And now... for something completely different.
The GFDL is a document with great goals, great principles, and I agree with all of the basic elements of Free Content. Let's just get that on the table first. I've read quite a lot of Stallman's texts, and works by other players in these sorts of debates like Lessig, Raymond, Boyle, etc., and I think I understand the terms of the debates over this, and some of the legal aspects, though I don't claim to be an expert in either domain.
But there are some ways in which it is not ideal in its implementation. Some of these have been gone over in some detail on here, but the basic summary is that it is a license made primarily for software manuals and is not always best adapted for Wikipedia's specific technical and informational purposes. There are, to say the least, sections which could be clarified, improved, or dropped altogether. And there are also places where we could imagine improvements added (such as allowing authors to be dis-associated with re-used content if used in a way which would be defamatory to them -- the old case of the Holocaust deniers using our articles as a base for their own literature and then claiming that Jimbo was an author).
But how to escape? The GFDL is viral -- once you license something as it, you can't undo it. What is made GFDL stays GFDL -- that's the purpose anyway. And, in principle, I agree with that: the goal of this is to make sure that what was "free content" not only stays "free content," but generates more "free content." And it improves Wikipedia's credibility to be committed to a license maintained by an external source: nobody can claim that someday WP will turn around, change its license, and suddenly have proprietary content. In order to accomplish that at present, we'd have to do a hostile takeover of the Free Software Foundation. Let's assume, our of the principles of practicality and good faith, that this will never be an option on the table.
So how to escape?, he asks again. I've been puzzling over this for some time (think of it as on of my hobbies, the sort of thing I muse about in the shower). Here are some thoughts I had.
The idea of multi-licensing has been pursued on the project at different times, whereby contributions are indicates as being licenseable under the GFDL or another, similarly "free" license (i.e., CC-BY-SA or CC-SA). There was also the big push, awhile ago, to get users to put templates on their user pages indicating that their present, future, and, I think, *past* contributions were multi-licensed as well -- I believe it had to do with making certain articles compatible with WikiCities' license. The basic idea was to run a bot to find all of the "authors" the articles in question and see if they would agree to this. I don't know how this worked out, but it was an interesting idea.
Based on this principle: can one really ask users to re-(multi)-license their PAST contributions? That is, can I say, "All those contributions I said were under the GFDL? Well, now I want them to also be GFDL or CC-BY-SA." Legally, I'm suspicious, but I'm also not a lawyer.
If this principle works -- couldn't we change the terms of use? That is, instead of every edit being licenseable under the GFDL, couldn't we change it to say that "this contribution, and any other contribution I have previously made, is licenseable under the GFDL or any other similarly 'free' license"? It wouldn't necessarily get *all* of the content out of the GFDL but, if we assume that many of the editors now were editors previously, it would potentially "free up" a very large amount of content. If an individual editor objected to this for some reason (I can't imagine why, but let's just say they did), then they'd be prohibited from editing, the same way we do when people suddenly claim that the intent to retain copyright on their edits.
It is just a thought I had -- the only one I could come up with which seems really plausible, aside from the possibility of the FSF being convinced to make updates to the GFDL (which I suspect they would be very dubious about, especially if the edits were primarily to benefit Wikipedia).
Just a thought, not a clarion call. Getting out of the GFDL may not even be necessary, but I think thinking about it as an option might be worthwhile (though again, not because I disagree with its principles in the slightest, just some aspects of its implementation).
I'm very interested in what others will have to say about this, and hopefully it will be seen in the spirit I intend it, which is more on the level of inquiry than policy.
FF