rendered moot.

I believe you are correct that this wouldn't affect packages that
implement a plugin interface via an instantiated object.  Only the
specific instance which was passed the GPL'd object would be affected,
so really, the package which must be GPL'd is the package that actually
includes the GPL'd package and instantiates the plugin object before
passing it on.

>     Do you have any ideas about how to approach this with rms?
> 
> Well, I can ask him with the rationale that it's conventional for CPAN
> and Perl modules.  I cannot predict his response.  If there are better
> freedom-promoting arguments, that would be good.  Popularity arguments
> (e.g., we'd get more users with it on CPAN) aren't likely to sway him.

I don't have any further objective arguments, but one of the things I
like about Perl is its friendliness and the friendliness of its
community.  The GPL has its place, and I support it on quite a few
projects, but I'm not inclined to start pushing it on the Perl folks,
who I think already do quite a bit for free software.  When in Rome...

Okay, here's an argument: I guess I see this like the LGPL.  I probably
couldn't make any arguments for using the LGPL that haven't already been
made, but I don't want to release a Perl package, basically a library,
that forces its license on all callers.

> Also ... how close is this to fruition?  I mean, can we wait until we're
> at the point of actually having texi2html in texinfo before waking up
> rms?

Um, well, I think I could convert the script and init files into a set
of Perl packages in a week, tops, and the CPAN build/distribution
structure would be a fairly trivial foundation I would lay down beneath
that, but I don't really want to start the work if I don't know that it
can be released under Perl's dual license.

Moving Patrice's test suite into the CPAN layout may be a little more
challenging, and I can't say precisely how long it would take since my
earlier analysis of it remains incomplete.  Last time I looked at it, it
appeared to be dependent on some specifics of Patrice's environment
which I never tracked down, but I would hope that would take another
week, maybe less.

I already sketched out what would need to be done to convert the scripts
to modules in another thread, and I expect Patrice could work from that
if he is so inclined, but I don't have a feel for how long it would take
him.  I also don't know how familiar he is with the CPAN build structure.

Assuming we can release the Perl package under the dual license, I would
wait until Patrice finishes up a few things he is working on, including
a few bug fixes, so I can test my changes without having to clean up
after him.  Patrice estimated he could finish those changes in about a
week, and I could start as soon as he is done.

As far as moving the script into Texinfo before doing this work, I think
reorganizing the directory and file structure and then moving it into
the new repository would be simplest, but it's probably no biggie either
way.

Waiting for Olaf Bachman's paperwork before consulting rms might make
sense, I suppose.  I don't have any reason to believe Olaf won't assign
his copyright to the FSF, but his refusal might render a positive
response from rms moot anyhow.

Derek
-- 
Derek R. Price
Solutions Architect
Ximbiot, LLC <http://ximbiot.com>
Get CVS and Subversion Support from Ximbiot!

v: +1 248.835.1260
f: +1 248.246.1176