Subject: Re: Re: Question Regarding GPL
From: Ben Tilly <>
Date: Sat, 21 Jan 2006 02:49:06 -0800

 Sat, 21 Jan 2006 02:49:06 -0800
On 1/20/06, Rick Moen <> wrote:
> Quoting Ben Tilly (
> > On the surface that's that looks like a pretty big jump.  Though when
> > you squint at it sideways, he's at least somewhat consistent.
> [...]
> > But while I see the inconsistency, I'd have to disagree with your
> > characterization that Linus' opinions changed drastically.
> Um... in 1995, he said the kernel module interface is treated as LGPL,
> and thus proprietary modules that use that interface are fine.
> In 2002, he said there's  nothing  licensing-privileged about the kernel
> or any other kernel programming interface, except (per the COPYING
> file's specific provision) function calls.
> That's an about-face.

The interface that he was talking about in 1995 was different than the
one he was talking about in 2002.  The last email at discusses this

: The Linux kernel modules had (a long time ago), a more limited
: interface, and not very many functions were actually exported. So
: five or six years ago, we could believably claim that "if you only use
: these N interfaces that are exported from the standard kernel,
: you've kind of implicitly proven that you do not need the kernel
: infrastructure".

: That was never really documented either (more of a guideline for
: me and others when we looked at the "derived work" issue), and
: as modules were more-and-more used not for external stuff, but
: just for dynamic loading of standard linux modules that were
: distributed as part of the kernel anyway, the "limited interfaces"
: argument is no longer a very good guideline for "derived work".

Given all of the other points which his opinion remained the same on,
I suspect that in 1995 his opinion about the 2002 interface would have
been the same as the one he later held in 2002.  Certainly in 2002 he
indicated that his feeling had not changed about the 1995 interface.

I grant that he did err badly in 1995 in the LGPL comparison.  And I
also grant that the immediate impression upon reading the two emails
is completely opposite.

> But I hope you're not missing the main point, which is that all this is
> an irrelevant sideshow from the main question, which is:  derivative or
> not?  That is disposative.

I'm not.  As far as I'm concerned we are carrying on parallel
conversations.  One is about the consistency of Linus Torvalds'
thinking on this matter.  Another is about whether it is legally
possible to have a proprietary loadable kernel module.  They have
little to do with each other.

> Sorry if it's an unfamiliar concept.  Oh well.

The concept is familiar.  The term was not.