"The creators of the GPL"? You mean the FSF? If you don't think Linus and the kernel devs have standing, what standing or authority do the FSF and Wikipedia have? Neither are a licensor of the Linux kernel either.
I'm saying the Linux kernel devs have seen fit to give their view, and they think it carries weight, and seems to govern the actual practice and use, and this would seem to be another analogous instance where they could give their view again.
> You are the one claiming that Linus reasoning implies the AFS code combined with the Linux kernel _can be distributed_. Linux is not actually saying that in the post you quoted. I am asking for where he says that.
Linus doesn't explicitly say this in that post. I thought that was clear enough from ... reading the linked post. However, I do feel it naturally follows from what he says. Otherwise, this discussion would have an angels on the head of a pin quality (which we're talking about the GPL on the internet, so...).
Perhaps we simply disagree about the underlying copyright jurisprudence? After all, you might say, the GPLv2 states: "But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License,..." and "a 'work based on the Program' means either the Program or any derivative work under copyright law". Maybe, you say, ZFS is a "derived work" under copyright law, if not a "derived work" given the plain meaning of derived work?
I happen to believe that courts would not be sympathetic to broad claims of what is "derived" software/infringement, like those made by the FSF, and have not been since Computer Associates v. Altai (and a long line of other cases). For example, you might see Sega v. Accolade, where a court of appeals held that Accolade could reverse engineer Sega’s video game system to create games that ran on the system, even though it involved copying Sega’s code. And I happen to believe the FSF has misled the FOSS community as to the state of the underlying copyright law with their (mis)interpretations of their licenses.