Re: [Gnu-arch-users] contributors' licensing conditions

[James Blackwell]

As it happens, I'm on one of the gplv3 boards. I'll do what I can to
answer questions. If my answers do not satisfy you, then please visit
gplv3.fsf.org/comments/ and bring up your points (its really easy to do;
select the text with your mouse, and type 'c').

I took out your question on propagation. I think thats pretty clear, that
propagation essentially means making copies of software.

> Philosophically, the section on interpretation vis-a-vis DRM is pretty
> disheartening.  In fact it says Thou Shalt Not Use Free Software for
> Privacy Protection, because what else is privacy protection but
> white-hat DRM?  Any system that can be used to protect personal
> privacy can be adapted to DRM.  Do we have to remove login and chmod
> from the GNU system?  ("Wheel bit" deja vu.)

The intended meaning of the DRM section is that if DRM is used to verify
the validity of software, that the tools are also made available to
perform the signing. 

My favorite example is the Tivo. The Tivo relies heavily on free software
(its a GNU/Linux system). As such, one can damand (and will get) the
source in preferred form for the Tivo. This doesn't mean anything under
gplv2 though because the binaries are DRM signed and replacing their
authorized version with your unsigned version will result in a tivo which
will not boot.

With the DRM clause in the GPL, Tivo will either have to forgo gplv3
licensed software or provide free software which allows anybody to replace
the signed-by-them free software on their machines with signed-by-you
software on their machines.
> 
> More fundamentally, this retracts a core principle of the GPL up to
> v2: it doesn't tell you what to do with the software, only the license
> terms under which you may redistribute it.  This violates Freedom 0,
> which is explicitly affirmed in dGPLv3[2] for the first time.

There is a section that grants you rights to use and modify the software
if you don't start a patent suit with others over the same program. Is
this what you mean?

> The section on DRM also says that "covered software is not part of a
> system" to deny users who can _run_ a program their other rights.
> Does this close the "ASP loophole" for systems where the webserver
> program is dGPLv3?  Does this go further, and require that any
> dGPLv3ed server run via inetd have source code available, so that the
> ASP loophole is available only to standalone daemons?  How does it

No. This is not the case. Output from a program, say one running on a
webserver, is not covered by the GPL.  To propagate a work, one must copy
and distribute it to a seperate party. 

> affect a company's ability to deploy covered software internally
> without thereby giving employees all dGPLv3 rights, including
> redistribution of the software and licenses on any patents?  And how

This is identical to GPLv2. You'll either have to make it bindingly clear 
that software used by them is not being propagated to them. This is more 
clear if you consider license agreements in a for-pay context: If I buy a
license to use propreitary software and have my user use it, I have not
transferred the license to them.

> about the user's side: if I can run but not copy a dGPLv3ed program
> using the SSH protocol, does that mean I'm not allowed to use lsh
> because lsh may not be part of a "system of DRM"?

No. ssh is a different work. Using a program to use another program has
nothing to do with the GPL at all. 

The point to the paragraph you're reading is that the software that is
distributed under the GPLv3 does not contain digitial rights management
and that it must include anything that is required to access the data
managed by the software. In simpler terms: 

  1. If you give me a program that can read and write a dataset
  2. And I change the program you give me
  3. I can still read and write the dataset.

This comes into play, again, with embedded systems in which if you change
the program, you can no longer access the data.

This has nothing to do with system management or authentication! Promise!

> I'm sure that all of these questions can be answered, more or less
> satisfactorily.  But given that they haven't been answered yet, and
> will presumably require changes in the language to be aswered
> satisfactorily, it is _way_ premature to consider using this refugee
> from _Fantasy & Science Fiction_ for any free software.[3]

They were in Boston a few days ago. There will be a FAQ that explains in
clearer terms. We're also trying to answer questions at
http://gplv3.fsf.org. 

> Socially, this is clearly going to be very divisive.  Andrew Suffield
> is the obvious person to comment, but I would worry about Debian's
> reaction if I were you.  On that ground as well I think it's best to wait.

Most of the Debian people that I saw in Boston seemed happy to me. The
people that weren't happy were the large companies present that have been
behaving in a rather naughty way.

That all said, the document still needs quite a bit of work. There are at
least three loopholes in the document, parts of it aren't as clear as
could be, there are grammar errors that need fixing. Its certainly going
to see changes over the next year or two prior to the final version. 

You can help fixi t up by hitting the comments page and selecting the
parts that you think have problems and clearly explaining what the
problems are.

-- 
James Blackwell's home :  http://jblack.linuxguru.net
Gnupg 06357400   F-print AAE4 8C76 58DA 5902 761D  247A 8A55 DA73 0635 7400

signature.asc
Description: Digital signature