Date: 22 Dec 2001 14:40:55 -0800 From: swear@blarg.net (Gary W. Swearingen) To: chat@FreeBSD.ORG Subject: Re: GPL nonsense: time to stop Message-ID: <bu8zbuvpxk.zbu@localhost.localdomain> In-Reply-To: <20011221150930.A78601@prism.flugsvamp.com> References: <local.mail.freebsd-chat/Pine.LNX.4.43.0112181134500.21473-100000@pilchuck.reedmedia.net> <local.mail.freebsd-chat/20011218110645.A2061@tisys.org> <200112182010.fBIKA9739621@prism.flugsvamp.com> <4.3.2.7.2.20011218180720.00d6e520@localhost> <20011219091631.Q377@prism.flugsvamp.com> <0en10ey5jo.10e@localhost.localdomain> <20011219215548.D76354@prism.flugsvamp.com> <lpellpwlhe.llp@localhost.localdomain> <20011220171739.J26326@prism.flugsvamp.com> <4.3.2.7.2.20011221131016.00d3dcc0@localhost> <20011221150930.A78601@prism.flugsvamp.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Jonathan Lemon <jlemon@flugsvamp.com> writes:
> No, Brett, that's not what I said. I said that the *ORIGINAL* work
> by *itself* is not and cannot be tainted by the GPL. Yes, I agree with
> you here that the Linux syslogd (a derivative of Allman's original
> work) is tainted (by some definition of taint) by the GPL. But the
> original pre-GPL work, which probably still is available in an archive
> somewhere, is not GPL'd.
>
> Gary does not seem to understand this.
I think Jonathan was correct when he said a few posts back "... any
further logic would be a waste of my time". :-) And I'll take his
generous offer to have the last word with him.
I don't know the history of syslogd, but if a collective work is
formed of a GPL'd part and anything else, then the GPL requires the
entire work to be GPL'd. No? If the entire work is GPL'd each and
every part of it is GPL'd. No? (At least. Parts could also be
dual-licensed.) If a part of it was originally only under BSDL then
the only way that part can become GPL'd is if the owner of that part
GPLs it. The deriver may not change the licensing for the owner of
the copyright on that part. Berne Convention, and all that stuff
Jonathan was explaining to us.
Again, he errs in thinking that copyrights and free software licenses
pertain to copies of anything. They only apply to the work themselves
and all of their parts (excepting fair use). It matters not whether one
get the work or a part of the work in a full, pristine (what he calls
ORIGINAL) copy or buried deep in the thirteen derivative of it under
twelve different BSDL-compatible licenses. The part still belongs to
the original owner and only he may say how it is licensed, but that
BSDL, BSDL+GPL, or GPL. And, if a compiler is incorporating third-party
GPL (and thus can't use a modified GPL), then all parts of if must be
under GPL and only by the action of all owners of all parts putting
their work under the GPL. And if they put a work under the GPL for
the deriver, they've put it under the GPL for the public, which means
that a holder of an original, pristine copy of the work now has the
work under BSDL and/or GPL.
Jonathan previously wrote:
> On Thu, Dec 20, 2001 at 06:55:19PM -0800, Gary W. Swearingen wrote:
> > Jonathan Lemon <jlemon@flugsvamp.com> writes:
> >
> > > Suppose I write a piece of software, and release it under the BSD
> > > license. It gets committed to the FreeBSD kernel. Someone else
> > > tweaks it and slaps a GPL license on it. You would have all copies
> > > instetaneously tainted by the GPL.
> >
> > This belies your claim to have understood me. You're scenario there
> > is not something that may legally occur. Let's call your work "A"
> > and the tweaker's work in a derivative of A, "B". You own A; nobody but
> > you has the right to slap a GPL license on it. The tweaker is allowed
> > by the BSDL to make a derivative of A. This is a work of authorship
> > which we'll call "C". It contains A and B. The tweaker is free to
> > distribte C because of the BSDL, but he may not slap a GPL on C, because
> > it is not all his to license.
>
> Well, if I am to believe you here, then it means that you have just
> proved that there is no worry about the GPL at all, since in your own
> words, "he may not slap a GPL on C".
>
> "C" being the FreeBSD kernel of course, which, when you get down to
> it, is simply a collection of works "A", which is all BSD licensed.
>
> Sorry, you just argued yourself into a corner here.
That's only a problem when the argument needs somewhere to go. I'm
comfortable with my position until someone shows me what's wrong with
it other than its being unconventional or having undesirable
consequences.
I have not "proved that there is no worry about the GPL". I've proved
that people err when they claim to apply the GPL to a work which
includes non-GPL parts without getting the owner of those parts to
license them under the GPL.
These false claimers of GPL don't infringe on the BLSLers; the BSDL
allows redistribution in a derivative. They infringe on the GPL'd work
they are incorporating. If the infringer is FreeBSD, I'd worry.
There's also the worry that the acquiescence to the use of the GPL will
later be regarded as an implied license.
Maybe the worries are negligable. Maybe we're willing to risk that
nobody will care that FreeBSD licensing is infringing on them. (I don't
say that it does, because I've already forgotten the relevant facts.)
> Either the someone can slap a GPL on the BSD code, in which this thread
> has relevancy, or they cannot, in which case the whole thing is moot.
Anone can slap a GPL on anything. The fact that so many think they may
do so, thus changing the licensing of someone else's work, without
obtaining the permission of the non-GPL owners, is a problem.
Side note: It may also be considered a problem with the GPL's language.
It could be fixed and still retain most of it's copyleftist power.
> > But if the derivation is done and it is distributed with your
> > cooperatation and you say things like "the kernel is under the GPL",
>
> But we (me, greg, RMS, other posters) are _NOT_ saying that, *you* are.
> It has been our position from square one, that the BSD kernel is not
> under the GPL. Don't try and put your words in our camp.
Jonathan, Greg, RMS, and others have said that repeatedly in the context
of this discussion about a GPL-infected ["BSD"] kernel. We'd have no
discussion if it wasn't said. Methinks Jonathan isn't paying attention.
Firstly, here's Jonathan's words from this same thread:
E.g.: kernel + (N)"options XXX" = non-GPL'd kernel.
kernel + (N)"options XXX" + "options EXT2FS" = GPL'd kernel.
That's what I'm discussing; a "GPL'd kernel" (Jonathan's words).
Whether FreeBSD people are distributing a derivative of a GPL'd work is
not at issue in the sub-thread started with the just above scenario, as
far as I'm concerned.
I responded (in more/other words): if you GPL the parts to satisfy the
GPL in his second equation, you've caused the first to also be GPL'd.
Socondly, consider the third-level quote of Jonathan above. He continues
the scenario with his supposition about a GPL'd FreeBSD kernel.
Thirdly, remember that almost a week ago when this whole thing started,
Greg and RMS were discussing a FreeBSD kernel with GPL code in it when
Greg asked about it and RMS said:
If you link some GPL-covered code into the kernel, the GPL's
conditions will apply to the kernel as a whole.
(I snipped the part where he erroneously implied that it doesn't apply
to all the parts.)
We aren't discussing a GPL-free FreeBSD kernel or even whether it is free.
> I believe I'll end this thread here, since you don't appear to be able
> to mount a cohesive argument, and any further logic would be a waste
> of my time. Feel free to have the last word. :-)
I believe that it only appears that way to him. I suppose that many
people are not able to make sense of my quickly composed and wordy
arguments, but there is fault on both sides. I wish Jonathan had
addressed my arguments in more detail. I've learned in the past from both
forming my arguments and having them soundly criticized. Sometimes one
can even learn from responding to criticism which is nothing but sound.
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-chat" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bu8zbuvpxk.zbu>