On Sat, 14 Oct 2006 15:51:59 -0700, Martin Albrecht
<[EMAIL PROTECTED]> wrote:
> On Sunday 15 October 2006 00:22, William Stein wrote:
>> William Stein said:
>> >> This was one very concrete thing I got out of the SAGE
>> >> foundation discussion on Sunday -- people want to keep the
>> >> copyright of the work they submit to SAGE, just like some
>> >> people do with the Linux kernel.
>>
>> Bill Page replied:
>> > Yes. Personally I think this is a poor choice. They do not seem
>> > to realize how difficult this might make it in the future to
>> > incorporate their work into software with different licensing
>> > conditions. :-( This sort of copyright policy is what prevents
>> > the Aldor program from being licensed as open source right now.
>>
>> Since you think it is a poor choice, I better respond.
>> My impression is that in fact the SAGE developers who want
>> to keep copyright do understand "how difficult this might make
>> it in the future to incorporate their work into software with
>> different licensing conditions." In fact, that difficulty
>> is precisely what they want. A major reason they are motivated
>> and willing to contribute their code to SAGE (under the GPL), is
>> that they know beyond a doubt that their work can never be
>> incorporated into a commercial product, (etc.), even if I
>> totally sold out and tried to do so.
>>
>> SAGE developers -- what do you all think?
>
> It's not a "commercial product" but a proprietary product which is ruled
Sorry, I meant "non-free". I shouldn't have written "commercial", since
that is misleading; I'm well aware of this distinction. I meant,
e.g., incorporating SAGE into Magma or Mathematica, say, and making tons
of bug fixes, but charging people for the version of SAGE with the bug
fixes.
> But concerning your question: Yes, I want that code I contribute stays
> free
> both as in free-speech (<insert math should be verifiable argument
> here>) and
> in free-beer (< insert that-drove-me-to-SAGE-in-the-first-place>). I
> definitely don't want to hand over copyright but I'm not sure if I'm
> totally
> opposed to sharing the copyright (as I do with you - William - IIRC
> anyway)
> with some to-be-founded foundation at some point.
Some relevant facts:
(1) You and David Harvey both feel very strongly that
you "definitely don't want to hand over copyright".
You are also two of the top contributors to SAGE.
(2) The GPL makes it possible for SAGE to develop and work
indefinitely in a wide variety of ways without
requiring a single entity have copyright.
(3) This copyright model, where all code is GPL'd (or compatible),
and individuals hold the copyright, is what is used by
the Linux kernel, which is generally considered to be
a successful open source project.
On Sat, 14 Oct 2006 15:41:25 -0700, Justin C. Walker <[EMAIL PROTECTED]>
wrote:
> Part of Bill's comment deals with the mixing of copyright policy.
> We've just finished dealing with the impact of mixing policies
> (Singular/omalloc), so it's pretty clear why this is A Bad Thing.
Agreed. My goal with SAGE is that all components be "licensed under
a free open source license as defined at http://www.opensource.org/.";
The licenses that qualify are listed here:
http://www.opensource.org/licenses/index.php
With open source, it seems that the license is much more important
than the copyright holder.
> On the other hand, I think mixing "BSD-like" and GPL isn't inherently
> bad ...
It's fine. SAGE has a number of BSD'd components. E.g., I BSD'd the
interrupt.h file in the SAGE library, so it could be used by scipy.
For various reasons, I think that BSD is the wrong license for the
majority of SAGE.
> There are a lot of murky areas here, so perhaps as part of setting up
> the SAGE Foundation, 'we' could deal with a lawyer who knows about
> this kind of stuff.
But a lot of this is also about understanding what SAGE developers
genuinely feel good about. I doubt a lawyer will help much with that.
I am personally very comfortable with releasing my code under the GPL
and keeping copyright, and I know a number of other people are also.
> One thing to consider, individually, is the question of the code you
> contribute, embodying an algorithm that you think up, could be
> something you want to capitalize on at a later time. I don't know
> what the issues are in this case, but it's worth a second thought.
If you design an algorithm, then implement it and contribute it to
SAGE under the GPL, but you keep the copyright (i.e., put "Copyright
Justin Walker...") in the file, the situation is very flexible.
At any point in the future you could take that same file (but without
any bugfixes that I might have added!) and release it under any
license you want. E.g., you could license it to Mathematica.
Then Mathematica could use it without showing the code
to anybody, and they could have their own proprietary bug fixes.
If you assign the copyright to the SAGE Foundation, then the above
relicensing to mathematica is not possible for you.
-- William
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~----------~----~----~----~------~----~------~--~---
