Unlike the GPL, the CCL is a license for content rather than software. It was first released in December 2002. Since then, it has undergone a series of minor revisions. In its current form, the license consists of baseline rights and restrictions about copying and redistributing a work, such as preserving the copyright, and a series of conditions that can be mixed and matched, except where they contradict each other. These conditions are:
- Attribution: All uses of the work depend on the creator receiving credit.
- Noncommercial: Only noncommercial use of the work is permitted.
- No Derivative Works: The work can be used only without altering it.
- Share Alike: All use of the work, including derivatives, must use the same license.
The CCL is developed and promoted through a non-profit corporation led by a board of directors that is chaired by Lawrence Lessig.
As is the case with the GPL, revising the CCL is a public process. According to Mia Garlick, general counsel for Creative Commons, members of Creative Commons International, the organization formed to help port the license to different jurisdictions, are consulted regularly, as are major stakeholders. General members of the public can provide their input through the cc-licenses mailing list and the Creative Commons blog. In addition, they can view changes to the license online.
The actual writing of the license is done by Garlick in consultation with the Creative Commons board of directors, many of whom have a legal background or experience with licensing issues, and with members of Cooley Godward, Creative Commons' external counsels. Unlike the writing of the new GPL, in which consultation ends as each draft is produced, the new CCL is being written in what Garlick describes as "an iterative process," with the language being discussed as it is written.
Issues in the revision
Several of the concerns of the GPL do not exist for the CCL. For instance, except for clarifying the anti-DRM language, the CCL revisions don't need to address issues created by new technologies such as BitTorrent. Nor is compatibility with other languages a concern, no doubt in large part because the flexibility of the CCL conditions means that dual-licensing is usually unnecessary. Similarly, Garlick notes that "CC licenses are about content, which typically doesn't raise patent issues."
A basic concern in the revisions of the CCL has been to improve the existing language. Clarity and simplicity have always been major concerns of Creative Commons, and are likely a major reason for the success of the license. In the changes to the license available online, those interested can see where ambiguities and unnecessary language have been removed and clarifications added.
Another concern has been to address the growing popularity of the CCL outside the United States. To address this issue, the former generic license is now specifically the American license. In its place, a new generic license has been written. Garlick explains that this new generic version "is based on the language of international Intellectual Property treaties and takes effect according to the national implementation of those treaties." This new generic license includes a recognition of moral rights for jurisdictions in which the concept exists, and tries to balance them against the right to produce derivative works. Because it is not specific to the United States, the new generic license should make adaptations of the CCL in other countries easier.
Many of these issues are straightforward, in the sense that they require only a discussion of details rather than an attempt to reconcile philosophical differences. However, a more intensive part of the revision process is addressing the concerns of major users or potential users, then getting the rest of the Creative Commons community to accept the changes made for these special interests.
For example, answering the concerns of MIT OpenCourseWare, the electronic publisher of free course materials for the Massachusetts Institute of Technology, is more or less straightforward. Besides some of the clarification of language that was already occurring, the new CCL includes a "no endorsement" condition that prevents the name of the copyright holder from being used to promote the product. Since a no endorsement condition is already available in the Free Software's version of the MIT License and the BSD License, the precedence for the condition is well-established, and it has apparently caused relatively little controversy.
However, satisfying members of the Debian distribution is proving more difficult, if not impossible. Famous (if not infamous) for its punctilious scrutiny of licenses, Debian does not currently accept the CCL as a free license. Debian's concern is that some of the possible variations of the CCL violate the Debian Free Software Guidelines, which govern what can be included in the distribution.
As summarized by Evan Prodromou, these possible violations include the ability of someone using the Attribution variant of the CCL to have their credit removed from a derivative work, and ambiguities in the language about DRM and trademark restrictions. More seriously, some variants of the CCL, such as Attribution-Share Alike and No Derivatives, are considered completely incompatible with the Debian Free Software Guidelines.
In the spring of 2005, at Lawrence Lessig's personal request, a Debian workgroup consisting of Evan Prodromou, Don Armstrong, Benjamin Mako Hill, Branden Robinson, MJ Ray, Andrew Suffield, and Matthew Garrett began consulting
with Creative Commons representatives to try to resolve these issues. According to Prodromou's report in August 2006, these meetings resulted in clarification of the anti-DRM language, but Creative Commons did not accept Debian's efforts to include language that would make an exception to DRM works that had dual licenses or "parallel distribution" that included a non-restricted form.
The trouble is, other groups involved in the process are resisting the idea of parallel distribution. According to Garlick, many do not see the need for parallel distribution, and worry that it would complicate the license. Many, too, view the concept as weakening the anti-DRM language. "Consequently," Garlick says, "CC is currently not proposing to include this new parallel distribution language as part of version 3.0." In theory, Debian could still decide to accept the CCL through a general resolution, but, given Debian's past record on licensing, such a move seems highly improbable.
Talking to NewForge, Garlick would only say that such issues are being resolved "by discussion." However, just as rejection of GPL 3.0 by Linus Torvalds and many kernel developers may undermine the efforts to revise it, so the CCL may face the specter of continued rejection by Debian. The attempts to resolve theses issues are undoubtedly a large part of the reason for the delay in publishing the final version of the CCL, which was scheduled to be introduced in July, according to the originally timetable.
For now, whether everyone can reach a consensus is impossible to predict. Considering the similarities in the troubles faced by the GPL and the CCL, observers might wonder if the free copyright communities have grown too large and too diverse to agree on anything so detailed as a license unless it is one with which they are long familiar. The choice seems to be to delay the release of the licenses, gambling that consensus will emerge from continued discussion, or to ignore the concerns of some stakeholders and proceed despite enormous blows to each license's prestige and credibility.
In the CCL's case, the willingness to reach an agreement seems to exist, but whether that willingness will be enough to reach a satisfactory agreement is still anybody's guess.
Bruce Byfield is a course designer and instructor, and a computer journalist who writes regularly for NewsForge, Linux.com and IT Manager's Journal.