CC 4.0, an end to porting Creative Commons licences?

Last week, Yours Truly attended the 5th edition of the Creative Commons Summit in Warsaw, Poland. This was an impressive event bringing together Creative Commons affiliates and enthusiasts from all over the world.

Anyone who has been following the rise of Creative Commons as an international movement may have noticed that there was a small hiatus between the Sapporo summit in 2009 and this year´s meeting. The break was needed, CC found itself at an interesting and unique position, it had grown from a small licensing project that would try to replicate the success of open source and free software in the creative arena, and it became a global phenomenon with millions of objects available online. CC went from being a niche geeky project to a global standard. This process clearly generated institutional stress, and the community had to stop to gather its thoughts and analyse where to go next.

The difference in both breadth and tone of the summits was striking, and I felt that the Warsaw summit saw CC coming back to its roots. The first summit in Harvard was a small affair, almost a family event. The Rio and Dubrovnik summits were massive, almost a “who’s who” of the open camps. Warsaw was a family reunion, true, a big family reunion, but it was clear that this was a more focused gathering of the core support for CC at an international level. This was a meeting of the very heart of the CC movement, from the large projects to the smallest and newest affiliates, it was humbling to see the energy on display.

While the summit was supposed to act as a gathering of affiliates, there were very big news in store. Undoubtedly, the biggest announcement to come from the conference was the launch of the drafting process for the version 4.0 of the licences (we are currently on version 3.0). There are various important reasons to undertake an update at this point in time, the main one being a need to overhaul the provisions on the sui generis database right, which is currently not covered by the licence. There are also some tweaks needed in language to make the licences more international, and there will be discussions on what can be done to make it easier for large institutions to adopt the licences and release their content under the CC standards. This is personally one of the most needed tweaks; think of governments, international organisms and memory institutions — these are organisations that have important amounts of content and data that would enhance the commons, and if shared under CC licences would allow more data to be freely shared for the benefit of all. Moreover, any institutional adoption brings us closer to wider recognition of CC as an international open licensing standard, and would make it easier for some reticent adopters to recognise the movement. I am thinking particularly collecting societies, which up to this day still constitute a stumbling block in the music arena.

But suggestions have not been devoid of controversy. One of the proposals made at the meeting is that it is very possible that we will finally see an end to the porting experiment. For those unfamiliar with porting, when Creative Commons started, its quick popularity arose the interest of legal scholars and lawyers from all over the world. As the licences were drafted specifically with U.S. law in mind, it was felt that the creation of a network of associates that might port the licences into a local language would be the best way to ensure the enforceability of the documents across jurisdictions. Notice that we are not talking about mere translations, but to take the existing text, translate it, and then make necessary changes required by local legislation. This process was undertaken by local legal experts in the areas of copyright and licensing, creating in my opinion an unprecedented experiment in legal drafting. Here we have a series of documents that have been adapted from one country to the other bringing together an enviable wealth of experience and talent.

By most measures, the porting experiment has been a great success. With version 1.0 of the licences, only two countries undertook the porting process: Israel and Finland. By now, there are CC projects and affiliates in 71 jurisdictions, with 55 of those having conducted a successful porting process. But it is precisely this amazing success what has made the porting process a bloated and unmanageable exercise for a small non-profit organisation like Creative Commons. Each new jurisdiction that joined the movement and decided to port would add a new set of 6 licences for each version, which has meant an uncontrollable growth of the total of licences out there. At the time of writing, there are 550 CC licences, not counting the ones in drafting process. Moreover, it has become clear that the first rush to create a national port is a time-consuming endeavour, and that not all jurisdictions have the resources to continuously undertake it. For example, only half of all jurisdictions have ported their licences to 3.0, with large markets still stuck with 2.0 (e.g. UK). The growth in licensing practices means that CC must spend more money in legal staff to support a more jurisdictions, as each new version will generate another round of harmonisation. Similarly, the threat of incompatibilities between versions cannot be neglected.

However, for me the killing argument is that the vast majority of users are bypassing national ports and using the unported version of the licences. Take the raw linkback data for 23/09/2011; this uses search engine data of how many objects link to any given licence, which gives an indication of percentages of use for each jurisdiction. There were a total of  263,014,105 objects using  CC licences, of which 189,741,280 were using unported versions. This means that 72% of all items released under a CC licence do not make use of the impressive set of more than 500 ported versions. The users are voting with their clicks.

These reasons have prompted Creative Commons to propose that version 4.0 should not be ported, but only translated. This will require a solid draft that should be very international in nature, mostly making use of international treaties and other harmonisation tools. This would be more resource-friendly, but also would be future-proof.

This announcement was met with some animosity from some quarters, and enthusiasm from others. Having participated in two different jurisdictions (3 if you separate Scotland from the UK), I have to admit that I believe porting is untenable in the future, but I understand why some people are uncertain about losing porting altogether. As far as I could tell by listening to those who were against the change, the strongest arguments in favour of maintaining porting are two: community and legal.

The community argument is that the vibrant international community that we saw on display at the Warsaw summit is precisely the result of a heavy porting process, which means that by losing porting, it will be more difficult for new countries to join the CC process. This is certainly the one argument that I find more troubling, but it can also be seen as a glass-half-full scenario. While porting a licence gets the lawyerly juices flowing, it does not a community make. It seems like there are many countries where there is clear community interest in CC, but the lack of legal expertise in the topic means that the community does not develop as it should. If the requirement to port a licence is removed, many organisations could simply join as CC affiliates, enhancing the community.

The legal argument goes something like this: the licences are mostly American in nature, so a national porting process is needed for the licence to be enforceable in the target jurisdiction. I have to admit that I find this a rather weak argument. Firstly, copyright law is highly harmonised internationally thanks to the existence of widely adopted international treaties such as the Berne Convention and the WIPO Copyright Treaty. This does not eliminate national quirks and requirements, but it certainly makes an international licence much easier to achieve if one follows the language and letter of the treaties. Secondly, those tricky topics where there is more international divergence can be worded just to apply in jurisdictions in which these require special rules, e.g. moral rights and database rights. Careful wording, such as the one that already exists in unported 3.0 versions regarding moral rights, would solve most discrepancies. Thirdly, copyright licensing already works quite well without porting. When was the last time you installed a piece of software that had a licence ported to your jurisdiction? If your answer is negative, do you think that licence is invalid? Finally, courts around the world have already been enforcing unported CC versions of their licences (see my presentation on the subject for the Warsaw summit). We have been applying international contracts for millennia, give the courts some credit.

So, while I will be very sad to see porting disappear from a merely selfish reasons (seeing the Costa Rican flag on top of a CC licence is one of the proudest moments of my legal life), I strongly believe that porting has served its purpose, and that we should adopt a model more akin to that present in Free Software with the GPL. The GPL v2.0 is as American a legal document as you will ever find, yet it has been implemented in various jurisdictions (mostly Germany).

So we should be brave and take this as an opportunity to improve the licences and tighten their application. Come join us, the water is warm (or the Kool-Aid is sweet, your mileage may vary).

18 thoughts on “CC 4.0, an end to porting Creative Commons licences?

    • That would be indeed cool. I think that the argument used against that in general is that CC is not supposed to be used for software, in which case incompatibilities are less of a problem.

  1. But that’s a bogus argument, because there are games distributed under GPL with artwork under cc att-sa. What do you do, now, if you want to turn the cool spiral image from the logo into an SVG and subsequently into a cool spiral generator?

    Or think fonts – or better still: souce-code documentation with the actual sources included which embeds the utilized graphics.

    Or graphics with logic – or programmed music. Having cc att-sa and GPL as incompatible licenses arbitrarily limits the ways how people can structure their creations, because they must not mix cc and GPL content.

    • Thanks Arne,

      Are these hypothetical situations, or can you give me specific examples of where these things take place? Are developers encountering these issues in real life? I'm really interested in case studies if there are any.

      The reality as far as I can see is that incompatibility is not an exclusive area of CC vs GPL, but that it happens all of the time even in software projects, where there may be source code from incompatible sources. This is precisely why ohloh and the Open Source License Checker exist.

      It is my impression that at the moment developers make do with all sorts of licensing arrangements. Do not get me wrong, compatibility is an important goal, but I strongly believe that sometimes it is overplayed, and that in case of conflict a court is perfectly capable of deciding based on the existing licences. The proof is that as far as I know, there has never been a single compatibility case for FOSS licences.

      • The most radical option would be to simply use as cc by-sa the plain GPLv3 or later with a special permission: Allow to distribute without sources and use a link to cc by-sa as license reference.

        That’s essentially the way of the LGPLv3 (just with a different permission), and it would give cc by-sa the legal foundation of the GPL — which has already been proven in court multiple times.

        • "That’s essentially the way of the LGPLv3 (just with a different permission), and it would give cc by-sa the legal foundation of the GPL — which has already been proven in court multiple times."

          So has CC, two weeks ago it was enforced in Germany.

      • I’m hitting this all the time, because I am writing a free pen-and-paper roleplaying system with an implementation in python, which is hard at the border between art and code.

        There is source code (scribus files, text sources, python-code) which gets parsed into obect code (PDFs and *.pyc files) and there are Python-scripts which include part of themselves into the deployed files (you know: name strings and very simple templates for parsed websites).

        To play it safe, we use the GPLv3 or later for everything, so we can use artwork from battle for wesnoth, but all the artwork from Ryzom is excempt for us (even though it would be a good fit), because we would not be able to mix it with GPL artwork from Battle for Wesnoth or even to overlay it with Text under GPL, if the result gets used as an image – for example on our website. The same is true the other way round.

        It is all and nice that there has not yet been a single case of compatibility, but this is part of my heart I’m working on, and I really don’t want to lose it because someone realizes he could hurt us by just trying to sue – we would not be able to withstand a lawsuite, if there is even a meeger chance that we might lose in the end.

        There are already all kinds of fishy copyright-based suing-companies. When their lwyers find someone who wants to give them money to down a free project, they will.

        The problem why compatibility between cc att-sa and GPL really hurts, is because they actually try to achieve the exact same goal (copyleft) just by slightly different means. For works where the final product is the source (scripts and images where the artists don’t keep an extra file for modification) the GPL and cc att-actually have the same implications: Keep everything free which you do with this. The GPL just adds a requirement in case the editing-format differs from the consuming-format.

        But there is one grave difference: The GPL wants you to use the GPL for the final product, cc by-sa wants cc by-sa.

        Since the GPL secures the same freedom as cc by-sa, it might be completely in line with cc by-sa to allow using by-sa licensed works under GPL, and this would simplify dirstibution for many free games. And it would allow us and Battle for Wesnoth to use the artwork from Ryzom – and vice versa, if Ryzom would simply allow complete works under GPL.

        Since we’re a small group operating at the border between code and art ( ) I believe that the case of Battle for Wesnoth and Ryzom is the stronger one: Both are free games, both fantasy-aspected (Ryzom in parts) and they cannot use their respective artwork, because Wesnoth uses GPL for everything from code over campaign-descriptions to artwork and music (which is the sensible choice, if you don’t want to worry bout licensing at all), while Ryzom uses AGPL for the code and cc by-sa for the artwork.

        • Thanks for the answer. I empathise with borderline cases such as this, and as I said, compatibility is a concern. However, I think that software developers are being too cautious. For example, if you use some CC licences under a sharealike provision, I think that many courts would find that this can be released under a similar copyleft clause. The law is more flexible in licensing agreements than people give it credit, it is not a completely binary issue.

          Even if that does not convince you (and I agree that you want to diminish your liabilities), you could try to find the authors and get a permission from them to reuse the work under GPL, which would also serve your purposes. In my experience, people who licence under CC are open to reuse.

  2. As long as there is no clear law which says that I can use cc by-sa under GPL, I won’t: We have no case law in germany.

    And asking the authors if I can use their works under GPL is not always possible – and it is unfeasible when working creatively, because it is a roadblock you need to resolve, before you can do anything: You need to write an email and wait for the answer – and when/if you get the answer, your creative flow or your free time is already gone.

    Which CC was enforced in Germany (I did not hear that, but if it is none of the nc versions, that’s great news!).

    What exactly would be an argument against explicitly allowing reuse of cc by-sa licensed works under the GPL?

    • Hi,

      This is the blog post of how it was enforced in Germany:

      As to arguments against explicitly allowing reuse under the GPL, there are none that I know, it just doesn't seem to be a priority at the moment, but I might be wrong. It could be something that will be included in 4.0, it is still too early to tell.

      By the way, for some reason my filters are picking up your replies as spam, so apologies if it takes time to get them approved.

      • Thank you for the link! I think I even remember seeing that case in german on …

        It’s nice to see it work out!

        Next step: Get the viral nature of by-sa held up in courts.

        I hope explicit GPL allowance will move higher up on the agenda, especially since with wikipedia one of the largest bodies of by-sa content only got possible because the FSF did something similar (but much more far-reaching) for GFDL.

        (maybe it does not like my email address. Some spammers did put it into their reply-to address a while back, pissing me off quite badly, because I got all that “address does not exist” spam :( )

  3. The ports have another useful feature: They are written in the language spoken in the country that is targeted.

    The real license text. Not some inofficial translation. That lowers the hurdle for using the license.

    Of course "official translations" could be provided, but then they would need to be legally as binding as the English text, not just some second class version.

Leave a Reply