licensing choice [was RE: D/BUS IDL compiler ... #2 ...]

[Mark McLoughlin]

Hey,

On Thu, 2004-03-25 at 17:07, Michael Meeks wrote:
>  I'm happy to commit to CVS, although still concerned by the
> ongoing, and deepening license uncertainty.

	So, I had real trouble understanding what uncertainty Michael was
referring to here, so we had a long conversation about it on irc and I
thought I'd summarise as best I could here.

	The issue is that both of D/BUS licenses look like they are going to
significantly hinder D/BUS's adoption and since the intent is that D/BUS
would be very much an integral part of free desktops we should be
choosing a license that makes it *easier* to build consensus around
using D/BUS, not harder.

	So, the easy part is the issue with using the GPL license - basically,
if D/BUS is used throughout e.g. GNOME, any licensee who uses the
chooses GPL license for D/BUS immediately causes the viral effect of
licensing the GNOME platform under the GPL, and thereby making it
illegal to use the GNOME platform from a proprietary application. GNOME
clearly intends it to be possible to license the platform under the
LGPL.

	The issue with the AFL 2.0[1] license is less clear and hinges around
Section 10 "Termination for Patent Action", quoted here:

  10) Termination for Patent Action. This License shall terminate
      automatically and You may no longer exercise any of the rights
      granted to You by this License as of the date You commence an
      action, including a cross-claim or counterclaim, for patent
      infringement (i) against Licensor with respect to a patent
      applicable to software or (ii) against any entity with respect to
      a patent applicable to the Original Work (but excluding
      combinations of the Original Work with other software or
      hardware).

	Obviously, if you really care about this you should seek legal
interpretation from a lawyer, but the way I read it is that the Licensor
(the joint copyright holders of D/BUS) agrees that the licensee may
continue to use D/BUS so long as:

  a) the licensee takes *no* software patent action/counter-action 
     against any of the D/BUS copyright holders and

  b) takes no software patent action/counter-action against *anyone* for
     infringement of any patent which is applicable to D/BUS

	As an example of the problems that may arise, assume for a minute that
IBM have contributed some code to D/BUS. If some small software company
has been building up a defensive patent portfolio over a number years
and is also a D/BUS licensee, then if IBM choose to take a patent
infringement action against the company, the company has the choice of
either not using its patent portfolio in a counter-action against IBM or
discontinuing its use of D/BUS.

	Now, the issue here is not whether such a strategy against software
patents is generally a good idea, but whether it is a good idea to
potentially limit D/BUS's adoption by using such a licensing strategy.

	The "ongoing" part of Michael's statement refers to the fact that as
more and more people contribute to D/BUS, it becomes more difficult to
change the license and resolve this worry.

Cheers,
Mark.

[1] - I thought D/BUS was using the 2.0 version now, but some of the
headers still refer to the 1.2 version.