[cc-licenses] Attribution license

Terry Hancock hancock at anansispaceworks.com
Mon Oct 2 14:03:19 EDT 2006

Evan Prodromou wrote:
>  So, a side question on the DRM issue: considering that the
>  requirements of the non-copyleft Attribution license don't extend to
>  Derivative Works -- it is the most liberal of Creative Commons
>  licenses -- are people as passionate about maintaining the anti-DRM
>  requirement in that license?

Personally, I think you're missing the point of the disagreement, Evan.

I was with you for awhile there, because the binary/source analogy to 
TPM/non-TPM is reasonably compelling. Considering only the consequences 
of that analogy, the parallel distribution argument appears to make sense.

But Greg London, IMHO, has demonstrated that Debian's analysis is 
*wrong*. To be precise, I mean that the requested change is not only a 
problem for CC, it *should* be regarded as a violation of Debian's own 
guidelines, the DFSG.  This idea needs to get bounced back to Debian 
Legal, because it's a mistake *in principle*.  I'd be more than happy to 
make that case on Debian Legal myself now that I see the problem.

(I wish that Greg could be less confrontational about this point, he 
appears to think you are "misunderstanding on purpose" -- please don't 
let that interfere with understanding his reasoning.  And yes, Greg does 
have some weird economic theories, but this isn't one of them).

Here are some specifics about the breakage of the analogy:

1) The force of DRM is imposed not by mere technological inconvenience, 
but by force of law (the DMCA). It is (in that sense) more comparable to 
the kinds of limitations imposed by software *patents*.  In any case, 
it's not the same problem as with binaries (or any form of mere 
encryption that is *not* legally considered a TPM/DRM).

2) The assumption that requiring end users to apply their own TPM if 
they need it, is onerous, is based on the faulty comparison with the 
process of compiling source to binary (which is an error-prone, complex 
process beyond the capabilities of many end users).  The application of 
TPM can be made completely automated and invisible, if so desired, thus 
the practical requirement of keeping the application of the TPM off of 
the end-user is lifted.

There may be other problems, but these are enough to trash the parallel 
distribution argument.

I have myself shown that GPLv3 will impose exactly the same requirement, 
so if you have a problem with the revised CC-By-SA 3, you will have the 
same problem with GPLv3 (to use your pragmatic/political argument, you 
now have two opinion leaders FSF and CC allied against the Debian 
position, so now Debian's the odd man out).

I say this, because it sounds like you're wanting to back off to a 
position where "By" is admitted to Debian, but "By-SA" is not. But I 
think (as a Debian user myself) that this is the wrong strategy.

I know it's surprising that after so much time, Debian Legal has not 
itself discovered this problem, but that's one thing that can happen 
when you get a radically different group of people to look at the same 
problem (and that's a good thing). At this point, I think it would be 
far wiser to return this back to Debian's court, and let it be reconsidered.

IMHO, Greg's "DRM Dave Test" needs to be added to the "Desert Island 
Test" and the "Dissident Test" as fundamental cases for consideration of 
the DFSG freedom of a license.


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

More information about the cc-licenses mailing list