Creative Commons is jumping on the license-rewrite bandwagon and planning to publish a draft of version 3.0 of their license modules. This has occasioned some discussion of the ways in which CC licensing can be improved (I hope to write more broadly about this later). For me, it suggested re-treading an idea that CC failed with -- the so-called "Founder's Copyright", and giving it a bit of new life via a better implementation and a little cross-pollination with free software business models. After much pounding on the mailing list, I think I've got a good idea of the shape this ought to take, and I'd like to make a condensation here of my case for "Sunset NC/ND" modules for Creative Commons.

In brief, I proposed this: why not add a series of "Sunset modules" that expand the commons not by expanding on 'fair use', but rather by expanding on 'limited times'. Let non-commercial and non-derivative license modules be assigned a reasonable sunset via modules with (say) 1, 3, 7, and 14 year sunsets. This would allow the artist to use conventional proprietary licensing fees while the NC/ND was in force, but then expire into the copyleft (or non-copyleft) free commons. Theoretically this already happens with copyright expiration -- but at 80 years and rising, I argue that that's too long to be useful: it provides no incentive to the commons to assist the artist in distribution or promotion.

I'm going to assume you are familiar with Creative Commons (CC) licensing. If not, please read the Meet the Licenses, from the CC website. None of this will make sense if you aren't familiar with these basics.

Free-licensing advocates lament the "fragmentation of the commons", a problem associated with "license proliferation". Richard Stallman hates it so much he's publicly spoken against Creative Commons for confusing the issue. CC plans to defuse some of that by better distinction of the free licenses in version 3.0. But I think something more definitive is necessary: we need a more free-commons friendly version of the NC license.

The problem with NC

The problem with the "NC" licenses (specifically CC-By-NC-SA, CC-By-NC-ND, and possibly CC-By-NC) are two-fold. First of all (and most importantly), they are incompatible with the free CC-By and CC-By-SA licenses. Second, they impede distribution, because they require distributors not to make any money off of the distribution -- even indirectly, via advertising. I can't use a NC work in this blog because of those ads you see on the left and right of this page. CC recently made this clear in its use-case analysis of the NC module (PDF, 91KB) -- although they've unfortunately buried it where it is almost impossible to find (a conspiracy theorist might imagine they were trying to hide the problem instead of fixing it, but I hope it's just ordinary bumbling).

The compatibility problem means, you can't take work under an NC license and combine it with free licensed works to make a new work -- because the new work couldn't be released under either the NC or the SA license.

This has caused a fair amount of acrimony -- the most recent being Stallman's public 'un-endorsement' of the Creative Commons. Most arguments are along the lines of saying that Creative Commons was wrong to provide the NC license module, and should withdraw it -- people should either go completely to free licenses, or they should just stick with 'all rights reserved' and not misappropriate the trademark on 'our' commons.

The inability to make money off of distribution exacerbates this, because it means that the distribution of NC works is very inefficient. That means fewer people will see the work in order to "try before they buy", which is the promise that people are buying into when they choose NC for their work.

But artists still need an NC option

Of course, the trouble is, NC is the single most popular license module that Creative Commons produces. Why is that?

The primary pay-off of a copyleft commons is what I call "collaborative leverage". I put in a buggy program that scratches a very important itch, and through the magic of open source, I (and the world at large) get back a fully-finished, nearly bug-free tool that scratches that itch even better. There are other advantages, but IMHO, this is Reason #1. It is, more obviously, the reason for using copyleft, instead of just a free license (CC-By-SA instead of CC-By, with Creative Commons licensing).

This is such a boon, that I wind up not caring that I lost the opportunity to make money from selling copies of the original program. I saved more on development than I lost in sales. More importantly, from an individual's point of view, I was able to develop what I would've needed to start a company to do if I wanted to use a proprietary license.

However not all projects are amenable to this approach. In particular, with aesthetic expressions, it's questionable whether that kind of collaboration is even desireable, let alone worth the cost. More common is the desire that works be combined with works of different media (e.g. a soundtrack with a film), or used in a collective of some kind (e.g. Wikipedia). But in any case, for a professional artist, it just won't do -- you need to be able to make money doing your work.

There are, of course, various methods to raise that money which allow a free license to be used. For example, one can use "rational street performer protocol" to raise money before or during production. One can look for grant-based funding. One can make pre-sales -- a significant source of income for the Elephants Dream free-licensed animated movie project.

These suffer the usual problems of public funding to various degrees: primarily in that they require the money to be presented before the work is, which makes it difficult for works to compete for funding based on quality (or existence -- a lot of vaporware could get made this way). But are we so anti-capitalist? What about private funding methods? What about selling copies after the work is created?

Of course, that's where a non-commercial clause can help: the artist may allow a certain amount of "informal" sharing, but they disable wholesale distribution. So the artist instead makes his money off of license sales. It's the proprietary business model, of course.

We shouldn't scoff at that system too hard -- it obviously worked for a long time, as we have very active publishing industries because of it, and there are a lot of artists and writers who make their money from that system. Of course, it can be (and has been) abused. The "limited times" promise in the US version of copyright for example has been made into a mockery by copyright extension acts, so that copyright is effectively unlimited now. Even the 80 years or so we have may not last -- as soon as Mickey Mouse is threatened with public domain status, Disney is sure to lobby for yet another extension. Who knows how long that madness will go on.

Bringing back 'limited times'

So artists want some kind of NC option to protect license-based income from their works, and I can't find too much fault with that, myself. But isn't there some way the NC license commons could "play nicer" with the free commons, and get some of the payoff from that community as a result (in goodwill, if nothing else)?

So far, Creative Commons' response has been to expand the range of 'fair uses' of copyrighted works, but surely it's not unreasonable to tackle the 'limited times' problem, too?

And that's the seed of the "Sunset" module idea: make minor modifications to the NC and ND license modules so that they recognize the possibility of a licensing sunset, and respond to new sunset modules, say "1Y", "3Y", "7Y", and "14Y". Then when you see a CC-By-NC-SA-1Y work this year, you'll know that next year, it'll be available as CC-By-SA, and a CC-By-NC-ND-7Y work will be available under CC-By in seven years.

I originally was waffling over whether the sunset should apply to the "Sharealike" (SA) module. But then I realized that it was pretty simply to let the artist choose. If the sunset applies to non-derivative, then artists who wanted works to expire without SA could just use ND instead of SA. The effect in most cases would be nearly identical, until expiration, when ND would expire, but SA would not (before expiration, the SA would have little effect -- and where it did, the artist probably would want to keep the SA option, anyway). So for simplicity, I suggest that NC and ND respond to the sunset, but SA is unaffected.

It's not perfect of course -- I'd rather work with stuff that goes immediately into the copyleft free commons, but it's good enough to raise some interest.

In fact, of course, there are a few "free-software" projects that actually work on a sunset basis like this. For example, the Python Imaging Library and the Ghostscript rendering software both retain the most up-to-date release under a proprietary license, while the older releases are made available under the GPL. This is basically the same idea, applied to art, and giving a actual time limit, instead of referring to "versions" (which might or might not make sense for a given work of art). So this might just be another case of applying a free software business model to art.

This idea has been argued for and against in the discussion on CC-licenses (and to some degree on CC-Community), which might interest you if I haven't provided a good enough summary. Objections have ranged from 'Founder's was a flop and so will this be -- no one will use it' to 'why not just do it with the license grant and leave CC out of it' to 'the commons is not fragmented, but extended by NC works, so this is a solution for a non-problem'. I remain pretty confident that it would be a useful idea, though.

What do you think?