Ed Felten comments:
If somebody violates the CC license on my website, I sue them for copyright infringement. The license exists not to give me a basis for suing them – copyright law does that – but to give them a defense if I sue them for doing something that is permitted by the CC license. If we end up in court, it’s the user of the website, rather than me (the publisher), who will be arguing for the validity of the CC license. And I can’t very well argue that I didn’t properly consent to my own CC license.
First, that depends on the relevant CC license and the relevant term. CC licenses give authors some rights that copyright doesn’t — most importantly, attribution/non-attribution rights. (Also, the commercial/non-commercial distinction maps only imperfectly to the “commercial” dimension of fair use.) If I use a CC-licensed “attribution-only” work (and Freedom to Tinker has an “attribution only” license) and I don’t attribute my copying to the original author, am I liable for copyright infringement? I don’t think so.
Second, as user, have I assented — so am I liable for breach of contract? It takes two to assent. The original publisher assents; does the user assent? Even under “notice and use” clickwrap-style assent models, the user has to be given notice, somewhere, that use means something more than use. The “Some Rights Reserved” CC notice, on its face, doesn’t do that.
Ernie Miller also defends CC in response to my post, noting that the point of CC licenses is to enable authors to give up rights that copyright reserves by default. That’s true up to a point, but (see above) the attribution right that CC covers is an extra-copyright interest. I don’t think that CC licensing can be justified generally on the principle that copyright owners can always give away what they start with.
More generally, Ernie argues (I infer) that copyright law means that you can’t design information goods in ways that defeat statutory rights, so my “DRM-limited-CD” is no different, really, than a book wrapped with a “no fair use” label (my simplifying example, not his). I wish it were so, but I don’t see a place in the statute where what I might call a “design preemption” rule kicks in, treating abusive design like abusive contracts. For labels and contracts and licenses, Section 301 on preemption is the starting point. What is the starting point for abusively designed information objects? We don’t have a good one in copyright law itself, which is why (I think) Lexmark and Chamberlain, for example, come out in the right place but on very shaky reasoning. Antitrust law offers some help, but only in the rare case, like Microsoft, where we can get worked up over market power — and even then, antitrust courts are reluctant to police product design.