[cc-licenses] Subject: Re: Version 3.0 - List Discussion Responses

Terry Hancock hancock at anansispaceworks.com
Tue Sep 26 21:09:19 EDT 2006


rob at robmyers.org wrote:
>  Quoting Terry Hancock <hancock at anansispaceworks.com>:
>  GPL-3 does not allow code licensed under it to be considered part of
>  a TPM.

True, and really important for the world of free software, but totally 
irrelevant to this thread.

[...]
>  GPL-3 allows far more than parallel distribution. It prevents systems
>  being locked in to DRM, ironising DRM law as earlier versions of the
>  GPL ironised earlier mechanical copyright law.

The "I am not a TPM, I'm a free program!" clause doesn't have anything 
at all to do with whether a TPM is used, it just can't be a TPM.  You 
could, in principle distribute a TPM'd GPLv3 binary and a TPM'd GPLv3 
source distribution.

It's only the definition of source code -- which insists on any 
necessary keys to read the source -- which protects you from that.

 > Whereas Debian want
>  the freedom to allow systems to be locked in to DRM. This means that
>  they want the *opposite* freedom for CC-3.

More precisely, they want the freedom for free packages to be used on 
non-free platforms that are already locked-in to DRM.  Clearly, no pure 
Debian system can have this property (because only a non-free package 
can actually implement the DRM -- although, in principle, that "package" 
might be firmware).

> > So, if no one can refute the analogy between "non-TPM/TPM" and
> > "source/binary" then I think any remaining objection to the
> > parallel distribution idea is nothing more than FUD. Please
> > convince me of the flaw in this analogy, if there is one. What's
> > different?
>
>  Apart from not being able to tackle TPM systems the same way as GPL-3
>  there is no source code in CC-licensed works so there is no natural
>  pre-existing break point for not introducing DRM.

Again, this is irrelevant. We aren't imagining a CC-licensed piece of 
software implementing DRM, but rather a CC-licensed piece of content 
that may be locked by DRM.  Whether or not CC covers the implementation 
of DRM is not an important use case, because we don't recommend CC 
licenses for program software.

 > For CC licensed
>  work, the work is both source and binary. I don't need the non-DRM
>  version of a work to make a new DRM work on my DRM system, whereas I
>  do need the non-DRM source code to make a new DRM version of the DRM
>  system. The flaw, then, is that each new version doesn't need source
>  and so parallel distribution is an artificial requirement.

Not true.  AFAICT, you *do* need the non-DRM version to make a 
*modified* version of the work (which is the only time you need source 
to make a new version of a program, too). For both program and content, 
the obfuscated form may be copied (at least in principle -- but here DRM 
is less viral than a binary, since it may artificially impede your 
copying it, whereas the binary may be freely copied. Hence, the DRM file 
is even less of a threat of out-competing the non-DRM copy than the 
binary is of out-competing the source).

>  Parallel distribution is either defeatable after one generation (if I
>  don't have to parallel distribute derivatives or offer the original
>  to others in the format of their choice) or a burden to downstream
>  users (if I do).

Yes, I vote for the burden, but note carefully: the only users burdened 
are those who use the DRM format. As presently licensed, they have no 
access whatsoever!

So the only "burden" is on the already burdened, and the burden is 
actually lighter than what they had.

Or put another way, it isn't the license doing the burdening, but the 
DRM platform, and the parallel distribution requirement makes it *less* 
burdensome for them, while having no effect whatsoever on anyone else.

 > Unless we put our faith in archive.org, or have a
>  GPL-style clause that requires distributors to offer non-TPM versions
>  by post for three years (a pointlessly short time for culture, and a
>  burden at odds with CC's freewheeling take on culture), parallel
>  distribution cannot have the effects claimed for it.

You appear to be dismissing my preferred case in the previous paragraph. 
But IMHO, you shouldn't have. The parallel distribution "burden" is no 
problem, because only those who have chosen to be burdened need to worry 
about it (you always have the *freedom* to do what the CC license 
currently *obligates* you to do: use a free platform and no DRM at all).

And honestly, IMHO, it's a *good* thing that DRM platforms continue to 
present a burden to their users, because we would like them to die out.

Cheers,
Terry



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




More information about the cc-licenses mailing list