Picking a license

[Paul Boddie]

On 10 Mai, 17:01, Patrick Maupin <pmau... at gmail.com> wrote:
>
> I'll be charitable and assume the fact that you can make that
> statement without apparent guile merely means that you haven't read
> the post I was referring to:
>
> http://www.gnu.org/philosophy/why-not-lgpl.html

Of course I have read it, and not just recently either. But this is a
position paper by the author of the licence, and it doesn't mean that
someone who has written a GPL-licensed library completely agrees with
that position. And anyway, it's a case of "take it or leave it" - it's
not like the author or the FSF are sneaking stuff into every product
and every corner of the market and then telling you that you can't
"unchoose" their stuff.

[...]

> Legally, I don't think they can dictate the license terms of, e.g.
> clisp just because it can link to readline.  But practically, they DID
> manage to do this, simply because Bruno Haible, the clisp author, was
> more concerned about writing software than spending too much time
> sparring with Stallman over the license, so he finally licensed clisp
> under the gpl.  clisp *could* use readline, but didn't require it;
> nonetheless Stallman argued that clisp was a "derivative" of
> readline.  That case of the tail wagging the dog would be laughable if
> it hadn't worked.  In any case, Stallman's success at that tactic is
> probably one of the things that led him to write the paper on why you
> should use GPL for your library.

Although it seems quite unfair, the e-mail discussion about the
licence does show that Stallman was not initially convinced that works
should be affected in such a way (with regard to the Objective-C
compiler developed by NeXT), and that Haible was not strongly opposed
to changing the licence. You can argue that Stallman overreached by
demanding a licence change and that consideration of such matters has
progressed since that time, but Haible always had the option of not
using or supporting readline - only the latter is contentious, and the
obligation of GPL-compatible licensing (as opposed to GPL-licensing)
now diminishes how contentious this is today.

[...]

> I think that, legally, they probably don't have a leg to stand on for
> some of their overarching claims (e.g. about shipping proprietary
> software that could link to readline, without even shipping
> readline).  But morally -- well, they've made their position
> reasonably clear and I try to abide by it.  That still doesn't make it
> "not really FUD."  I'd call this sort of badgering "copyright misuse"
> myself.

Again, you have to consider the intent of the licensing: that some
software which links to readline results in a software system that
should offer the "four freedoms", because that's the price of linking
to readline whose licence has promised that any system which builds
upon it shall offer those privileges.

> > As for rst2pdf, what your modifications would mean is that the
> > software would need to be redistributed under a GPL-compatible
> > licence.
>
> That's parsing semantics rather finely.  In practice, what it really
> means is that the combination (e.g. the whole program) would
> effectively be GPL-licensed.  This then means that downstream users
> would have to double-check that they are not combining the whole work
> with licenses which are GPL-incompatible, even if they are not using
> the svg feature.  Hence, the term "viral."

Once again, I refer you to the intent of the licensing: if someone has
the software in front of them which uses svglib, then they need to
have the privileges granted to them by the GPL. Yes, if the software
also uses some component with a GPL-incompatible licence, then this
causes a problem.

[...]

> http://www.gnu.org/licenses/gpl-faq.html#GPLInProprietarySystem
>
> "A system incorporating a GPL-covered program is an extended version
> of that program. The GPL says that any extended version of the program
> must be released under the GPL if it is released at all."
>
> This makes it clear that the overall work must be GPLed.  Now, all of
> a sudden, downstream users cannot do some things they could have done
> before.  Can you not see that taking a preexisting MIT-licensed
> project and adding code to make it GPL could negatively affect some of
> its users and that that is not necessarily an unalloyed good?

Well, I have referred several times to WebKit without you taking the
hint, but that provides a specific case of a project which is LGPL-
licensed despite being based on (in GPLv3 terminology) libraries which
were distributed under the GPL and combined with that software.
Similarly, the effort to ensure that CPython's licence was GPL-
compatible had a lot to do with the right to redistribute with GPL-
licensed code (actually readline, if I remember correctly).

[...]

> > Well, even the FSF doesn't approve of trivial projects using the GPL:
>
> >http://www.gnu.org/licenses/gpl-faq.html#WhatIfWorkIsShort
>
> Sure, that's a pragmatic view -- copyright might not even be permitted
> on something that short that is mainly functional.  However, length is
> not the only arbiter of trivial.  To stay with the same example,
> personally, I would consider readline "trivial" within the context of
> a lot of software which might use it, regardless of whether the
> readline implementation itself used all sorts of fancy neural net
> technology to predict what word the user was going to type or
> whatever.  But whether it was trivial or not, if I ship software that
> *could* link to it but doesn't *require* it (like the case of clisp)
> without shipping readline, I think it's FUD and an attempt at
> copyright misuse to call my software a derivative work of readline.
> But obviously YMMV

Is readline trivial? Was readline trivial in 1992? Does it even
matter, because the author is more or less saying that they don't want
their code incorporated in a proprietary system? It's interesting to
see that GPLv3 doesn't talk about derived works or derivatives (at
least not as much as GPLv2), but instead talks about things being
"based on" other things, but as I've already said, at the point of
someone running a bunch of software components together, the intent of
copyleft licences is to say that the user should be able to take that
(or part of it, in the case of "weak copyleft" licences) and change,
recompile and distribute its sources, modified or not.

Paul