- Some Updates from Jun 7, 2020, 8:08am
As some people have been arguing over the so-called “ethical licenses” such as the Hippocratic License or MongoDB’s Server Side Public License, I don’t find that line of discussion particularly useful, but have started thinking more about what I’d like to see in a modern Free Culture license.
To be clear on my disinterest in the ethical licenses, I sympathize with the desire, but don’t believe that a so-called “liberal” (in the sense of fewer terms, not social philosophy) software license is a useful way to enforce “good” behavior and that the whole space opens the door for regressive moralizing attached to code. And that doesn’t even get into the strange compartmentalization that we often see outside of software, where philanthropic initiatives often enable and obscure the most unethical corporate behaviors. I would much rather see ethics become a part of the software industry instead of constantly disclaiming responsibilities. If a developer has the toolkit to refuse a task—not just because that work will hurt people, but because it’ll damage the developer’s career prospects—that’s a real solution.
That all said, this may never come to anything and I have no interest in pushing anything here, I’d like to kick around some ideas for a hypothetical Free Culture license of the future.
As I mentioned early on in the life of this blog, the easiest way to envision a public license working and the easiest way to make it function in a court of law is to treat it and write it like a contract: If the recipient provides the consideration specified in the terms of the contract, then they receive a consideration in exchange, a trade of value.
Too many licenses—particularly the “liberal” licenses—don’t have any mutual value exchanged. The code is given…and that’s it. And I don’t think that makes them defensible in court, if it ever came to that.
For thirty years, we’ve watched organizations create licenses and then turn around to tell us about all the kinds of works that their licenses should not be used for.
The Free Software Foundation’s original position was that we should only worry about software running on our personal, general-purpose computers. Art and text weren’t interesting. Firmware wasn’t interesting. Hardware wasn’t interesting. Servers weren’t interesting unless you were running them. Even high-level software wasn’t interesting, as long as it ran on a dedicated device where the users couldn’t easily run arbitrary software of their own. Some of those deficits became problems—TiVo’s use of Linux triggering the GNU GPL version 3 and the rise of interactive web sites triggering the GNU Affero GPLv3—while other issues are still ignored as someone else’s problem.
Likewise, Creative Commons still firmly recommends against using their licenses for software. And there are unrelated open hardware licenses. And fonts. And games. And databases. There are others, too. Many others.
What Is “Source Code,” Anyway?
One of the reasons I lean so hard on the phrase “Free Culture” is that, while we tend to refer a lot to “free and open works” (for example, see FaOW, a catalog of such content), the idea of “open art” is…odd.
For an easy example, Bicker is software, so its source code is easy to find. That’s why the term “open source” makes some sense, here.
However, now take Seeking Refuge, my novel. I make the Markdown files available and have an entire section describing the inspirations and sources of many of the elements that contributed to the creation of the story. Is that “open”? Have I shown my “source code” for the novel? There’s an entire cultural context and life-long education that would make it difficult for someone to reproduce the final story.
And now let’s take a step further and consider a photograph. Hang on, do I have a photograph that I can use as an example, here? Sure. Here.
OK, that picture is now released, just like the rest of the blog, under a Creative Commons Attribution-Share Alike 4.0 International License, just like it says at the bottom of the page. I would call it “Free Culture,” to the extent that the picture is culture, but it’s not likely anybody’s definition of “open source,” because…there really isn’t any “source,” except if you count the full-resolution version from the camera. I was walking and, since I was carrying my camera with me, took a picture when I saw that angle, because I liked it and am pretty sure there are no copyright entanglements in showing just the bricks and foliage.
So, the moral of that little story is that a universal Free Culture license needs to understand that distinction and require as much “source” as possible without requiring something that might not exist.
Strong Copyleft and Reciprocation
To get the definition out of the way, copyleft is a term used to contrast with copyright, typically to imply that a consumer has been granted limited rights over the work, provided that they extent those same rights to others. In other words, I share this blog with you and grant you the right to edit it and share it with others, if you want, but only if you give them the same rights as I’ve given you.
I tend to suspect that most people who choose non-commercial licenses and even the ethical licenses are really looking to not be exploited. In that sense, I tend to agree with Question Copyright in the analysis that copyleft prevents exploitation more efficiently than commercial restrictions: Disney might easily roll the accounting dice to claim that their motion picture adaptation of your short story is technically a non-commercial use, but they will not put a public license on their movie for other people (and companies) to use.
Similarly, I suspect that we can solve a lot of (though not all) Hippocratic-style issues by requiring that a derivative work publicly contributes its modifications back to the source project, assuming that they’re going to abide by the terms of the license at all. Even if they’re not ashamed to admit to what they’re doing with the code, contributing back gives the source project the ability to shame them in the public sphere.
So, I’d like to see the license not only require making changes available, but offering them back to the parent.
Clear Attribution Terms
The MongoDB problem, as they’ve stated in various places, is a fear that an Amazon-like company will take their free product and sell access to it as part of a larger service like Amazon Web Services.
However, this seems to be the opposite of the intent of releasing a database management system at no cost. Hence, they relicensed their product to a custom license that is largely contradictory in that you can use the product for any purpose, but not for some purposes.
However, this is a problem of competition and monopolization. However, if we replaced the SSPL with a strong copyleft license requiring reciprocation (as in the previous section), then Amazon (or whoever) is required to contribute back the changes they make to more easily make the database available, so the database company knows about this “threat” (assuming it’s a threat) and can compete on technical terms. And to take this a step further, if you can’t sign up for maybe-Amazon’s hypothetical not-MongoDB service without seeing a clear link to MongoDB, who can then try to offer the same service for a competitive price, then maybe-Amazon no longer has the feared unassailable monopoly position.
There’s a related but distinct problem when it comes to physical objects. If I download a 3D model for the above book ring and print one on a 3D printer, the license requires attribution. That’s fine, but…where does one put legible attribution with a URL to the designer’s profile on an object so small, where the largest spaces should be smooth? The object has no attribution attached, so it technically can’t be used in a way that complies with the license without making changes. Licenses unfortunately ignore those issues, right now.
Clear Patent/Trademark Guidance
Mozilla often comes under fire for not wanting their brands associated with compiled binaries for which they can’t manage quality. This is technically true under most Free Software licenses (the GPL has an explicit option to refuse to “grant rights under trademark law”), but people only seem to notice it when certain organizations and licenses are involved.
Patents are a similar issue, with some licenses specifically making patent grants and others…not so much. For example, the GPLv3 says:
When you distribute a covered work, you grant a patent license to the recipient, and to anyone that receives any version of the work, permitting, for any and all versions of the covered work, all activities allowed or contemplated by this License, such as installing, running and distributing versions of the work, and using their output. This patent license is nonexclusive, royalty-free and worldwide, and covers all patent claims you control or have the right to sublicense, at the time you distribute the covered work or in the future, that would be infringed or violated by the covered work or any reasonably contemplated use of the covered work.
If you distribute a covered work knowingly relying on a patent license, you must act to shield downstream users against the possible patent infringement claims from which your license protects you.
The Apache license works similarly.
Subject to the terms and conditions of this License, each Contributor hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable (except as stated in this section) patent license to make, have made, use, offer to sell, sell, import, and otherwise transfer the Work, where such license applies only to those patent claims licensable by such Contributor that are necessarily infringed by their Contribution(s) alone or by combination of their Contribution(s) with the Work to which such Contribution(s) was submitted.
All this is to say that a good license should at least be clear about the rights under patent and trademark law, since software carries aspects of all forms of intellectual property.
As a parallel example, one of my concerns with using Creative Commons for Seeking Refuge is exactly my trademark problem: In theory, I would like a trademark on The League of the Silver Bat, as a franchise name, to protect against a hypothetical large media company from using the name and turning around to sue me when they find out that I exist. However, doing that would require that I sue the people I encourage to create derived works if they use the name prominently, to prove that the trademark is mine, which defeats the purpose of sharing.
Absent an “open trademark” license (I see a draft of one in Japanese, but I haven’t seen any in actual use), I would minimally want my copyright license to include a trademark grant in the context of the derivative work, possibly requiring some variation on the trademark to distinguish the derivatives from the original; likewise, nobody accepting the terms of the license should be able to sue me over those trademarks. There are no patents in this example, but the situation seems analogous.
One of the things I like least about Creative Commons is their continual treatment of non-commercial licenses as first-class citizens of their system, since it creates two bodies of mutually unintelligible works. This isn’t “I want to use Pioneer One as part of a space opera franchise” (even though that would be an excellent idea for someone willing to work with the non-commercial license), but rather—as a brief example—they can’t adapt content from Stardrifter and Stardrifter can’t adapt anything from Pioneer One, since each license requires perpetuating terms that contradict the other license.
There is a similar problem in the relationship between Creative Commons licenses and the GPL. There is currently an agreement in place carving out exceptions for GPL-licensed software to use content licensed CC-BY-SA, but the inverse isn’t true. Therefore, apart from Fair Use considerations, it’s not feasible to include source code from a GPL-licensed project in a Creative Commons-licensed story or (in some cases) even use GPL-licensed software in a Creative Commons-licensed video. I don’t include screenshots of software in this blog (except for my own) for this reason.
If you’d like a more concrete example of GPL/Creative Commons conflicts, consider that most of the books on FLOSS Manuals are licensed under the GPL in the interest of being able to integrate the documentation with the software, at some point. But because of this, it can’t be integrated with other content under strong-copyleft licensing (I’ll continue to use this blog as an example, though Wikipedia and related projects come to mind as stronger contenders) without revising the license to the GPL.
So, a modern license needs to be aware of the needs of other licenses, so that incompatible terms can be supported on at least some limited basis. For example, if I was creating a movie (I’m not) where the characters use a lot of Free Software, it would work well enough if the software couldn’t be modified for the purpose and marked as licensed separately. Likewise, I wouldn’t mind if my license for Seeking Refuge included terms that allowed for people to use extended excerpts, regardless of their license, with tighter integration than that requiring tighter conformance to licensing terms. Better would be if its terms were flexible enough to integrate GPL and CC-BY-SA content and reciprocate instead of being a one-way (or no-way) transfer.
Contributor License Agreements are terms required for contributing to a project that cede all rights to the owners. Generally, these are used for companies that want to eventually change the license of software under the GPL, eventually taking later versions proprietary once the market has been proven. Sure, there are other theoretical arguments about why a company might impose a CLA on people giving them gifts, but in practice, it’s hard to find an exception.
There’s a related but lesser issue that, as the holder of the original copyright, up until people contribute to my projects, I am not bound by the terms of their licenses. I can (don’t worry, I won’t, but I could) write a series of blog posts—licensed as CC-BY-SA—that are just SlackBackup code, whereas you can’t, because you have received the software under the terms of the GPL.
I’m not sure if the latter problem can ever be solved, assuming we didn’t just abolish copyright entirely. However, a policy of “copyleft in, copyleft out,” suggesting that code contributed under the terms of the license can’t be re-released under other terms might handle both. Bear in mind that the latter problem is always going to have a weaker solution, just because the entire point of copyright is a limited-term monopoly on what you create. But, it’s still a start and would scuttle any attempt to impose a CLA outright.
Digital Rights Management
This is, unfortunately, a much more dicey issue than the others. Digital Rights Management are “locks” on content, designed to ensure that access to the content is exclusive. DVDs, for example, are (trivially, these days) encrypted to make it more difficult to watch a movie on your PC where you might copy it. Most streaming services use Encrypted Media Extensions, another form of DRM, to similarly make it difficult to consume the book/music/movie/whatever on any device at any time.
Because it takes research and development to implement these locks, prices go up. And, as you probably know, they don’t really work and it’s really easy to find just about any current property on some pirate distribution network, if that’s the sort of thing that excites you.
As has been said many times before in this context (I believe originally by Cory Doctorow, but can’t find the quote), nobody wants to pay more for a product whose distinguishing feature is that it’s harder to use.
So, I’m not at all sure what form I would like this to take, but at least the option of forbidding distribution of a work through restricted channels seems well in the spirit of this. Selling a work is fine. Expecting that sold work to be locked to a specific device or user, however, is fantasy, and (if nothing else) creators should have a say in what happens to the works created using their work.
One of the biggest objections I’ve heard regarding the GPLv3 is that it clocks in at 5,644 words, long enough that few developers have read it (almost twice the length of this interminable post) and, therefore, most developers don’t actually know what terms they’re offering or agreeing to, except in the broadest sense. By contrast, the MIT License hits around 170 words and, while it’s a weak license, there’s a better chance that you’ve read that than the GPL.
While a lot of the difference obviously involves added features, a modern license should also be laid out in a way that encourages reading and comprehension, ideally also including a summary like the Creative Commons “deeds.”
It occurs to me that I don’t think that I have ever seen a public licensing statement attached to a public license. In fact, the GPL’s authoritative copy has the following line just before the Preamble.
Everyone is permitted to copy and distribute verbatim copies of this license document, but changing it is not allowed.
In other words, we’re not allowed to create derivative works based on the GPL.
By contrast, the Creative Commons licenses, such as CC-BY-SA, have a section at the end including:
The text of the Creative Commons public licenses is dedicated to the public domain under the CC0 Public Domain Dedication.
I’m not saying that either is better or that this is the most critical thing in the world, but I’m actually the tiniest bit uncomfortable using the GPL, now, because I’m distributing a copy with my software without an explicit, formal public license. The way I read the disclaimer, it would be forbidden to make an audio recording of the license, since that’s not a copy of the document, but rather an interpretation.
Obviously, you shouldn’t name your derivative license after the original (see the foregoing about trademarks), but the idea that licenses don’t have licenses is…odd. It won’t change how I work, but it’s something I would like to see resolved.
So, keeping in mind that nobody appears to be working on a project for a next-generation copyleft license and that I’m just some guy who happens to use copyleft licenses, that’s a decent outline of what I would like to see if those conditions weren’t true: Written like a contract, not written for a specific kind of product, strong reciprocation requirements as compensation for work, clear attribution guidelines, patent- and trademark-aware, not adversarial towards similar licenses, adversarial towards CLAs and DRM, easy to understand, and available (ideally) under its own terms. I think that would resolve a lot of current problems and set the community up for better collaboration in the future.
Would we see adoption of a license like this among big corporations? Probably not, and that may be a problem for some developers. But as I’ve mentioned previously, driving out the companies who use Free Culture as a source to be harvested without compensation is a benefit to me, not a drawback.
Does this solve the problems that the “ethical” licenses seek to solve? Not directly, no, but I also don’t think there are many people and organizations who are willing to violate human rights and commit war crimes, but are going to balk at violating the terms of a license that may not be binding. It would be like imagining that we can stop murders by reducing the speed limits in dangerous neighborhoods to make it harder for the murderer to escape, because nobody ever speeds.
Don’t get me wrong. I appreciate the sentiment behind these licenses. But I think the better approach is to make it unacceptable to violate human rights, rather than to release something under a “liberal” license and idly wag a finger at organizations that enables arbitrary arrest. And there are certainly ways to do that, but when the pitch for a new license includes a lengthy discussion about how nobody cares about licenses, the new license isn’t much of a protest.
And if you’ve read all the way down this far, this is obviously a field that interests you, so…what features do you want in a license?
Credits: Untitled header photograph from PxHere, made available under the CC0 1.0 Universal Public Domain Dedication. The book ring by Frank Erfurth is available under the terms of the Creative Commons Attribution-Share Alike 3.0 license.
Tags: license meta opensource freesoftware gpl creativecommons copyleft
Sign up for My Newsletter!
Get monthly (or thereabouts) notifications on Entropy Arbitrage
posts, additional reading of interest, random thoughts, and previews
of upcoming projects, delivered right to your inbox. I won’t
share your information or use it for anything else.
You can view previous issues.
If you disable trackers (like I do), this form won’t work, so you’ll need to sign up on Mailchimp's site. Sorry!