What open source licensing could learn from Creative Commons

Creative Commons
Image by opensource.com
submit to reddit
 
(9 votes)

The arrival of the ten-year anniversary of Creative Commons is an opportunity to express gratitude to an organization that has done so much to promote the sharing of cultural works and to challenge traditional assumptions about the appropriate use of copyright.

I have been a critic as well as an admirer of Creative Commons. Last year, here on opensource.com, I noted that the CC license suite, though inspired by open source licensing, was at odds with norms of libre culture licensing by embracing, under a single legal brand, form licenses that prohibit commercal use and creation of derivative works. The result, I complained, was "a general confusing dilution of the meaning of 'openness' in the context of cultural works" and confusion on the part of both authors and users of CC-licensed material. Creative Commons has recognized at least some aspects of this problem in the course of its work on the 4.0 license series. (For example, there has been an interesting recent proposal to relabel the controversial NC licenses with "Commercal Rights Reserved.")

In the same article I acknowledged that "[f]rom the perspective of open source, the very different legal culture established by Creative Commons is not only tolerated but welcomed as an enormously successful experiment that is safely cabined to non-software content". Since then, I have come to have a greater appreciation for this experiment, particularly with respect to its potential applicability to the improvement of open source licensing.

Free software licenses were developed in a largely distributed and organic fashion. This has given open source licenses something of a baroque and quirky character, which, while appealing (to some of us), has some drawbacks. The coordinated, centralized manner in which CC licenses are conceived, drafted and revised, and the successful occupation of the full policy field of open and quasi-open content licensing by Creative Commons, offer a model that, if applied with the greatest care, might be beneficial some day for the open source licensing world. For example, CC licensing has allowed the open content world to avoid some of the problems of license proliferation that are commonly thought to plague open source. The variety of CC licenses has actually been reduced over time. Moreover, while the various CC licenses implement different policies, for a given localized or international version they have common textual elements, which may reduce the difficulty of interpretation and revision. I was once more skeptical about this approach (and its possible future application to open source) than I am today. Therefore, to quote again from my 2011 article, if "someone [were] to propose that a seemingly simple Creative Commons-like suite, designed by some central committee, should replace all traditional FLOSS licenses as they exist organically today", what is most likely to concern me is the central committee part (and who gets to be a member). The danger for open source, at least, is excessive influence from my own guild, excessive influence by what (for lack of a better term) I have thought of as "FLOSS establishment institutions", and insufficient participation by developer constituencies.

A notable feature of CC licenses that has no precise counterpart on the open source side is the great emphasis on simplification of use, understanding, and identification of the various license categories. Here as well I tended in the past to have a skeptical view of the CC approach. Most troubling to me was the CC practice of separating the so-called "deed" (also formerly, and quite unfortunately, called "human readable code") from the (invariably lengthy and complex) license text proper, still described, somewhat annoyingly, as the "legal code". A particular  concern is that authors will be inclined to read the "deed" (which is an imperfect summary at best) as the actual license.

Yet there is no question that the "deed" abstraction of license complexity is partly why Creative Commons licensing has been so accessible to so many content authors. I have more recently begun to wonder whether the serious copyleft wing of open source has been suffering disadvantage because of concerns about undue complexity among developer constituencies. That is a complicated topic and well beyond the scope of this article. But the example of Creative Commons shows one way in which complexity may be managed to promote license adoption and reuse. If CC "deeds" appeal to free software authors, it may be due to natural preferences for simplicity and minimalism. Perhaps open source would benefit from a unified system of smaller, simpler, and more easily apprehendable licenses.

""
Creative Commons CC0 Universal Public Domain Dedication.

7 Comments

Bradley M. Kuhn's picture

As always, I am glad to read what my good friend Fontana has to say. :) However, I must admit that even after studying the CC license suite for the last decade, I'm not sure there is all that much learning from its rise that we can easily apply to Free Software licenses. What I've mostly found are a series of “glad Free Software licensing doesn't work that way” responses.

I specifically agree with Fontana's earlier points about the diversity and baroque nature of drafting of Free Software licensing, but I like that aspect. CC is a single organization able to (more or less) dictate policy for the entire licensing infrastructure in the Free Culture community.

One of the major (and, IMO, insurmountable) problems is the mere existence of ND and NC versions. If CC has succeeded in any educational efforts, it has been to convince potential licensors that NC and ND versions are morally equivalent to the pure BY-SA options. CC is ultimately a centrist, neutral organization. CC takes the air out of the room for the more radical elements of the Free Culture licensing advocates.

This is where the Free Software world has things right. Brian Behlendorf of the Apache Software Foundation once told me: I'm glad you're around to be a radical, because if you weren't, I'd be a radical by default, when I'm actually a moderate. This concept applies to the licensing infrastructure: The Apache License is seen as the moderate, permissive alternative, but Apache Software Foundation and its license would be radical, if FSF and GPL weren't around.

Indeed, Fontana, while it's true that your copyleft-next project hasn't received as much interest as you'd have liked, wouldn't a similar project to create a replacement for CC-BY-SA have been an entire non-starter in the Free Culture community? As you know, I've tried to participate with you on copyleft-next even though I'm on the Board of Directors of FSF, but do you think that if you tried to make a CC-BY-SA alternative, that one of CC's directors would come help you?

Frankly, I think what CC's suite still shows more than anything else is the danger of licensing monoculture. That problem drowns everything else. I like CC-BY-SA-3.0-USA as a reasonably good strong copyleft for non-technical works, and CC-BY isn't too bad of a permissive license for the same. But, with CC controlling all Free Culture licenses, I think it's really difficult to draw these distinctions with newcomers. I still see “Licensed under the Creative Commons license” regularly. Licensing monoculture is, in some ways, more confusing than what we have the Free Software world.

I do agree Free Software licensing wonks can help simplify and improve understanding, building on the “deeds” model. I began a brief collaboration with noted user interface researcher, Michael Terry about how we might think about presenting the content and message of GPL better to users. He's got some amazing ideas, but sadly I couldn't get any copyright licensing lawyers interested in working with us on the project, so I had to abandon those efforts. But, honestly, I think most people don't really grok CC licenses, anyway — despite CC's efforts in this regard. I ask every CC licensor that I meet to explain the differences between the various CC licenses, and they usually can't. (I'd love to see a survey study to back up and/or refute my anecdotal assessment there, BTW.)

IMO, this whole problem goes back to Noam Chomsky's point about concision. Namely, it's very difficult to engage the public in discussions on important issues that require deep and complex context. I believe the Free-Software-using public's general inability to easily navigate the baroque set of licenses comes down to the simple, unfortunate (but understandable) “this makes my brain hurt” response.

drew Roberts's picture

Even now, if the FSF were to come out with a strong copyleft license (it would need to be stronger than the current BY-SA) for non-software works there would likely be a good amount of uptake. Especially if it could be crafted to be compatible with cc BY licenses.

It would also help such uptake if the FSF and RMS would give up on the idea that some works are ok being non-Free and that especially in light of the current copyright situation.

So perhaps the copyleft-next license / project could aim to fulfil this role and be a good license for both software and non-software in one. If it does not do so already.

If it already does this, let me know and I will try to find time to get more involved.

all the best,

drew
http://zotzbro.blogspot.com/

Bradley M. Kuhn's picture

Fontana is the dictator of copyleft-next (although the dictatorship is starting to wane a bit :), but it's up to him for the moment if he wants to try to make it a universal-work license.

Meanwhile, I don't really agree if FSF is necessarily the right place for non-software licenses. FSF's work in this area has mostly been superseded by CC already (FSF even uses CC licenses for some of its own stuff now). Meanwhile, there's plenty of software licensing work to be done.

David A. Wheeler's picture

This article closes with, "Perhaps open source would benefit from a unified system of smaller, simpler, and more easily apprehendable licenses." I would argue that there is already such a set. The problem is that not everyone realizes that.

I think the OSI (and others) should do a better job emphasizing that people should prefer to use a small set of widely-used mutually-compatible licenses. Don't forbid other licenses (you can't anyway)... just make it clear that there are serious advantages of choosing one of the common standard ones. But what are the common ones?

The DoD OSS FAQ specifically notes that, "nearly all open source software is released under one of a very few licenses that are known to meet this definition. These licenses include the MIT license, revised BSD license (and its 2-clause variant), the Apache 2.0 license, the GNU Lesser General Public License (LGPL) versions 2.1 or 3, and the GNU General Public License (GPL) versions 2 or 3. Using a standard license simplifies collaboration and eliminates many legal analysis costs."

If you stick with those licenses, then these licenses are already mutually compatible (Apache 2.0 isn't compatible with GPL2, but it is with GPL3 and in most cases that works).

OSI's list of licenses "that are popular and widely used or with strong communities"
includes all those, and is a step in the right direction. One problem with OSI's list is that it has some extra ones, which have known problems: Mozilla Public License 2.0 (MPL-2.0), Common Development and Distribution License (CDDL-1.0), and Eclipse Public License (EPL-1.0). These licenses are "wrong" in any sense, but the CDDL and EPL are GPL-incompatible, and since the GPL is the most popular OSS license, it is unwise to choose a GPL-incompatible license for a new project. The new MPL helps considerably, since the new MPL (unlike the old one) is GPL-compatible; I know of no efforts to fix the CDDL and EPL.

Blazing everywhere a list of "common licenses", instead of hiding it, would help. Not including the CDDL and EPL would be best, but if that's politically infeasible, an asterisk that noted "GPL-incompatible, which can inhibit collaboration because the GPL is the most common OSS license" or some such.

mlinksva's picture
Community Member

Hear, hear.

I think there's room for long-term improvement in apprehension, but that'll have to come from guidance outside of licenses as you suggest, and improvements in existing licenses in the most popular public domain->AGPLv3+ compatibility path, or aggressive compatibility from those out of it (such as MPL2 achieved) and new ones (such as the current copyleft-next draft).

It is nice that there are no fundamental-purpose incompatibilities among FLOSS licenses, unlike among CC licenses. This is a usability plus, even if not evident by examining individual licenses in isolation.

Note there is a possibility of a one-way CC-BY-SA-4.0->GPL compatibility mechanism. I think that'd be a very good thing long term, practically for things like games and hardware designs, and normatively for encouraging people outside the software world to consider and demand modifiable forms, and to collaborate more closely with FLOSS.

There's also the problem of incompatible and newish copyleft licenses targeted at databases and hardware designs...

gunnar's picture
Open Source Evangelist

I think of CC as compelling not because of the "brand" or the command-and-control necessarily, but because it's very easy for the recipient of the license to understand what they are and are not allowed to do – as Richard mentioned. It's focused on the end-user. There may be confusion at the margins, but the terms of a CC license are generally more easily grasped that your average OSI-approved license.

The ease with which I can express my intentions as a property owner under CC is something that would be very valuable in the software world. I don't think this necessarily requires a top-down regime. I know this idea has been flirted with before, but I think it would be immensely helpful to have a consensus set of signifiers, oriented to the end-user, for specific, recurring software licensing terms.

gunnar's picture
Open Source Evangelist

A timely update from Creative Commons on the future of the -ND licenses:

http://creativecommons.org/weblog/entry/35773