Re: Thoughts about gnuplot license and copyrights

[<tim...@en...>]

Dave Denholm wrote:

>Timoth=E9e Lecomte <tim...@en...> writes:
> =20
>
>>Robert Hart wrote:
>>   =20
>>
>>>>IMHO the gnuplot license is very close to the standard BSD license, w=
hich
>>>>imposes fewer restrictions than the GPL.
>>>>       =20
>>>>
>My understanding is that GPL comes with lots of words of legalese,
>but most of the restrictions are designed to prevent any one person
>from restricting anyone else from doing what they want with the code.
>
>BSD has few legal restrictions, but means that someone can modify the
>source and then distribute with more restrictions (including binary only=
).
>
>The one thing I do tend to feel strongly about is that if I put in
>time developing open-source software, I don't want someone else
>exploiting it for profit. BSD license doesn't prevent that, though GPL
>probably does.
> =20
>
Yes, I agree with this. My feeling is that the gpl is really the result
of a deep thinking to preserve copyleft, not only a long-and-unreadable
text.

>>>The purpose of the GPL, is not (just) to allow forks, or other indepen=
dent
>>>developments and patches, but also to allow code to be reused in entir=
ely
>>>separate projects. AFAICT, there is nothing in the gnuplot license tha=
t
>>>would let me (say) use the gnuplot terminal drivers as the start point=
 for
>>>some other graphical program.=20
>>>
>It may prevent the gnuplot terminal drivers from being
>distributed with another program. (Maybe not : it says you cannot
>distribute complete modified source tree, but doesn't see you cannot
>distribute a partial set of unmodified sources..?)
> =20
>
To my mind, a 'partial set of unmodified sources' is to understand as a
'complete modified source tree' according to the current copyright.

>However, it doesn't prevent you from designing and distributing a
>program which uses the same interface, allowing people who download it
>to link it locally with gnuplot terminal drivers. Well, that was one
>of the notions behind the terminal driver changes described in
>term/README
>
>I guess an analogy would be with gnu readline : the GPL
>prevents us from distributing a gnuplot binary that links with it, but
>someone building from source can link *their* copy with gnu readline.
>
>For a while, we (well, I) was considering breaking gnuplot into layers
>of libraries, with the lowest layer offering access to the graphics
>drivers, and the highest layer offering most of the gnuplot
>functionality. The gnuplot program would then just be a thin client of
>the library. But that never happened.
> =20
>
Well, that may happen one day ;-)
I agree that linking is possible provided some conditions (mostly a
'local linking') but it's definetely complicated !

>>>whilst happy to use gnuplot, are reluctant to get too involved in
>>>developing it because of that.
>>>
>[ reordering slightly]
>
>> am exactly asking myself the same thing : is it worth getting
>>seriously involved in gnuplot, if it someday vanishes just because
>>Thomas won't be joinable to give his agreement ?
>>   =20
>>
>Depends on your motives : when I got involved with gnuplot, it was
>because *I* was using it a lot, and I wanted to improve it for my own
>use. Contributing the changes let others benefit from them, but more
>importantly, meant I didn't have to rework the changes locally when a
>new version came out. At the time, I didn't realise how rare new
>releases were ;-)
> =20
>
On my side, I had some time this year to contribute to an open-source
project, and focused on gnuplot after having used octave for my physics
studies. Letting others benefit from my work is one of my primary goal !
And I would rather give my energy in something that I'm sure it has a
stable future.

>I was going to say that even if we can no longer make releases,
>there's always the CVS that people can download the source
>from. However, looking at the copyright again, could it be said that
>having the CVS available on the internet is a violation of
>
> * Permission to modify the software is granted, but not the right to
> * distribute the complete modified source code.
> =20
>
You're probably right. I assume that cvs is used with the private
agreement of Thomas.

>>(I can't imagine somebody writing a script that patches the original
>>source code, and then uses this modified code to patch his own code...
>>as we might guess from the current copyright),
>>   =20
>>
>
>Again, the license prevents distribution of such a work, but you can
>do anything you like with the sources locally.
> =20
>
On the contrary, I was thinking of a way to satisfy gnuplot's license,
ie to distribute the work as a patch plus the original source code...
But anyway it's not reliable.

>>* gpl, bsd, lgpl, mpl and other all allow to distribute modified code (=
I
>>mean in usual form, not as a patch, and without asking for permission).
>>This allows forks, but the most important is that it allows gnuplot
>>developpers to release when they want, without relying on the agreement
>>of somebody which may one day be unjoinable.
>>
>Maybe one way out is to persuade Thomas Williams to grant permission
>to a trusted nominee to release as and when required. Thomas would
>retain the option to revoke that permission at a time of his choosing,
>but it does mean that if he does vanish, there is one other person can
>release..
>
>
>dd
> =20
>
Yes, it should be better in the short run than the current situation.
But it's not really satisfying... as we will then rely on this trusted
nominee, which can also become unjoinable one day... unless the nominee
can name his successor if he wants to give away his responsibility...

I would really prefer to solve the whole licensing issue (easy release +
code reusability), either by choosing a well-known license (gpl or
compatible), or by allowing redistribution of modified source code and
binaries without restriction.

Thanks for your precious point of view, and your help in this situation !

Regards,

Timoth=E9e Lecomte