classpath
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Fwd: Sablepath packages available to download]


From: Etienne M. Gagnon
Subject: Re: [Fwd: Sablepath packages available to download]
Date: Thu, 13 Sep 2001 09:36:46 -0400
User-agent: Mutt/1.3.20i

On Thu, Sep 13, 2001 at 02:07:49PM +1200, Bryce McKinlay wrote:
> >I would like to do so, but the Classpath project has yet to accept my 
> >offer to contribute my modifications under my own Copyright (but 
> >identical license).
>
> I agree that its annoying, and actually I would personally much prefer 
> if we didn't have to go through the hassle of getting copyright 
> assignment from contributors. However, there are some fairly convincing 
> arguments as to why the FSF wants to hold the copyrights on its major 
> projects, in order to be able to defend the license in court if it ever 
> came down to that. 

Are you telling me that the FSF would not be able to defend their own work?
Probably not.  Just that they would have trouble defending mine, which is
my own problem, no yours or the FSF's.

The current situation is that I will not assign away my Copyright to anybody
or any corporation or charitable organization.  You must also consider that
my future graduate students will work on improving SableVM and its class
library, and I will not be in a position to conditionally accept their
contributions based on whether they assign their copyright to a US based
charitable organization or not.  Just try to imagine the consequence of
this in a Quebec, French speaking university...

So, take for granted that I will not assign my copyright, nor will my
students.  This means that the Classpath project, not the FSF or anybody
else, has to decide whether it will accept contributions without copyright
assignment (or public domain release of these contributions).

The Classpath license was designed to allow contribution without copyright
assignment (this is the whole point of the GPL: give freedom of choice),
but the project can chose to act differently for the maintenance of the
official tree.  It is the project's choice.

Personally, I will not stop my research, or refrain myself from fixing
Classpath bugs, or improving its build process and code organization, if
you do not accept my contributions; I will simply do it in another
developement tree.  Do not worry, I will conspicuously leave intact all
copyright and license clauses, and fully document which files I modified,
and so on, as required by whatever license Classpath finally chooses.  I
was simply trying to discuss whether you would be willing to relax your
rules somewhat as to avoid splitting the development tree.

> All of the current and past authors of classpath - 
> some of which have no doubt committed far more time and energy than you 

Let's not get into that either... I'm simply contributing a complete state-of-
the-art research VM.  It's a minor contribution, but some think it has some
value...

> >5- distribute/Sell the executable without distributing the sources of 
> >Classpath nor the modifications.  In fact, there's even no obligation to 
> > disclose that Classpath was ever used.  In other words, the current 
> >Classpath license is weaker than the BSD license (without adv. clause)!
> >
> Could you please explain this further. This is certainly not the intent 
> or spirit of the classpath license as I understand it, and if true would 
> indicate some kind of loophole.

Brian has summarized pretty well our 4 way discussion (RMS, Brian, Paul & I).

My proposal would simply be: as all we want it "to get rid of the LGPL
minus the upgrade clause, why not license Classpath under the "LGPL +
an exception about the upgrade clause"?

Something like:

Classpath is licensed under the terms of the GNU ... (LGPL) ...

As an exception to the terms of the GNU LGPL, you can replace
section "6 a)" with the following:

    a) Accompany the work with the complete corresponding
       machine-readable source code for the Library including whatever
       changes were used in the work (which must be distributed under
       Sections 1 and 2 above).


This would probably clear-up the matter, as it simply removes the
requirement to provide the proprietary modules in "re-linkable form",
which was the second part of section 6.a).

How about it?


Etienne
-- 
Etienne M. Gagnon                    http://www.info.uqam.ca/~egagnon/
SableVM:                                       http://www.sablevm.org/
SableCC:                                       http://www.sablecc.org/



reply via email to

[Prev in Thread] Current Thread [Next in Thread]