Martin Husemann wrote:
> I don't think there has been a public, official statement yet, but
> TNF considers CDDL to be not suitable for in-tree code.

http://www.netbsd.org/Foundation/
http://www.netbsd.org/Foundation/bylaws.html

With respect, I'm saying TNF is not competent to make such a decision i.e.
it does not have that authority, and we're mistaken to look to it for an
announcement, position, or statement.  

That doesn't mean I think Christos and the others have no voice.  It means
TNF, per se, the legal entity, is not afaik in charge of what's in the
tree and what's not.  That decision is up to the project participants.  

There are no legal consequences to adding something like zfs to the source
tree, because licenses apply to individual source files and resulting
binaries.  NetBSD's own code (TNF's purview) is of course not affected. 
Because the binaries are affected, I can easily understand why TNF in its
license-administration role would decide not to distribute binaries that
included CDDL-licensed code.  That would mean that someone who wanted a
kernel with zfs would have to roll his own, hardly a NetBSD novelty.  

Mark asked where to take this.  I don't know.  If it were up to me, I'd
change the documentation in the way I suggested and wait for someone to
object.  

We should only hope the zfs project moves to such a point that the
licensing question becomes material.  

Regards, 

--jkl