On Wed, Nov 07, 2001 at 09:52:41PM -0700, Richard Stallman wrote:
> > I think "scour" is too strong a word. The invariant sections have to
> > be listed in the notice that says the work is under the GFDL.
>
> Yes, but again, you're relying on the honor system and hoping authors
> will be principled.
>
> That is always true, for any license. I explained that in my previous
> message, and you responded "point taken". So I am surprised you are
> now once again presenting this as unique to the GFDL.
No, I said "point taken" when you demonstrated that Debian developers
should be dilligent in scrutizing a GPL'ed work for the existence of
code that *isn't* under the GPL within it. If I'm looking around in a
source tree and I find something with a copyright notice and a different
license, I know what's going on. The work isn't entirely under the GNU
GPL.
That is not the same thing as having a work that is purportedly under
the GNU FDL, and contains no evidence of any other copyrights or license
terms, but which has critical material marked as Invariant Sections, or
fails to properly identify Invariant Sections or Cover Texts.
If I do detect the existence of such things, what am I to do? What is
their licensing? Do they indeed remain invariant? Is the license
nullified because the author failed to apply it properly? Recall that
(unfortunately) legal tradition holds that errors in licensing are
resolved in favor of the author, and restricting the permission of
others to copy, modify, or redistribute a work.
This is altogether a different sort of problem from the one you pointed
out, and *is*, in my opinion, a characteristic of the GNU FDL that the
GNU GPL does not address, since the GNU GPL does not place these sorts
of restrictions ("identify this part of the work as thus-and-so") on the
copyright holder.
> The restrictions that come into force under the GNU FDL in the "Copying
> in Quantity" section, and the restrictions that are always in force for
> "standard (paper) book form" under the OPL look very, very similar to
> me.
>
> The difference is that the GPL permits publication of modified
> versions, even on paper, with certain labeling requirements. The OPL,
> if either of the options is used, does not permit this. That is the
> fundamental difference.
Usage of the OPL with either of the options, frankly, does not interest
me since it flatly and uncontroversially fails the DFSG when this is the
case. So, for convenience, feel free to pretend in our discussions that
those options do not exist when discussing the OPL.
In that form, I think the OPL stands as a viable competitor to the GNU
FDL.
> Any copyright holder can give permission to disregard this requirement
> if he wishes to. I don't see a reason why they should do so, but if
> you convince them to do so, they can do it without any change in the
> GFDL itself.
In the same manner as the "linking-with-Qt-permitted" riders that used
to be added to GPL'ed works before Trolltech dual-licensed the library.
Okay. Thanks for this clarification.
--
G. Branden Robinson | The first thing the communists do
Debian GNU/Linux | when they take over a country is to
branden@debian.org | outlaw cockfighting.
http://people.debian.org/~branden/ | -- Oklahoma State Senator John Monks
Attachment:
pgpgthz5LtMJP.pgp
Description: PGP signature