Are implicit dual-licensing agreements inherently anti-open?

[Gregor Richards]

On Tue, 12 Jul 2005 21:22:39 -0700, "David Barrett"
<dbarrett at quinthar.com> said:
> I guess my big question is:
> 
> Is it fundamentally "not open" to license source code in such a fashion 
> that the initial developer has the additional right to relicense the 
> code and all modifications under any license of his choosing, at any 
> future date?
> 
> We've determined that it "is open" to do this in two stages: by first 
> releasing under any license (such as the GPL) and then executing 
> separate "contributor agreements" on an individual basis.
> 
> But my question is if combining these two steps into a single license 
> somehow "crosses the line" between open-ness and non-open-ness.  If you 
> believe it does, then which of the OSI principles do you think such a 
> license would violate?  Based on my reading, I see no conflict with the 
> two relevant clauses:
> 
> http://www.opensource.org/docs/definition.php
> -----------------------------------------------------------------------
> 3. Derived Works
> The license must allow modifications and derived works, and must allow 
> them to be distributed under the same terms as the license of the 
> original software.
> 
> 7. Distribution of License
> The rights attached to the program must apply to all to whom the program 
> is redistributed without the need for execution of an additional license 
> by those parties.
> -----------------------------------------------------------------------
> 
> I don't dispute, there are "more open" licenses than the one I propose. 
>   And I further acknowledge that some might be unwilling or 
> uncomfortable to submit code under these terms.  But it's not my goal 
> here to convince you otherwise.
> 
> It is my goal, however, to determine if what I want to do is 
> fundamentally incompatible with the principles of the OSI.  Once I 
> determine this, then I can decide on the next step.
> 
> Thanks!
> 
> -david

My opinion on this would be no, and here's why:

Imagine if you released it under the GNU GPL.  If you got any
contributions, now nobody would have the rights to make a proprietary
version.

Now imagine if you released it under the MIT/X11 license.  Regardless of
contributions, everyone would have the right to make a proprietary
version.

Under an "implicit dual-licensing agreement" like you're proposing,
regardless of contributions, only you would have said right.

So, for most people, it would be "more open" (whatever that means) under
your scheme than under the OSI-approved MIT/X11 license.

While a scheme like yours may be considered a bit intellectually
dishonest, the alternative (assigning copyrights) is more likely to lead
to unnecessary forking (I don't want to sign a form just to make a
simple fix, I'll just fork it).  That being said, if your pro{je,du}ct
is at all popular, and you do make it proprietary, I guarantee you it
will be forked (can anybody say "Sourceforge"?)

 - Gregor Richards
-- 
  Gregor Richards
  grichards at ml1.net

-- 
http://www.fastmail.fm - Same, same, but different