Re: The QPL

[Brian Ristuccia <brianr@osiris.978.org>]

On Mon, Mar 29, 1999 at 11:28:24PM -0500, Peter S Galbraith wrote:
> 
> Brian Ristuccia wrote:
> 
> > On Mon, Mar 29, 1999 at 01:08:19PM -0500, Peter S Galbraith wrote:
> > > 
> > > He was almost ready to use the GPL, but I pointed out that once
> > > people send in patches his work is no longer his alone.  He can't
> > > turn around, modify the work and sell it without hunting through
> > > for all patches and re-writing them.
> > > 
> > 
> > Nothing in the GPL prevents an author from selling copies of his work, even
> > when the work is combined with substantial patches or other software under
> > the GPL. 
> 
> Nothing prevents him from selling copies _under the GPL_.  But he
> cannot use the patches in a parallel project which is sold under
> a different license.  Right?
> 

If the patches were substantial and were licensed to the original author by
the contributor under the GPL then the author is restricted to distributing
copies under that license (unless he makes a different agreement with the
contributor to purchase his patch or whatever). This still doesn't mean he
can't sell the software, just that he can't change the license on code he
didn't write. 

Minor patches like "change the > at line 31 to >= to fix a border condition
bug" are not considered substantial. 

> Actually, when I re-read the QPL, perhaps it doesn't allow that either. 
> The QPL seems to allow the upstream author to sell the
> _same_ software under a different license, but perhaps not
> incorporate the patches into something else.
> 

This depends on the license that the contributor puts their patch under. The
last draft I saw of the QPL did not require the patch to be licensed under
any particular terms. In fact, I think it might be possible for someone to
release a patch for a QPL'd program or library under a totally incompatible
license just to make things difficult. 

> Maybe the GPL is sufficient in this case, since the QPL may not
> offer the protection I initially thought it did.  I guess the
> author of gri has to evaluate the likehood of very cool patches
> being contributed, and the likelihood that he would desire to
> incorporate these patches into his potential-commercial parallel
> project.

In this case, he should consider buying the patches from their authors, or
omitting them from the commercial version.

> 
> Thanks to all who posted (and are pehaps still posted) for your
> enlightening comments.
> 

No problem. IANAL. 

-- 
Brian Ristuccia
brianr@osiris.978.org
bristucc@baynetworks.com
bristucc@cs.uml.edu