The fundamental dilemma, in my view, is that some uses of a work essentially destroy the openness of the work.

And then Bill said:

The idea of copyleft/SA is not to restrict those who would remix in good faith, but to prevent profit-taking landgrabbers from removing things from the commons.

First, let’s be clear that there is nothing anyone can do to remove a public domain work from the commons. Period. Even though Penguin Classics prints and sells copies of The Adventures of Huckleberry Finn, it will always be part of the public domain, and will continue to be available from places like Project Gutenberg. There is nothing that Penguin, Disney, or anyone else can do to move Huck Finn or any other verbatim public domain resource out of the commons.

Second, let’s also be clear that derivative works of public domain resources can be fully copyrighted, can be placed in the public domain, or can be licensed GFDL, CC, or any other license. Disney movies (which are fully copyrighted) based on stories (in the public domain) are an example. So, while there is nothing anyone can do to move a work out of the public domain, derivative works based on the public domain can certainly be copyrighted.

By contrast, ShareAlike (or copyleft) keeps derivative works in the commons by mandating which license they must be licensed with. This was the topic of my previous post.

So the discussion of copyleft approaches versus public domain approaches comes down to a simple question: do we choose to privilege people, or do we chose to privilege content? In the copyleft model, we privilege content (we guarantee it stays in the commons) at the cost of author freedom (to choose which license to use). In the public domain model, we privilege authors (we guarantee their freedom to choose which license to use) at the risk that some derivative works may leave the commons (be copyrighted).

So should we privilege people or content? For me, this is a very simple question. Content is simply a means to the end of supporting people’s learning. Content is never the end in itself. The idea that we might privilege content over people is frightening to me. As educators, people should always be the first, most important focus of everything we do.

More importantly, in the context of current realities where most of the open educational resources in the world are being produced by the so-called “developed world,” the copyleft requirement is even more sinister. With copyleft we once again find the “developed world” mandating solutions for the “developing world” (‘if you adapt our open educational resources for use in your country, you are required to license them the way we tell you to.’)

So… a few days ago I outlined the problems with copyleft / share-alike around license incompatibility in which GFDL and CC materials cannot be remixed, and most CC materials cannot be remixed with each other. Today I have argued that copyleft privileges content above people – and is especially problematic in the way it imposes “developed world” solutions on “developing world” users without flexibility or sensitivity to local needs.

Perhaps it is time for us to seriously discuss whether or not copyleft is good for the open education community?

Permissible Restrictions

Apart from these allowed restrictions, the license must not include clauses that limit essential freedoms. Especially, it must not specify any usage restrictions (such as prohibiting commercial use of the work, restricting use depending on political context, etc.).

• Attribution of authors: Attribution protects the integrity of an original work, and provides credit and recognition for authors. A license may therefore require attribution of the author or authors, provided such attribution does not impede normal use of the work. For example, it would not be acceptable for the license to require a significantly more cumbersome method of attribution when a modified version of the licensed text is distributed.
• Transmission of freedoms: The license may include a clause, often called copyleft or share-alike, which ensures that derivative works themselves remain free works. To this effect, it can for example require that all derivative works are made available under the same free license as the original.
• Protection of freedoms: The license may include clauses that strive to further ensure that the work is a free work, notably by enforcing some of the conditions specified in the paragraphs below: for example, access to source code, or prohibition of technical measures restricting essential freedoms.

This paradox has been rolling around in the back of my head / subconscious for months now, but something about reading this section in this context crystalized the problem in my tiny brain: how can a community so focused on freedom approve of any restrictions? Specifically, when expressing concern about restrictions making it difficult to reuse works, how can this community approve of the copyleft or share-alike concept?

Legal Hatfields and McCoys

The biggest legal problem facing the open educational movement is license incompatibility, as I discussed recently. For example, the millions of articles in Wikipedia and the 1.6 million media files in the Wikimedia Commons cannot be remixed with the thousands of courses and tens of thousands of media files coming from university OCW projects around the world. Why not? Because the “copyleft” scheme of the GFDL says that all derivative works must be licensed with the GFDL, and the “copyleft” scheme of the CC By-NC-SA license says that all derivative works must be licensed with the CC By-NC-SA. Hence, these materials cannot be remixed. It should be quite obvious that the sole culprit preventing this reuse and remixing is the copyleft restriction that is so happily permitted under the FreedomDefined definition of “free” (as well as the Free Software and many other definitions).

Although I am a huge fan of the open source definition and the free software definition, from first principles of freedom and openness I can see no reason to approve of any restrictions. In the free software definition, Stallman writes:

For example, copyleft (very simply stated) is the rule that when redistributing the program, you cannot add restrictions to deny other people the central freedoms.

If copyleft were implemented in this manner, then the Creative Commons licenses and GFDL would be compatible with one another. But copyleft is not implemented in this manner. The text of Stallman’s own GFDL reads:

You may copy and distribute a Modified Version of the Document under the conditions of sections 2 and 3 above, provided that you release the Modified Version under precisely this License.

Creative Commons similarly talks the right talk in the their human-readble definition of ShareAlike, “If you alter, transform, or build upon this work, you may distribute the resulting work only under the same or similar license to this one.” But in the Legal Code version of the document we read:

You may Distribute or Publicly Perform an Adaptation only under: (i) the terms of this License; (ii) a later version of this License with the same License Elements as this License; (iii) a Creative Commons jurisdiction license (either this or a later license version) that contains the same License Elements as this License (e.g., Attribution-NonCommercial-ShareAlike 3.0 US) (“Applicable License”).

So, in both the Free Software Foundation and Creative Commons contexts, copyleft or share-alike means “if you’re creating a derivative work, you have to use our license – and only our license.”

Why, when we are so worried about preserving freedoms, do we prohibit choice on the part of downstream users as to how they can license derivatives works they make? Why don’t we want to protect that user’s freedom to choose how to license his derivative work, into which he put substantial effort? The copyleft approach of both the Free Software Foundation and Creative Commons makes creators of derivative works second-class citizens. And these are the people we claim to be primarily interested in empowering. I can’t stress this point enough: the ShareAlike clause of the CC licenses and the CopyLeft tack of the GFDL rob derivers of the basic freedom to choose which license they will apply to their derived work. ShareAlike and CopyLeft privilege creators while directing derivers to the back of the bus.

CC Infighting

The following image shows which CC licenses are compatible with one another; in other words, it shows what content can be legally remixed with what other content.

You should notice that on this 10×10 grid there are only 33 smiley faces. In other words, only 1/3 of Creative Commons’ own licenses are compatible in terms of cross-license remixing. It’s difficult to believe, but true. This is so difficult to believe, and the detail so confusing that almost all people (and many people in the OER world, for that matter) simply believe that any page sporting a CC logo is compatible with any other in terms of remix-ability. Legally, of course, this is not true in the majority of cases. And, “But they both say Creative Commons at the bottom!” is not a viable legal defense. Do we really believe that “normal users” can navigate this morass? How often have you personally consulted a chart like this when remixing CC-licensed content? How often have you instead just shrugged your shoulders and said, “who cares?”

Public Domain as the PeaceMaker

You should next notice in the CC compatibility graphic that there is only one license that is compatible for cross-license remixing with all Creative Commons licenses (and with the GFDL, although the graphic doesn’t show it): the Public Domain license. Technically, this isn’t a license – it is a dedication of copyrights to the public domain. It is a formal rejection of all restrictions, and a complete freeing and opening of content. It is the ultimate expression of the ideals of openness and freedom. So why do we settle for only four freedoms? Why isn’t our goal to have all OERs eventually exist in the public domain? If we really care about ease of reuse, if we really respect creators of derivative works, and if we really want to see the open education movement succeed, we should have the public domain as our final goal.

Now, DON’T MISUNDERSTAND. I’m not saying that everyone who isn’t using a public domain dedication as of 5:00p tonight is an evil loser. I’m still very much a believer in and respecter of a step-by-step path into full participation in the open education community of practice. However, once upon a time I believed that the end legal goal of the progression was a license of some sort. Now I believe that the real legal end goal for open educational resources should be dedication to the public domain.

So the obvious question, then, is when will Mr. Big Talker’s (my) courses on USU OCW be placed in the public domain? Very soon now. More about that in another post.