[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: The draft Position statement on the GFDL

Raul Miller <moth@debian.org> writes:

> My copy of the DFSG does not say "The license must allow all
> modifications and derived works, ..."

True, though it's hard to argue that was not actually the intended
meaning of what was written, I think.  But granted, it is possible.

> The issue I've been addressing is the distinction between what the
> DFSG actually says, and how it is interpreted.

That's a feature, not a bug.  Or, perhaps I should say that the fact
that we're explicit about that is a feature rather than a bug, since
that's a characteristic fundamental to language and not really unique to
the DFSG.  We just acknowledge the fact and work with it, rather than
papering over it with a lot of pointless verbiage.

A similar issue (the distinction between guidelines like the DFSG and a
definition like the OSD, and that d-l believes that a definition of Free
just Doesn't Work(tm)) has been discussed in quite a lot of detail.

The interpretive process is part and parcel of the DFSG.  We have a sort
of case history (though only lately being formalized), various
principles that we apply, etc.  The Debian Free Software Guidelines are
a fluid and flexible process for determining the freeness of a work.
The document with the same title describes and directs that process, but
isn't itself the process.

> What it actually says isn't enough for our purposes -- you could say
> it's too tolerant of licensing problems.  Unfortunately, the way that
> we express how it's interpreted is also inadequate -- what we say we
> do is actually less tolerant than what we actually implement.  There's
> too many corner cases.

If a particular corner case is significant enough or causes a lot of
controversy, it's probably worth explicitly eliminating via a GR that
modifies the DFSG.  But it's important that we not throw up our hands
and say "Ahh!  Corner case!" whenever we find one, because we'd be
making GR's all the time.  Especially given all the nit-picking we have
here; we'd likely need a GR for every license decision.

Jeremy Hankins <nowan@nowan.org>
PGP fingerprint: 748F 4D16 538E 75D6 8333  9E10 D212 B5ED 37D0 0A03

Reply to: