Re: GNU Arch moves to GPL v3 or later

Thomas Lord <[email protected]> Fri, 07 Sep 2007 21:29:36 -0700
Newsgroups gmane.comp.version-control.arch.user
Message-ID <[email protected]>


I think that that's a solid complaint in the abstract.   A good rule of 
thumb
would be to do a little extra work and keep "supporting" GPLv2 just
because -- well, it can only help.   But, not in this case, probably.

It's the specifics of the case here that matter.  I love a great deal of 
all that
code but, you know, not *that* much.  It just isn't *that* precious.    
If anyone
were to aim for heavy duty re-use of some part, I'd almost certainly say 
"copy from
that, kind of trace some of the ideas, but basically rewrite for a 
cleaner form".  
This isn't a code base I think needs to be overly protected for maximum 
possible
copyleft re-use.   

If it does turn out that someone is really screwed because they need a 
v2 form,
hopefully they can speak up and arrangements can be easily made.

-t


Stephen J. Turnbull wrote:
> Stefan Monnier writes:
>
>  > > Er, why make Arch code license-incompatible with a lot of GPLv2 code
>  > > that's out there, such as the Linux kernel and (for now) XEmacs?
>  > 
>  > Last I looked XEmacs still was distributed under the GPL>=2, so it is
>  > perfectly compatible with GPLv3.
>
> Sure, if somebody wants to incorporate XEmacs in a future version of
> tla, they're free to do so, and we'll applaud their taste.  But free
> software is about cooperation, and respecting others' goals.  If a
> project, for any reason, wishes to continue to use "GPLv2 or later",
> then a move by Arch to GPLv3 makes Arch code unavailable to that
> project.[1]  I would think that any free software advocate would
> consider that in principle a bad thing, and thus want to balance it
> with specific advantages of the new license.
>
> Linux and XEmacs are simply examples of that kind of project and of
> the variety of reasons why people might wish to continue to use the
> same licensing.
>
>
> Footnotes: 
> [1]  Note that this argument is isomorphic to the logic that Richard
> Stallman uses for claiming that XEmacs's failure to collect
> assignments makes XEmacs code "unavailable" to Emacs, although the
> specific obstacle is different.
>
>
>
>
> _______________________________________________
> Gnu-arch-users mailing list
> [email protected]
> http://lists.gnu.org/mailman/listinfo/gnu-arch-users
>
> GNU arch home page:
> http://savannah.gnu.org/projects/gnu-arch/
>
>