Development of Creative Commons licenses

[Terry Hancock <hancock AT anansispaceworks.com>]

Evan Prodromou wrote:
>  On Mon, 2006-02-10 at 10:10 -0400, Greg London wrote:
> > CC-ShareAlike is intended for community projects that need copyleft
> > protection to prevent a proprietary interest from taking the
> > project private and using that copy to compete against the open
> > project.
[...]
>  For myself, I know that, as the author many Attribution-ShareAlike
>  works and organizer of one of the largest Attribution-ShareAlike
>  projects around, my intention is to ensure that developers of
>  derivative works grant the same rights that they were granted. I'm
>  not afraid of competition if I can re-incorporate any additions back
>  into my work.

This is why this issue needs to be bounced back to Debian Legal. The 
promotion of the parallel distribution idea fails to provide this 
assurance, as Greg has demonstrated with his "DRM Dave" example.  That 
example needs to join the "Dissident" and "Desert Island" tests, because 
it shows why the parallel distribution model violates the DFSG.

What has happened here is that the analogy of TPM/non-TPM to 
binary/source has been pushed too far -- past fundamental breakages, in 
fact.  I challenged Greg on precisely those grounds: that he needs to 
show how the analogy is broken if he wants to make his case.

IMHO, he did so.  I now see that the analogy is broken in at least two ways:

1) It's law, not code that makes DRM unbreakable, and that, in 
particular, denies the user the right (freedom 1) to make a modified 
version of a received copy and play it (receiving the non-TPM version 
doesn't help, because you still can't apply the TPM to run it).  This is 
the reason in principle why parallel distribution shouldn't be 
permitted. The CC-3 licenses won't allow it, and I believe I've shown 
that GPL-3 won't either.

2) Unlike binaries, which must be compiled from source in a complex and 
error-prone process which makes it very inconvenient for end-users to 
have to work from source distribution only, TPM is completely 
automatable, and has no complex system of dependencies: there's just an 
original work and a couple of keys.  Thus, a TPM-only platform can be 
supported by automatic application of TPM if tools are provided to apply 
it (this has the further advantage that it discourages -- though it 
unfortunately doesn't actually prohibit -- the TPM-platform-owner from 
creating TPM-based monopoly of free content works on their platform).  
The principle point of this breakage is that it kiboshes (to use your 
word) the argument that end users "need" to have DRM distribution.

>  There are a number of projects that offer software implementing
>  patented techniques under the GPL. Yahoo! DomainKeys comes
>  immediately to mind:

This is a red-herring.  If the patents were enforced against users, the 
"liberty or death" part comes into play in GPLv2, and of course, GPLv3 
has quite specific language against this case.

Cheers,
Terry

--
Terry Hancock (hancock AT AnansiSpaceworks.com)
Anansi Spaceworks http://www.AnansiSpaceworks.com