Re: Bad license on VCG?
On Sat, 2002-08-31 at 00:54, Nick Phillips wrote:
> On Fri, Aug 30, 2002 at 10:31:52PM -0500, Jeff Licquia wrote:
> > Please point out exactly which section of the GPL would grant us such
> > rights. Remember, rights not explicitly granted are withheld under
> > default copyright law.
>
> 1. You may copy and distribute verbatim copies of the Program's
> source code as you receive it,
>
> See, we're fine.
>
> 2. You may modify your copy or copies of the Program or any portion
> of it,
>
> Also fine.
Under the license as you quote it, yes. Under the license in reality,
no.
Debian requires the right to distribute binary executable versions of
these programs, which are only granted in section 3. This section
requires that "the complete corresponding machine-readable source code"
be distributed.
> As I said, I think all those who are saying otherwise are guilty of
> confusing what we're allowed to do with what we want to do.
Or, possibly, you're not seeing a crucial aspect of the whole debate.
That's my opinion, anyway.
> Look again at the GPL chunks I quoted, paying particular attention to
> 'as you receive it' and 'your copy'.
I see no such words in section 3 of the GPL.
> > Well, the author's stated definition of "source" is "the preferred form
> > for modifying a work". Since the .o (or the obfuscated C) isn't the
> > preferred form for modifying the work, we cannot fulfill the conditions
> > for distribution, since we cannot provide the preferred form for
> > modifying those C files.
>
> Again you are bringing into the equation something which as far as we are
> concerned does not exist (the original commented unobfuscated source).
I don't recall anything in the definition in the GPL about "preferred
from someone's point of view". The unobfuscated source is the preferred
form for modification, whether we have access to it or not. Certainly
everyone involved in this debate would much rather modify the
unobfuscated source given a choice.
> > The author does not consider the obfuscated C files to be source. How
> > do I know? I read the author's license statement, which reads:
> >
> > "The source code for a work means the preferred form of the work for
> > making modifications to it."
>
> And given the package with which we have been provided, that is the obfuscated
> C.
Please point out where in the GPL "the package we have been provided" is
allowed as a factor for determining "the preferred form for modification
of the work".
> > > Quite. But in this case there is no such obligation on the author. The only
> > > question is what *we* could and should do with it.
> >
> > I notice you clipped the "no foul" part, where I agree with you. The
> > author is not obligated to do anything. Are you trying to manufacture a
> > debate where none exists?
>
> Um, it wasn't me who started claiming that we *can't* distribute this package.
Read my sentence again, and see if you can figure out what I was trying
to say.
> > I also note that you agree with me on the definition of "source" when a
> > third party's rights are involved.
>
> The definition of source is "the preferred form of the work for making
> modifications", selected from those forms which are available to you.
Please show me where in the GPL this definition appears, specifically
the part about "those forms which are available to you".
> > > What I'm claiming is that he didn't intend to do something that he knew he
> > > had no power to do in the first place.
> >
> > Which is what?
>
> Force authors to reveal anything, at all.
As I have said now for the third time, the author is never forced to
reveal anything at all, or to do anything else, for that matter. Are
you confused about this?
> You may be confused...
Indeed I am; that's the point. So far, it would seem that the rest of
the Project shares my confusion.
> ...so feel free to ask the
> author what your rights are (hell, I certainly can't stop you ;). I'm
> pretty sure that they will point out exactly what I have. They do want
> to license under the GPL, but they also want to at least discourage
> modifications to the files in question.
>
> They might like to prevent such modifications, but presumably value the
> ability to distribute under the GPL slightly more highly.
No; they wish to have their cake and eat it too. They want to publish
source code and build a community, but they also want to withhold parts
of the code.
I consider such behavior to be deceptive. They could have simply
compiled the "secret" code into binary modules and provided an exemption
clause for the GPLed code; this would have been more honest and
straightforward. Instead, they are playing semantic games and trying to
confuse the issue.
> > Now, this whole mess can be straightened out if the author provides a
> > license that allows the obfuscated C source to be used. They could
> > provide, for example, a GPL exception statement to that effect. One
> > could claim that such a statement is implicit; one would, I think, have
> > to be a lawyer to make such a claim, however.
>
> It's not a mess; there are just several over-zealous people here who are
> seeing what they would like to see (a legal reason why we can't distribute
> this) rather than what is actually before them (a potentially tricky question
> which needs to be answered by Debian developers as a whole, not just those
> who participate here).
>
> This, whilst understandable, is not desirable.
In other words, "people who don't agree with me are over-zealous and
blinded; I certainly can't be wrong".
I have patiently attempted to explain our position, and addressed your
objections without resorting to prejudice. It's too bad that you seem
to need to resort to insult to try to prove your point.
> > If you don't believe me, think for a minute about why this code couldn't
> > be linked to third-party GPLed code (which you've already admitted) and
> > distributed.
>
> By us, as an extra modification, it could. By the original author (which is
> what we previously discussed), it couldn't.
So let me get this straight: You believe that we, as third parties,
actually have *more rights* to the author's code than the author has???
Before I believe such an amazing claim, I will have to see some
evidence, and boatloads of it.
> In this particular case, given the material which the author has distributed,
> it is crystal-clear what they want. Whether or not we go ahead and do what
> they want (i.e. distribute it) is the question, and should be answered
> alsewhere.
You assert this repeatedly in the face of opposition, but you provide no
proof; indeed, you look down on those who do not share your
enlightenment. Please prove this logically, or go away. The policy I
am advocating has been a policy Debian has followed for a long time, and
for good reason; if you wish to overturn it, you will have to do a much
better job of arguing your point than you have so far.
Reply to: