Re: Why licenses don't need to be free (was: Re: Why licenses *are* free)
Hello Manoj and everyone,
I concede that we can't be sure under which circumstances we are allowed to
derive from a copyright license, without major legal assistance. I thnk this
argumentation is obsolete anyway (and more about at the end of this mail).
On Sat, Aug 15, 1998 at 03:10:13PM -0500, Manoj Srivastava wrote:
>
> Marcus> Things are not black and white in the software world,
> Marcus> too. nevertheless Debian has a clear definition, and
> Marcus> everything that fails is in non-free, regrdless of the reason
> Marcus> it fails. I like this.
>
> True. :et us then, have one standrd. Not one for the GPL, and
> one for everything else.
I think we have already agreed that we have to treat different types of data
entities different. And I'll show further below why licenses are completely
different from anythjing else.
> >> Sorry. The same reasoning applies to the FSSTND. I know of no
> >> standard out there that would be allowed in main; that should be
> >> enough reason too.
>
> Marcus> Huh? We can stop to ship the FSSTND in main. This is no
> Marcus> reason to stop following the FSSTND. We also try to follow
> Marcus> POSIX, although I've never seen a copy of it.
>
> I think it is a bad decision to stop shipping the FSSTND in main.
Wasn't it your suggestion to make "main" as modificable as possible? Or do
you mean "Debian" here instead main?
> >> We should not flip-flop from a high moral ground and accept
> >> anything pragmatism based on what document we look at.
>
> Marcus> I see the "verbatim" section as a pragmatism.
>
> I think we differ here. There is a difference, in my mind,
> between documents that are non-modifiable (and there are lots of
> categories we have not considered that fit there as well -- opinions,
> stories, magazines, etc) and propreitery software. Verbatim is merely
> recognizing the difference.
Maybe. Still I feel that standards should not part of "verbatim". And you
are right about the other things, we haven't talked about them yet. For many
things, I can see a benefit from the right to derive from it (everything
that is in some way "functional", has somebody a better dictionary than me?
German word is "Sachtexte"). However, I don't want to start the discussion
about them now, in this mail, as I first want to clarify the license
issue.
> >> Ship the GPL in verbatim, which is a part of debian.
>
> Marcus> We have not agreed on this. It was exactly my proposal, to
> Marcus> allow some very special and well defined exceptions in
> Marcus> main. This would make the verbatim section unnecessary from
> Marcus> my point of view.
>
> No. Main should be sources that everyone may modify with
> impunity. I think it is about time we took a stand on midifieability,
> and created a verbatim section.
So you do think that FSSTND should not be part of main, although you say
above "it is a bad decision"?
> >> We can
> >> still ship stuff. As you say later, either we take a stance on non
> >> modifiable documents, or we don't. Putting them in verbatim seems the
> >> best solution practically.
>
> Marcus> And I say we should allow some exceptions, but generally hold
> Marcus> the line.
>
> That is a contradiction in terms.
Again, I think the special characteristics of licenses allow to make the
exception (see below). A further one would be email snippets. Do you want to
make an extra package in "verbatim" for the mail from Linus about the
kernel-headers? I think so small and obvious exception do not warrant the
extra work. However, I think the main difference between us is the type of
documents we would like to see (and not to see) in this verbatim section.
IMO, only very little would fit in there (and the rest in non-free), in your
opinion verbatim would be used for more things. This is also why I think it
is unnecessary, because it would contain only very few things.
> Marcus> Furthermore I say that license documents are a valid
> Marcus> exception.
>
> And I say it is not. Licenses have copyrights, just as
> standards do. Licences can too be put under rename if changed
> clauses.
>
> >> We can say that an immutable document, bundled with software,
> >> does not prevent the inclusion of the software in main. (Software
> >> programs can nver go in verbatim). So, packages can happily include
> >> the GPL with no changes. But a stand alone package, (say, containing
> >> /usr/doc/copyright/GPL), should go in verbatim.
>
> Marcus> This assumes that the verbatim section is part of main,
>
> No, this assumes verbatim is part of Debian.
> free enough for our purpose. See again below.
Sorry for the typo. I sure meant Debian. So you agree that standards should
be free, but you also say that immutable standards are "free enough for our
purpose"? I think I disagree with the latter.
> I can write a standard that kinda does the same as the FSSTND,
> as long as I don't quote verbatim. Same thing. Licenses and
> standards are exactly the same.
This is not true. Stand-alone license documents are exactly the same as
other data entities, but licenses are not written for this purpose, they are
always written to *apply* to a certain software or other work. You say above
"We can say that an immutable document, bundled with software, does not
prevent the inclusion of the software in main." This makes sense. However,
what are you trying to discuss here then? We only ship licenses bundled with
the software, this is the whole purpose of licenses. We are not in the
business creating a legal repository.
Removing the license from the software would be illegal, so we can't do it
(and I wouldn't want to do it). So, we have to ship the license with the
software in the main distribution. We don't have a stand-alone GPL package,
so this whole discussion is non-sense. This makes me wondering what you are
trying to argue. Licenses are at least not a valid reason to warrant a
"verbatim" section.
> Marcus> This will not bring us any further. We should always consider
> Marcus> common sense, too. For example, fair use is always allowed,
> Marcus> it does not have to be stated explicitely in the copyright.
>
> Fair use means only copying a reasonably small portion of the
> document for *personal* use, or as excerpts in something that talks
> about the document.
This is not true. Fair Use is granted for every(?) purpose, as long as you
acknowledge the originator. At least for more than personal use and
secondary literature.
> Fair use does not condone plagiarism.
True enough. However, how many licenses contain standard phrases? Is anyone
able to claim copyright for them?
> No, the license and standrds are not distinct, at least under
> the laws of the united states.
You are speaking about stand-alone licenses. We have to consider solely
licenses that are applied to software and data entities. License documents
are not the data entities we are talking here about, they are "meta" in the
sense that the spoeak about the date. We can't ship the data without the
license, and furthermore the license is the only thing that grants us our
freedom.
We can't strip the license from the software, so either we have to drop the
Debian project or break the law. I'm not in favour for any of these, but I
suggest we leave the licenses in the software and in main were it belongs.
Our freedom to modificate the original source is restricted:
GPL:
"You may copy and distribute verbatim copies of the Program's
source code as you receive it [...] provided that you [...] keep intact
all the notices that refer to this License and to the absence of any warranty;"
Is this bad? No. Has this any influence on our work? No. Does it make sense
to create an extra section for license documents? No, why should we ship
license documents seperately from the software?
Thank you,
Marcus
--
"Rhubarb is no Egyptian god." Debian GNU/Linux finger brinkmd@
Marcus Brinkmann http://www.debian.org master.debian.org
Marcus.Brinkmann@ruhr-uni-bochum.de for public PGP Key
http://homepage.ruhr-uni-bochum.de/Marcus.Brinkmann/ PGP Key ID 36E7CD09
Reply to: