Sun, 26 Jun 2005

Fascinating how lucky Free Software got, in a lot of ways. We got fairly good before we got popular (so hard these days); and our rock stars were skilled creators of Free Software first and rock stars secondarily- for a long time they led mostly by example. Outside Gilberto Gil and Lessig (who despite being an author does not consider himself an artist), Creative Commons seems to lack creative stars who are creators who happen to use CC, and it seems that this is fairly problematic for them. Contrast with Hal Abelson (who I had a great conversation with last night) who has been with FSF since day one, or Stallman, who as author of emacs had a great deal of cred with other programmers (though arguably he has squandered it.) It is not coincidence, I think, that CC is ragingly successful in Brazil- Gil’s influence there must be huge in ways that can’t be understood in the US. [One wonders if BBC can become this going forward, though their licenses are not CC.]

Got luck in other ways too. AFAIK, GPL was originally intended to be US-only; by the time it was widely used it was global by accident and it was too late for the lawyers to ‘fix’ it by fragmenting it and making it regional. And now, 20 years later, it is in many ways the most global social contract ever- as someone said last night, many smaller governments would now actually consider ‘fixing’ their copyright law if GPL was problematic. And it has IBM and other heavy hitters are behind it. In contrast, the lawyers are trying to ‘fix’ CC (for fairly good reasons, even if you’re not a lawyer) before it has a chance to establish itself as a global social contract. [The GPL, in this sense, was an example of optimistic concurrency– and the lawyers are just not predisposed to be optimistic.]

Some other random thoughts that I want to write down:

  • Because I am an idiot, it only finally struck me yesterday the correlation between the CC’s NC and the GPL’s dual-licensing trick, as practiced by QT, mysql, etc.
  • CC’s ‘machine readable’ RDF-based license structure is more interesting/important than had struck me at first. It would be sort of interesting if every .c/.h file had machine-readable license data and the compiler verified that the linking/mixing process was legal. That of course would require a RDF-based description of whether or not licenses were mixable, which would be… well, politically charged if nothing else. *cough*OpenSSL*cough*
  • GPL is a single, global, social contract– much more a social contract than a license, really. One voluntarily enters into it, subscribes to its terms by choice, and gains benefits as a result. Ideal social contract. It would sort of be a shame if CC became ‘just’ a legal license and lost that additional dimension.
  • I’ve long been unhappy with our trademark situation, but when I realized that Creative Commons has nasty trademark issues too, I basically said ‘if they can’t figure it out, we probably can’t either.’ Turns out that they are going to try to fix it, which exites me no end, because if anyone is going to pull a legal rabbit out of their hat, it’ll be CC. It might be a while, though.