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

Re: PHP-Nuke License Conclusion?

Scripsit Branden Robinson <branden@debian.org>
> On Tue, May 13, 2003 at 06:46:31AM +0200, Henning Makholm wrote:

> > Digging in the archives turns up that it has not always been you who
> > made the false claim that GPL+more restrictions is necessarily
> > internally inconsistent. I apologize for implying that.

> Calling it a "false claim" is overreaching given that you've never
> actually constructed an argument illustrating its falsity.

I wonder how the arguments I pointed to came into being, then, if I
did not construct them.

> Okay, so you and Nick Phillips will cotton to "the GNU GPL" being a
> variable rather than a constant,


> > I am completely unconvinced by your tacit leap from "most people do
> > this wrong" to "it is impossible to do it right, and Debian always
> > interprets even cases where it is done right as non-distributable".

> This is refutation by assertion.  You haven't addressed my points at
> all.

Your only point seems to be that *sometimes* the description of
such almost-but-not-quite-GPL licensing terms is phrased in unclear
and possibly inconsistent ways. This in no way entails that *every*
set of almost-but-not-quite-GPL licensing terms must be unclear and

>   [I have asserted that the way most people add restrictions to the GNU
>   GPL] -- at least in cases that have come to my attention via the
>   debian-legal list -- results in a license that is internally
>   inconsistent, and thus impossible to satisfy while exercising one's
>   traditional freedoms with a copyleft.

Even here, you seem to argue that because "the way most people add
restrictions" is wrong, and often internally inconsistent - which I do
not deny, mind you - it is impossible to add restrictions.

>   Observation One
>   ---------------
>   The scope of the terms and conditions, a.k.a. restrictions, is clearly
>   defined by the text of the GNU GPL.  Any restrictions beyond that are
>   forbidden by section six.

Section six has absolutely no power over how an individual, sole,
author may or may not chose to license his work.

>   It is obvious to me that "THESE TERMS AND CONDITIONS" is meant to refer

No. On the contrary, it is obvious that if an author knowingly and
explicitly licenses is work under "the conditions of the GNU GPL, with
the following additional restrictions: [bla bla]", then the author
intends every occurence of "THESE TERMS AND CONDITIONS" to refer to
his entire combined license conditions.

>   Observation Two
>   ---------------
>   The FSF claims copyright in the text of the GPL itself, and does not
>   extend permission to anyone to modify it.

Actually, the FSF does permit people to use the legal language in the
GPL as the baseline of other licenses (see your own quote from the GPL
FAQ below). They will not like it, and they will call people who do it
nasty names, but they do not claim the legal power to prevent it.

>   Therefore, making modifications to the area between "TERMS AND
>   AND CONDITIONS" might be considered by the FSF as an infringement of
>   their copyright in the text of the GNU GPL.

Wrong (see above, and below). But even if *making* the modifications
were illegal, it is beyond the powers of the FSF to forbid people to
*imagine* modifications, and use such imagination as a succint way to
describe their own licensing terms.

>   Observation Three
>   -----------------
>   The FSF does not want people "forking" the GPL, and think that license
>   proliferation is a bad thing.

This is true, but it does nothing to argue that a fork would be
internally inconsistent.

> 	Although we will not raise legal objections to your making a
> 	modified license in this way, we hope you will think twice and not
> 	do it.

Note this excerpt from your own FSF quote.

>   However, in practice, most people don't do what the FSF recommends in
>   their FAQ.  They just say "it's under the GPL" and slap extra
>   restrictions somewhere else -- perhaps in a README file.

Which I can easily agree is internally inconsistent and may make the
work undistributable. But this still does not mean that adding
additional restrictions is OF NECESSITY internally inconsistent.

>   It is this practice that I call internally inconsistent,

The words you used were

|| Debian interprets "this License" and "herein" to mean the
|| conditions of the GNU GPL expressed in its text; no more and no
|| less.

which does not seem to me to acknowledge that there is a right way to
add extra restrictions. (Right as in "probably legally sound", not
right as in "free").

>   And, of course, even if one feels that this isn't a
>   problem, there is *still* the issue of incompatibility with all other
>   GPLed works in the world.

I do not deny this. But an incompatible license can still be
*internally* consistent.

>   I therefore think it's a good idea for people forking the GPL to
>   rename the license,

But I still do not think you have offered any convincing arguments
that "GPL with the following additional restrictions: [blah blah]"
is a bad name to rename it to. On the contrary, I think it is the
honest way to go about it, rather than reproducing the entire load
of legalese as "the Foobar Public License" and hiding the
incriminating restrictions in the midst of vaguely-familiar licensing
text that nobody is going to read closely anyway because it just looks
like a revamped GPL anyway.

> To do that, all you have to do is show that the phrase "THESE TERMS
> AND CONDITIONS" in the GNU GPL is not meant to, and does not in
> fact, refer to the text between "TERMS AND CONDITIONS FOR COPYING,

It is clear as day that if the author starts his source files with

/* Copyright YYYY N.N.
 * This file can be copyed freely under the GNU general public license
 * with the added restriction that everyone who distributes it (or
 * derived works) must pet a cat.

then OBVIOUSLY he means for "THESE TERMS AND CONDITIONS" to include
his additional restriction. And it is what the copyright holder means
that counts. It does not matter what those who happened to write the
words that the author chose to refer to to partially explain what he
intends meant, as long as the author's intentions demonstrably

Henning Makholm               "Hele toget raslede imens Sjælland fór forbi."

Reply to: