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

Re: PROPOSED: interpretive guidelines regarding DFSG 3, modifiability, and invariant text



On Tue, Nov 27, 2001 at 08:22:44PM +0100, Henning Makholm wrote:
> Scripsit Branden Robinson <branden@debian.org>
> > On Tue, Nov 27, 2001 at 04:14:24PM +0100, Henning Makholm wrote:
> > > Scripsit Branden Robinson <branden@debian.org>
> 
> > > > it is impossible for an author/copyright holder to
> > > > misapply the GNU GPL to his own work.
> 
> > > /* Copyright (c) 2001 J. Stupid Luser
> > >  *
> > >  * This program is free software and is governed by the GNU public
> > >  * license. It comes with no warranty and must not be sold for profit.
> > >  */
> 
> > Whatever this is, it isn't the GNU GPL.
> 
> It is completely similar to the situation you worry about where an
> author purports to use the GFDL but adds extra, inconsistent,
> restrictions - for example marking a section as "Invariant" when that
> section cannot be "Invariant" at all due to the explicit definitions
> in the GFDL itself.
> 
> If that happens we can just copy your GPLish reasnoning and say
> "whatever this is, it isn't the GNU FDL" and be done with it. No new
> situation here.

You may be right about this.  However, I think it's potentially more
difficult to discern such misapplication in the case of the GNU FDL than
it is in the case of GNU GPL.  Maybe that's because the GNU FDL is a
more complex license, and/or the community simply has more collective
experience with the GNU GPL, because it's been around in its current
form for over 10 years.

> A side note: Having read the GFDL I'm left feeling less than comfortable
> with its treatment of Invariant Sections. Specifically, you're not
> allowed to delete Invariant Sections (of any kind) in derived works.
> This seems to pose a practical obstacle to reusing documentation
> language.
> 
> Scenario. Suppose I'm writing a silly little tool that basically does
> the same thing as xfontsel but allows the user, along the way, to edit
> the sample text. This editing uses emacs-like keybinding (however, I'm
> implementing the actual code from scratch for some convenient
> technical reason that I'll leave it to the reader to think up).

Because you're writing in C and not elisp?  :)

> The manual for my tool will be very small, because I do not really
> know enough X font naming conventions to explan to the user what their
> choices actually mean. But it would be nice to reuse (slightly
> reworked) a couple of pages from the Emacs manual that describe the
> editing keybindings. However, I'm not allowed to do this without
> adding to my manual things such as the GNU Manifesto, which would then
> completely dwarf the actual technical content of the manual, rendering
> it unsuitable for inclusion in Debian!

I think this is an excellent point and deserves a response from the FSF.
Traditionally, Fair Use permits quotation of limited sections for the
purposes of scholarship or review, but 1) you wouldn't be using sections
of the GNU Emacs Manual for those purposes, and 2) if you implemented a
fair number of key bindings, you might be pushing the concept of
"limited sections".  The GNU Emacs documentation is very extensive and
even a extremely small proportion of the total can run to many pages.

> On the other hand, I don't really want the Emacs manual to disappear
> from Debian just because of this. But we might give it a thought
> whether it would be a Good Thing to have a semi-official set of
> guidelines that explicitly bless such obnoxious terms, instead of
> just letting them slip by unnoticed for documentation we Really,
> Really want to have in Debian.

I think we need to be careful how much we bend DFSG 3.  As much as I may
be in sympathy with what I understand to be the FSF's ultimate goals
(radical reform -- or even elimination -- of traditional copyright
protections for software), I have to abmit I'm less than sanguine about
permitting some entities to attach whatever Manifestos *they* like to
documentation they license under the GNU FDL.  I sure would hate to have
to ship 100 megabytes of advertising materials in a documentation .deb
just because it was licensed under the FDL and the advertising was
marked as Invariant.

> Yes, I know, I'm arguing for inconsistency and double standards here.

It is the FSF that has thrust this issue upon us.  I can't blame you
for wanting the GNU Manuals in main.  They are, on balance, excellent (I
have issues with parts of the GNU C Library Reference Manual) and, to my
mind, wholly inoffensive.  What worries me is how we set a precedent to
achieve this desired end that is not discriminatory in favor of GNU
FDL-licensed material from the FSF.  We may have turned a blind eye to
this issue in the past, but the genie's out of the bottle now and we've
got to decide how to face it.

-- 
G. Branden Robinson                |
Debian GNU/Linux                   |     Music is the brandy of the damned.
branden@debian.org                 |     -- George Bernard Shaw
http://people.debian.org/~branden/ |

Attachment: pgpfj4xnKVYeJ.pgp
Description: PGP signature


Reply to: