Re: [Evolution-hackers] Evolution: Taking forward...

[Patrick Ohly <Patrick Ohly gmx de>]

On Fri, 2008-07-11 at 04:21 -0600, Srinivasa Ragavan wrote:
> We have had a set of problems that we are carrying around for some
> time like :
> 
>       * Copyright assignments, which is not the best way looking for
> the future of Evolution. It sucks and sort of limits contributions to
> Evolution and we wanted to drop it.

I agree with Paul S. that this can become problematic during later
license changes. I suggest to either got for a "LGPL 2.1 or later"
license (thus basically choosing GNU licenses forever) or changing the
copyright agreement so that the copyright is transferred to some
independent, non-profit organization instead of Novell. The GNU
foundation perhaps?

>       * The current licensing incompatibility issues of Evolution with
> Samba4/libmapi (GPLv3). Evolution needs to link with libmapi/samba4
> for the new mapi based connector being developed for Exchange 2007.
>  
> So here is the plan :
> 
>       * Drop Evolution copyright assignments and make it really easy
> to contribute to Evolution
>       * Move Evolution licensing to  "LGPL v2 and LGPL v3" to let us
> re-use the code more easily around the platform.  This also moves us
> closer to Thunderbird's MPL/LGPL model.

This should be "LGPL v2.1 or later or GPL v3 or later". As long as the
code is not linked against libmapi (which is GPL3 as Paul B. pointed
out), the resulting binary can be released as LGPL. Code can be moved
from main Evolution into libs and vice versa, which wouldn't be the case
if the code of the executable is only licensed under the GPL.

As soon as source is compiled into an executable which is also linked
against libmapi, then the resulting work and its source has to be
distributed under GPL3. The license of the Evolution source allows that,
so you are fine. The shared libraries can remain under LGPL: GPL
binaries can use LGPL libraries.

> It would be really helpful if you can post a public/explicit mail with
> permissions to do it, or code pointers - if you think you wrote a
> piece of Evolution code & object.

I haven't contributed much code and when I did, it was with copyright
transfer. So no objections from my side.

-- 
Bye, Patrick Ohly
--  
Patrick Ohly gmx de
http://www.estamos.de/