> > Any discussion of license issues __has__ to take this scenario into
> > account, or it's simply not addressing the points at hand.
> 
> Cetainly.  I think the main problem we are seeing here is that there
> was not adequate discussion of this issue before we found ourselves
> committed to a course of action.

But, the turning point down the wrong course of action happened at the
start of the project, not recently.

The problematic decision wasn't the one to allow the change in my
license terms, that's just what ended up bringing the matter up for
public discussion.  To any large commercial organization, it should be
obvious why my license doesn't really make a difference; either the
organization is going to do their best to follow the licenses (in
which case they're going to suffer as-is), or they're not going to
bother (in which case they're legally in the wrong and can be pursued
about it).

The problematic decision _was_ the one, at the beginning of the
project, to allow individuals, rather than the organization, to hold
copyright on things in the source tree.  Of course, since the project
started from other individual-copyrighted source (386BSD + the patch
kit, really), the decision was, at least to a minimal degree, out of
our hands from the beginning.  However, we could have made the
decision to allow no _new_ individual-copyrighted source, then.

It may be time to reverse that decision; _that_ is not my choice.
However, even if that decision is reversed, it'll still be almost
impossible to rid the tree of invidual-copyrighted source...



chris