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

Re: flowc license

Glenn Maynard <glenn@zewt.org> wrote:

> I can deal with the line of reasoning that says "the 4-clause BSD
> license would be non-free, because forbidding anyone mentioning the
> software in banner ads, etc. is insane, but due to its widespread use,
> an exception was made for this license".  (It means that there should
> be no fundamental issue with removing it from DFSG#10 with a GR some
> day, once the license is no longer so prevelent as it was when the
> DFSG was drafted.)

There's no evidence whatsoever that that's what was intended. Checking
the discussion, nobody even seemed to question whether the BSD license
would be considered DFSG-free or not.

> I do have problems with "forbidding mention of the software in banner
> ads is horrible, but we have to consider the restriction "free" because
> it's part of a license in DFSG#10, which trumps all else (and therefore
> we must allow restrictions similar to it, such as an explicit "don't
> mention the product in banner ads"--a trivial extrapolation from the
> OAC, by my understanding of it.)

That's not the way the reasoning should work. The DFSG describe certain
conditions that software has to meet. DFSG 10 makes it clear (though
indirectly) that the 4-clause BSD license was meant to be a free license
- it's an example, not a prescriptive comment. The authors of the DFSG,
and those who voted to accept them (along with the social contract)
intended it to be clear that the 4-clause BSD license was free.

Things have changed since then, and we've tightened up in various ways
as people came up with new issues with existing licenses. But I don't
think there's any way that anyone can claim that 4-clause BSD has issues
that people didn't know about at the time of the writing of the DFSG.

> Of course, the question of whether DFSG#10 is a grandfather clause (the
> former) or an interpretation guideline (the latter) is a long-standing
> one which will probably never be resolved.  (I don't think it was
> initially *intended* as either, since--by my far-from-first-hand
> understanding--it wasn't even intended to be a clause.)

The archives of debian-private are instructive. It's a shame that the
discussion didn't take place more openly.
My point is pretty much this - it's possible to claim that Debian's
standards of freedom aren't changing, merely our understanding of
licenses. That's fine, and a logical consequence of that is that some
licenses may be decreed non-free. But trying to extend that argument to
4-clause BSD is a reach. It was an old and well-established license at
the time of the writing of the social contract.

I agree that considering the 4-clause BSD license non-free nowadays
probably wouldn't lose us anything of any great significance, and I
agree that it's a pain to deal with. But any attempt to make the
4-clause BSD license non-free is an unambiguous redefinition of Debian's
standards of freedom, not a mere reclassification of a license due to a
new discovery.

Matthew Garrett | mjg59-chiark.mail.debian.legal@srcf.ucam.org

Reply to: