Re: Termination clauses, was: Choice of venue
Matthew Garrett <mjg59@srcf.ucam.org> writes:
> On Wed, Jul 14, 2004 at 12:03:40PM -0400, Brian Thomas Sniffen wrote:
>> Matthew Garrett <mgarrett@chiark.greenend.org.uk> writes:
>> > Why not? Again, what practical difference does it make to our users?
>>
>> Right now, not much -- but it makes it harder for us to mistake
>> non-free licenses for free ones. The patent situation sucks, but to
>> take it as precedent would lead to there being no free software.
>
> So we declare it non-free despite the fact that it makes no difference
> to our users? Does this not sound a little ridiculous?
I just explained to you how it makes a difference, as did several
others. The patent situation is unavoidable and, ideally, temporary.
The termination clause situation is avoidable. Also, it imposes a
burden on our users -- each user now needs to check
/usr/share/doc/*/copyright for termination clauses, then check to make
sure his license to each of those products has not been terminated.
>> > Yes. Why does the test say this? Which aspect of the DFSG is represented
>> > by this test?
>>
>> The DFSG give some guidelines for how to identify Free Software. They
>> are not an exhaustive definition.
>
> The Debian social contract states:
>
> "We provide the guidelines that we use to determine if a work is "free"
> in the document entitled "The Debian Free Software Guidelines". We
> promise that the Debian system and all its components will be free
> according to these guidelines."
>
> If you can't relate these tests to the guidelines, they are effectively
> guidelines themselves. If they're guidelines, then they should be in the
> DFSG. There's a defined process for changing the DFSG - you propose a GR
> and you convince the developers to vote for it. Effectively extending
> the DFSG without developer agreement is not acceptable behaviour.
The DFSG would need amendment to *allow* some sorts of software -- to
declare things free which would otherwise be obviously non-free. They
don't need to change in order to forbid something already clearly
non-free.
Now, it's clear to me that the QPL, with its requirements for
communication with the initial author, its choice-of-venue clause, its
asymmetry and provision for the initial author to take others changes
and sell them as his proprietary works, and its eternal burden of
managing a log of all changes ever made to QPL'd software is non-free.
It appears clear to you that it is free, despite the fact that it
imposes costs equal to the worth of the modifications on distribution
of modification, infinite cost (epsilon storage cost * O(n) tracking
cost * infinite time) on modification, and a cost on acceptance
(travel at the author's whim to Norway, France, or wherever some
mutant version of the QPL sends you).
-Brian
--
Brian Sniffen bts@alum.mit.edu
Reply to: