SablePath (was Re: Classpath future?)

[Etienne M. Gagnon]

Hi Mark,

This will be my last post about the licensing issue.  As I state below, 
I cannot live with the current practices of the Classpath team (unless 
Mark is not talking in the project name).

To me, all contributions are important, as little as they might be.  For 
obvious purpuses, I agree with Classpath's requirements for "clean-room" 
status, but I disagree with the Copyright assignment policy, and the 
extent of the GPL exception text (that allows redistribution of derived 
work without the source code of the "Free" parts).

Sadly, for practical reasons, I will continue to maintain a parrallel 
class library version, called "Sablepath", as primary class library for 
SableVM, and potentially other VMs.  This library will be (is) based on 
Classpath, and will continually integrate any changes appearing the the 
Classpath CVS tree.

[My policy for contributions will be much more relaxed, and discussion 
about everything in it is open, including: project name, project 
management, build process, etc.  Every contributor, as small as it may 
be, will be recognized as a very important person to the project.  To 
any potential contributor:  I still believe that if you have the 
patience and willigness to assign your code to the FSF, then you are 
better of contributing to the upstream project "Classpath".

Hopefully, one day, when the sume of all our little contributions will 
be imortant enough, the Classpath project will reintegrate our code 
within its main stream development tree, without requiring Copyright 
assignment, nor unnacceptable licensing requirements.]

If you are interested, you can contact me privately, off the classpath 
mailing-list (which is clealy innapropriate for this type of discussions).

I insist that I am sad that it should come to this.

Mark Wielaard wrote:

> I do do think the wording you suggest is clearer then the current
> wording we use now. But I don't think we should change the license (now).
> The license has been agreed upon and I am only willing to discuss changing
> it if there is a very clear benefit. So if you are saying I have written
> a java.nio or java.swing package and I am only willing to donate it to
> the Classpath project if you change the license then I am all ears.


So, I'll speak for all those little, unimportant contributors, that 
would simply like to contribute a few things to Classpath easily, 
without going through months of awaiting for paper work, and would like 
to get their name in the Copyright holders list (as sole reward for 
their little, yet maybe important effort).

I disagree with your attitude.  I disagree with the "GPL exception" 
text. I would like to make small contributions easily.  I do not want to 
allow redistribution of executables containing some of my work without 
the source code to my work.  I do not want to donate my Copyright to others.

I think that every contribution, as small is it might be, is important 
to a Free software project.  It is with many little contributions that 
you might end up with a big stable project.  I might not have the time, 
or the resources, to write a full blown java.swing package, but maybe I 
can go and fix a few but difficult bugs/issues within my area of 
expertize.  Also, as a professor, I might encourage some of my students 
to contribute to the project, which would finally add up to a big 
contribution, overall.  But, by refusing my little help, and that of 
others, you are refusing a lot.

> The license was also choosen because it was based on the
> wording that was used in the C++ standard library licence.


I do not care about the origins of the license, I care about its 
content.  [The BSD license, for that matter, is probably older than your 
exception text.  Does that make it acceptable to you?  Probably not, 
because you don't agree with the content.]

> We should only discuss changing the current license if we can guarantee
> that all these projects and the GNU project as a whole will be better off
> ...


I was only discussing the exception text within the scope of the 
Classpath project.  Sure, if you want to involve a few hundred of 
developers, then you are never going to build a concensus.

> Please give me a clear (concrete) incentive to even consider the question
> of a license change.


Because it is too permissive, and does not protect the Freedom of the 
code (which is the freedom of users receiving the "Free" part of the 
software to inspect that part's source code, among other things). 
Mainly, this exception contradicts freedoms 1, 2 and 3. 
(http://www.gnu.org/philosophy/free-sw.html)

But, this only applies my my "potential" contributions.  I cannot talk 
for others.

> I do appreciate all the issues you brought up Etienne and I am happy to
> see that there is so much discussion about all the topics now on the list.
> So I hope this email doesn't sound to arrogant, but I have seen license
> discussion often being very unproductive (flamewars) and take up a lot
> of energy (because it is an important issue) but without a clear indication
> of any gains to the project.


As I said, my intention was not to start a flamewar, nor a long license 
discussion.  I even went as far as suggesting solutions, including a 
"fixed" exception text, to shorten the discussion.  But I gather that 
these issues are not important to the current Classpath members.  I will 
not keep you away from your good work anymore to try conciling divergent 
views.  I will do, instead, as I think is best for everybody.  Work on 
my side and let you work on yours, and take it on me to continually 
update my code with your newer development tree.

I know that the FSF is pretty keen on the Copyright assignment issue, so 
unless you can find a quick solution to it, I see no reason to try avoid 
this fork.  Anyway, unless I am wrong, I am sure most people will prefer 
working on the official tree, and assign their code to the FSF.  I just 
won't assign mine, and I need to apply some patches to Classpath, so I 
have no choice to fork.

Respectfully,


Etienne
--
+--------------------------------------------------------------------+
| Étienne M. Gagnon                    mailto:address@hidden |
| Professeur adjoint            Téléphone: (514) 987-3000 poste 8215 |
| Bureau: PK-4930                        Télécopieur: (514) 987-8477 |
| Département d'informatique, UQÀM          http://www.info.uqam.ca/ |
| Auteur de SableVM                          http://www.sablevm.org/ |
| et de SableCC                              http://www.sablecc.org/ |
+--------------------------------------------------------------------+
| Etienne M. Gagnon                    mailto:address@hidden |
| Assistant Professor                Phone: (514) 987-3000 ext. 8215 |
| Office: PK-4930                                Fax: (514) 987-8477 |
| Department of Computer Science, UQAM      http://www.info.uqam.ca/ |
| Author of SableVM                          http://www.sablevm.org/ |
| and SableCC                                http://www.sablecc.org/ |
+--------------------------------------------------------------------+