From: Raffael Cavallaro on
On 2010-03-25 06:06:09 -0400, Andrew Haley said:

> There's nothing ironic about it. The FSF seeks to maximize freedom,
> so licenses code whichever way works best. Libraries sometimes have
> different needs from applications.

Which is why I suggest that Pascal's lisp libraries would be more
useful licensed under the LLGPL than the GPL. It's ironic because the
FSF is the creator of the GPL, and even they recognized that the GPL
was a poor fit for libraries which is why they created the Library (now
Lesser) GPL.
--
Raffael Cavallaro

From: Hyman Rosen on
On 3/25/2010 9:44 AM, Raffael Cavallaro wrote:
> It's ironic because the FSF is the creator of the GPL,
> and even they recognized that the GPL was a poor fit
> for libraries which is why they created the Library
> (now Lesser) GPL.

The FSF does not believe that the GPL is a poor fit for
libraries. They believe that when there are good non-free
alternatives to free libraries, they should use the LGPL
for the free libraries so that users will have at least
some freedom. When there are no good non-free versions
available, they will use the GPL to maximize freedom for
users.
From: Tamas K Papp on
On Thu, 25 Mar 2010 09:44:14 -0400, Raffael Cavallaro wrote:

> On 2010-03-25 06:06:09 -0400, Andrew Haley said:
>
>> There's nothing ironic about it. The FSF seeks to maximize freedom, so
>> licenses code whichever way works best. Libraries sometimes have
>> different needs from applications.
>
> Which is why I suggest that Pascal's lisp libraries would be more useful
> licensed under the LLGPL than the GPL. It's ironic because the FSF is
> the creator of the GPL, and even they recognized that the GPL was a poor
> fit for libraries which is why they created the Library (now Lesser)
> GPL.

I disagree -- I don't think that the FSF considers the GPL a "poor
fit" for libraries. Quite the opposite (see [1]). They just
recognized that in certain situations, some people would prefer
something like the LGPL, and I guess that they wanted to give them the
choice. But the GPL is still the option they recommend, even for
libraries.

[1] http://www.gnu.org/licenses/why-not-lgpl.html

Regarding the broader issue (of how people license their libraries): I
think this is an optimization problem where people have heterogeneous
objective functions, and thus trying to convince people to pick
another license is not always a worthwhile.

It is possible that someone using a GPL/LGPL/LLGPL/BSD/MIT/... license
is perfectly aware of the advantages and disadvantages, it is just that
they decided to make a different choice. In which case, threads like these
are unlikely to be fruitful.

Regards,

Tamas
From: David Kastrup on
Hyman Rosen <hyrosen(a)mail.com> writes:

> On 3/25/2010 9:44 AM, Raffael Cavallaro wrote:
>> It's ironic because the FSF is the creator of the GPL,
>> and even they recognized that the GPL was a poor fit
>> for libraries which is why they created the Library
>> (now Lesser) GPL.
>
> The FSF does not believe that the GPL is a poor fit for
> libraries. They believe that when there are good non-free
> alternatives to free libraries, they should use the LGPL
> for the free libraries so that users will have at least
> some freedom. When there are no good non-free versions
> available, they will use the GPL to maximize freedom for
> users.

More pragmatically: they want their licenses to be taken seriously.
That involves being able to go after violations in court and/or
settlements with good chances of success. Licenses covering a work "as
a whole" are hard to press when the material they cover is functionally
a drop-in replacement of existing non-free libraries. That makes "mere
aggregation" a really good defense.

--
David Kastrup
From: Hyman Rosen on
On 3/25/2010 10:05 AM, David Kastrup wrote:
> Licenses covering a work "as a whole" are hard to press
> when the material they cover is functionally a drop-in
> replacement of existing non-free libraries. That makes
> "mere aggregation" a really good defense.

This is completely wrong. The GPL applies to work as a whole
only when the GPL-covered work is made part of a combined
work and that combined work is copied and distributed.

Your statement sounds as if you continue to believe incorrectly
that a program which uses a dynamically linked library covered
by the GPL is subject to the GPL even when it is copied and
distributed without that library. That is not so. Copyright law
is about copying, and when a GPL-covered work is not being copied
and distributed, the GPL cannot come into play. What the program
does when it runs is not relevant for falling under the GPL because
the GPL does not restrict running covered works.

Similarly, mere aggregation is irrelevant to libraries which
are statically linked into programs. Such a combined work is
not a mere aggregation of the library and the other components.
Mere aggregation refers to including a covered work on a medium
of distribution along with other works.