Re: Thoughts about gnuplot license and copyrights

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

Dave Denholm wrote :
> Lars Hecking <lhe...@us...> writes:
>
>  =20
>>> I have concluded that Thomas Williams is the one who has to give its
>>> agreement before an official release. Can you confirm ? Would it be
>>>      =20
>>  That is correct. As the other copyright holder, Colin Kelley, is miss=
ing
>>  in action, Thomas Williams is our only contact.
>>
>>    =20
>>> possible to have further details on his position about the license ?
>>>      =20
>
>  =20
>>> Archives say that the gpl was refused. Is there a precise reason, or =
is
>>> Thomas William simply opposed to a change ? Another open-source licen=
se
>>> may fit the needs.
>>>      =20
>
>
> My recollection is that Thomas Williams is (or was) opposed the notion
> that gnuplot might ever fork into multiple, incompatible versions.
>
>  =20
>>  I have discussed development issues with Thomas in the past, and the =
only
>>  time I ever mentioned license changes (and the "single point of failu=
re"
>>  issue) there was no reply. I'm sure that particular email was receive=
d,
>>  and we were in contact again for the 4.0 release - which he liked a l=
ot :)
>>
>>    =20
>>> By the way, as the FAQ doesn't mention it, the actual position of the
>>> author against licenses could be a good addition.
>>>      =20
>>  I really have nothing to back this up, but people like Dave Denholm o=
r
>>  Alex Woo might.
>>
>>    =20
>
> I've dug out a very old mail folder "gnuplot.legal" which contains a
> thread from around August 1977 (wow, that's old...)
>
> But I suspect it's obsolete, since it was arguing about changing the
> license to allow distribution of modified binaries. Previously, this
> was completely prohibited, but it looks like the Copyright in the 4.0
> sources at least does grant permission to distribute modified
> binaries, with a few restrictions.
>
> At the time, this was enough to happify Richard Stallman (who actually
> approached me around this time to write a truly free gnuplot clone if
> the license wouldn't get changed)
>
> It's currently just a collection of separate mh files, but if anyone's
> interested, I can pack it into an mbox and upload it somewhere.
>
>
> Since I never actually got round to making a gnuplot release during my
> time as maintenaner, I didn't have to go through all this pain myself!
>
>
> dd
>  =20
Well, that's interesting information !
(Wow, 1977 ! I was not born, and my parents just get married !)

So, some years ago,Thomas Williams refused the gpl because forks would=20
be authorized. I may understand this position, as there is a risk to=20
lose the control over a program if one allows forks for it. However,=20
gnuplot is a long-standing software, and a slightly more permissive=20
license allowing direct redistribution of modified source code is a=20
decision for a brighter future, not a decision to make it disappear...

Unfortunately, as far as I know, there is no license which at the same=20
time is compatible with the gpl and forbids forks. Do you think Thomas=20
may have a different opinion about the gnuplot license now ?

As Ethan says, the current license is just enough to make distributions=20
package their binaries and apply a small patch over the source. But,=20
apart from distributors, I am concerned about the work done on the=20
trunk. Do you know if Thomas will ever be there to give its agreement to=20
a new release ? What will happen if someday he becomes unjoinable ?

Thank you very much for these details, Dave, I find it interesting to=20
know what happened to such great piece of work in time.

Best regards,

Timoth=E9e Lecomte