My concern with non-commercial and non-derivative licensed works is having someone now gate keep access to the work. This has a chilling effect on use as anyone using it now clearly knows the author didn't want their work used for anything outside of "showing to your friends".
Did I use the work when talking about it in a blog post? Am I in violation because there's advertising on it? Did I use the work to help make a video that gets a million views? Am I now in violation because I get a small income stream off of some other piece of work that used your non-commercial work on its periphery?
You say you want to target hobbyists. Say I'm spending my time investing in learning your library and I want to use it in my job because I think it's cool and I've gained familiarity with it. Now I'm in the position of asking you for permission of whether I can use it or not. My time would have been better spent investing in something that told me from the outset I had permission to use it.
You don't want people using your explicit approval, that's your prerogative and NC and ND are for you. If you want people to have control over the tools they learn, use and deploy, then it needs to be under a libre/free license.
It's more than aesthetics. Putting a non-commercial license on works, especially software, is a death knell.
I'm old enough to remember having this exact argument on dial up bbs's on why the BSD and MIT licenses were the only true free software licenses and the GPL wasn't. After all the GPL limits my freedom as a developer to not have to share all the source code I write as part of a program.
The argument is as wrong then as it is now: the freedom isn't for you, it's for everyone.
It doesn't help that that a bunch of squatters came into the commons and stole the words for free software and open source, those words were used long before the FSF and OSI stole them from the rest of us and they didn't mean at all what they were redefined to mean.
At least the FSF had the self awareness to realize that they needed a new word and tried for libre.
No, it’s not wrong. The GPL does restrict what a developer can do with code. It’s not an invalid argument. But like most/all arguments, how you interpret it depends on your point of view. Like you said, if you view freedom from the developer’s POV, you choose BSD/MIT. If you view it from the POV of the commons, you choose GPL.
It’s not wrong, but it is a choice.
But I also feel like we’re simply rediscovering why societies have laws, which are incompatible with perfect freedom. For example, is the GPL’s restriction sufficiently beneficial that we can overlook its abrogation of personal freedom? I’m not sure that’s widely agreed-upon even today, but clearly it is considered worth it to its author(s), as any good constitution ought to be.
We’ve discovered this week why France has a decades-old parasitism law, that allows you to recoup damages when someone else tries to build their fortune using your work and reputation. It serves the same intent as the AGPL attempts to - you can’t resell my work as if it’s yours without paying me - but it is written far more precisely, in a way that does not default-prohibit all commercial uses, but merely those uses that try to feed on someone else’s reputation. A licensing argument would say “then isn’t all commercial use parasitic?”, and the answer in nation-law terms is essentially “a technical analysis of the law might allow for that, but a reasonable analysis would not”. This is the crux where licensing and nation-law diverge: in licensing, to date, we have generally acted as if we must close every possible loophole because there are no judges in our system to evaluate ‘reasonable interpretation’.
So, to tie this back to the main point of the post above, if the GPL and AGPL (which appears to be longer than the French parasitism constitutional law!) are trading freedom for benefit, then they are not free, but they’re still trying to be beneficial to society. Have they made the correct tradeoffs in their currency swap of freedom for benefit? Well, I’m not sure. Similar restrictions in nation law would be intolerable to extreme libertarians, who would contest that all nation laws are unacceptable, but the rest of society seems comfortable with a reasonable set of restrictions.
Is the AGPL reasonable?
This is where the software licensing as legal constitutions model falls down, because we simply are not equipped to judge. We have one or two institutions that offer commentary and blessing, but our laws as written are evidence that we have no judge or jury able to provide reasonable evaluations and judgments. So the AGPL restricts people from building small ISPs just as much as it restricts people from building a multinational ISP. That wasn’t reasonable in a pre-software world, where you had to purchase objects to build things with them, but now that digital work can be reproduced and reused effortlessly, creators of source code no longer get paid for their source code (not that they did very often before!), and so the AGPL tries to fill that void.
Is getting paid reasonable?
Sometimes I think that we would have been better off tapping into music lyrics copyright and licensing, rather than trying to invent our own field of licensing from base metal principles. Lyrics writers get paid, though of course as with all things megacorps pay them poorly. Cover songs are allowed, but blatant knock-offs are not. Performances require permission, which may entail payment, and certainly does if not granted.
Music lyrics aren’t free for all commercial reuse, but they are free for reasonable commercial reuse. So if we try to bridge the gap between existing nation laws, and the ideals of the GPL/AGPL, a simple solution presents itself. One that provides the desired benefit of the AGPL without needing to create new restrictions on freedom. One that has a long and storied history of court judgements to provide human reasonable outcomes when the laws could never be precise enough anyways.
Source code is just lyrics to a song played on a computer instrument, and deserves the same protections and freedoms and limited reusability as any other song.
(I am not your lawyer, internet forum posts are not legal advice.)
This is where the GPL shines. It doesn't prevent commercial use in the text, it even IIRC explicitly allows it. You don't even have to think too hard to know if you're in violation (did I distribute and did I include source code?). OTOH the requirement to include full source code of derivative works under the same license clearly goes against a lot of commercial use cases.
I think the distinction between software and "content" is the utility of software. With content, distribution or viewing/listening is the primary purpose of it. I suppose showing GPLed source code in a video might raise questions of weather the video needs to be GPLed, but that seems like strange edge case (can we brush it off with "fair use"?) even most authors wouldn't care too much about.
This reads as "I don't like it, so how much hand wringing can I do?"
If the project doesn't provide a clear road to commercial use like MySQL did or SQLite(Public domain but sells licenses for places that don't recognize public domain) does, then it is hobby only. Don't bother picking it up if you are looking for something that you could use in the day job. Why? Because even if they will sell you a license they aren't setup to support a business.
I disagree. If you don't do that and Amazon kills your business model as a result, you now have to fire all your developers and abandon the software. Now that's a death knell!
I'm happy to use open source licenses for hobby software that I may or may not continue maintaining. For anything I hope to build a business around, I would never consider an OSI license anymore in this new cloud world. Amazon basically killed that open source business model.
There is the AGPL, but that isn't compatible with the majority of Free Software out there.
Maybe not, but such a license will stop me from including it into Debian or other distros that provide Free Software only, and I won't be able to use your code in or mix with projects under a copyleft license such as GPL. If that's what your intention is, then go for it. If not, maybe rethink.
Sure, it might not be useful to you, but you’re not the one putting in the labor in this hypothetical scenario.
> If that's what your intention is, then go for it. If not, maybe rethink.
Which is pretty much what I said as well, read more carefully.
So what?
This is a discussion worth having, instead of sweeping it under the rug with "well that's how we've always done it".
Also, it seems to me that "non-free" is already a good and descriptive tag for non-free source available software.
https://blog.wolftune.com/2011/07/brain-parts-song-video.htm...
The core point:
> The Creative Commons Non-Commercial (CC BY-NC) license (and also the BY-NC-SA license) substantially hampers the creation of non-commercial derivative works!
In short, it's because BY-NC is incompatible with BY-SA, and the latter is probably enough to discourage exploitative commercial use, anyway.
Duplicacy by Gilbert Chen https://github.com/gilbertchen/duplicacy
Dome by Federico Milano http://faraday1.ucd.ie/dome.html
Fman by Michael Herrmann https://fman.io
From what I've gathered, success has been mixed. Duplicacy seems to be doing well based on forum activity and release history. Fman never made much money. If I recall correctly the Fman author was turned off by the number of people who criticized it for being fully open-source and wished he'd stuck with a closed-source full commercial model. I'd like for the open-source community to be more accepting of this model as there is potential value in it. As an example, Mathematica still refuses to let anyone not employed by the company see the source code, which creates a reproducibility/validation problem for math researchers. While Steven Wolfram understandably rejects pleas to make the software FOSS, there is real benefit in letting people see the code.
Edit: Fixed names
The post you mentioned about being criticized for not being open source: https://fman.io/blog/dear-comrade/
The software wasn't open source but, as a practical matter, for solo developer projects it worked pretty much the same way.
Gilbert Chen is a different person to Raymond Chen, unless I'm missing something?
Less so in 2020s real life where we are a hair width away from smart dust total surveillance.
I agree with this. I'll go further: any software license, other than "do whatever you like with this work", imposes some restrictions on what you can do with it, and therefore is, in some sense restricting freedom.
However, the sense that it is restricting freedom is not a particularly useful one.
Restricting other people's freedom to violence means I'm free to walk the streets without fear of harm, likewise, restricting your freedom to close source and sell my software means your end user is free to know what's running on his/her computer.
That would reduce the usefulness of the word "freedom" as absolute good though.
They ensure that software licensed under the GPL (or similar license) grants users the four fundamental freedoms (https://www.gnu.org/philosophy/free-sw.html.en) and that those freedoms cannot be taken away by copies, forks, or derivative works.
If you want your software completely unrestricted, then the GPL is not what you want, but don't be surprised if someone creates a proprietary work using the completely unrestricted software.
The choice is yours.
Yes, the F.S.F. has decided that those four freedoms are the only fundamental ones to care about and calls it some absolute semantic of freedom.
Anyone can decree that one of those is not important, or add a fifth.
And even within the F.S.F.'s framework there are problems. The G.C.C. had to be given a licence exception for the runtime it links against, otherwise arguably anything compiled with it would become a derivative product of it, and thus subject to the G.P.L.. The argument is that that prohibits the freedom to run the software for any purpose; the F.S.F. has always said that writing a clause in a licence to stop software from being used to make proprietary software would make it nonfree.
So even there, the G.P.L. without such exceptions would create nonfree software if applied to compilers with a runtime.
>> Anyone can decree that one of those is not important, or add a fifth.
Freedom 0: The freedom to run the program as you wish, for any purpose
Freedom 1: The freedom to study how the program works, and change it so it does your computing as you wish. (Access to the source code is a precondition for this.)
Freedom 2: The freedom to redistribute copies so you can help others.
Freedom 3: The freedom to distribute copies of your modified versions to others. (Access to the source code is a precondition for this.)
Which freedoms do you disagree with and why?
What freedoms are missing that you would like to see added?
>> The G.C.C. had to be given a licence exception for the runtime it links against, otherwise arguably anything compiled with it would become a derivative product of it, and thus subject to the G.P.L.
You are conflating GCC as free software with software created using GCC as a compiler.
The runtime exception has been in place for quite a while and is part of the reason why GCC is very widely used.
I would argue that the four freedoms are also part of the reason why GCC has been so successful.
> What freedoms are missing that you would like to see added?
I am not so small a mind to be concerned with semantics debates and the definition of “freedom”.
Purely in the sense of what I personally desire, as in rational egotism, I would wish for all software to exist for me free of charge, with all source code available, with a copious options system, and that it be written in a programming language I am competent in and follows the style guides I find easiest to reaed.
That is a wish, not a moral dogma; moral dogmata are Stallman's domain and I consider any man who would seriously believe that things can be morally right or wrong to be of quæstionable rationality.
> You are conflating GCC as free software with software created using GCC as a compiler.
I'm doing no such thing; I'm simply pointing out that the G.C.C. would probably not be free software without the exception.
> The runtime exception has been in place for quite a while and is part of the reason why GCC is very widely used.
> I would argue that the four freedoms are also part of the reason why GCC has been so successful.
I'm fairly certain it's the biggest reason is simply that it's a very good and gratis compiler.
Lots of companies, for example, extract value out of Linux without thinking much about the GPL. Because the "linking" bit never comes into play. On a smaller scale, GnuPG is similar, because the default interface is pipes and not a linked api.
>Copyright law today grants software developers that power, so they and only they choose the rules to impose on everyone else—a relatively small number of people make the basic software decisions for all users, typically by denying their freedom. When users lack the freedoms that define free software, they can't tell what the software is doing, can't check for back doors, can't monitor possible viruses and worms, can't find out what personal information is being reported (or stop the reports, even if they do find out). If it breaks, they can't fix it; they have to wait for the developer to exercise its power to do so. If it simply isn't quite what they need, they are stuck with it. They can't help each other improve it.
A non-commercial license still covers all of these "freedoms." The only caveat is that you can't sell it or derivatives thereof.
What do you mean by that? Aren't "permissive" and "copyleft" opposites?
That said, sharing your code to be used in certain ways, not others, or for purposes you hate is maintaining control. That is relevant to "is it free?" I don't think it's morally bad, unless you take an absolutist stance that all software must be free.
To the normative point... As more of the software industry is concentrated in monopolies, the benefits of Free Software tend to concentrate there too. At that point, it's harder to feel good about it and sharing your code selectively often feels better. I also think that server side software breaks some of the Free Software logic. Users of the software aren't technically users, and they don't inherit Free Software's rights. Again, this makes the normative point bland and unattractive.
It isn't morally bankrupt but it is in contradiction to "share my code freely". Libre sharing of code with others invites contributions, which may or may not take the form of commercial enterprise around the fringes and/or of the code in-question itself. Think WordPress and Linux/RedHat; and not Redis and ElasticSearch. I agree that companies like Docker that created real value couldn't capture it all for themselves, but that's down to the rest of the ecosystem catching up to them and eventually overhauling them (some would argue, this is how it is supposed to work). If it is any consolation, they did end up creating an enormous amount of value with an industry changing innovation, but that happened on back of permissive licensing.
You cite Amazon as an example, but as one counter, AWS now employs the RTOS lead developer and continue to fund the effort. Something similar happened when Rust-lang engineers were let go by Mozilla. So, it isn't all bed of nails.
I hack on FOSS full-time, and in my view, favouring permissive licenses over restrictive ones is how the FOSS ecosystem flourished in the first place. Besides, unlike other forms of work, digital work inherently wants to be free. There's no caging it. And in letting it be is how we let it grow.
Is it? Can you truly support this argument?
I can't work on open-source code full time without getting paid. I have bills to pay. If I release a project, try to make some money on the side, and Amazon swoops in, takes all my code, and offers it for less than I can afford, I will have to stop working on it.
Does the software still grow now?
On economics of software, these posts make for a good reading:
https://www.joelonsoftware.com/2002/06/12/strategy-letter-v/
http://dtrace.org/blogs/bmc/2004/12/16/the-economics-of-soft...
On FOSS and permissive licensing, see these:
https://blogs.apache.org/foundation/entry/the-apache-way-to-...
http://dtrace.org/blogs/bmc/2018/12/14/open-source-confronts...
> Is it? Can you truly support this argument?
The word "can" is doing a lot of heavy lifting in this sentence.
Does it do that, in practice?
That scenario is clearly wrong and absurd. You don't need a license to use your own intellectual property, and thus cannot be bound by that license in any way.
This is a Very Bad Thing because it keeps really good software from making it into commercial use over fear of the restrictions in the license. (APGL bans are great example.)
Yes.
This is where the AGPL (at least for code) comes into play, which mostly is enough to prohibit Amazon from taking your code and hosting (a modification of it) on their servers.
So yes, that is apparently already thing now and distinct from "free" or "open source".
[1] https://perens.com/2019/10/12/invasion-of-the-ethical-licens...
[2] https://perens.com/2019/09/23/sorry-ms-ehmke-the-hippocratic...
Look, don't pretend that you don't know that the FSF doesn't exist, or that that is what is conventionally meant by Free Software.
Yes, English does have plain meanings for the word "free". However, you're not really doing yourself any favors here. You just sound like you're being willful or stubborn about the term Free Software. I agree that it's a poor term, but it's the term that has been selected. Complaining now in 2021 about computer terms devised in the mid 1980s is fruitless at best.
> From a developer's perspective: Why should code be 100% Certified Free™ instead of 95%, or even 50%? Is it a moral imperative to make code truly free, or an aesthetic ideal? What harms come from a CC-BY-NC license relative to MIT and why do those harms matter? I think answers to these questions are much harder and more interesting to come by than splitting hairs over the semantics of freedom.
Okay, so, if you look at the FSD link in the article [0], the very first thing is the four software freedoms:
> - The freedom to run the program as you wish, for any purpose (freedom 0).
> - The freedom to study how the program works, and change it so it does your computing as you wish (freedom 1). Access to the source code is a precondition for this.
> - The freedom to redistribute copies so you can help others (freedom 2).
> - The freedom to distribute copies of your modified versions to others (freedom 3). By doing this you can give the whole community a chance to benefit from your changes. Access to the source code is a precondition for this.
The second thing in the FSD article explains why NC isn't Free:
> “Free software” does not mean “noncommercial”. On the contrary, a free program must be available for commercial use, commercial development, and commercial distribution. This policy is of fundamental importance—without this, free software could not achieve its aims.
> We want to invite everyone to use the GNU system, including businesses and their workers. That requires allowing commercial use. We hope that free replacement programs will supplant comparable proprietary programs, but they can't do that if businesses are forbidden to use them. We want commercial products that contain software to include the GNU system, and that would constitute commercial distribution for a price. Commercial development of free software is no longer unusual; such free commercial software is very important. Paid, professional support for free software fills an important need.
> Thus, to exclude commercial use, commercial development or commercial distribution would hobble the free software community and obstruct its path to success. We must conclude that a program licensed with such restrictions does not qualify as free software.
In other words, your open source, NC software will never achieve success because it's chosen to exclude one of the major causes for software success and distribution: commercial adoption. By extension, you've told all commercial systems to go build their own proprietary system and sell that instead. You've forced commercial systems to compete with your software. You've created a monster, because commercial enterprise is always going to have more resources than you do. You've ensured that if your software's purpose is necessary and important, your software will fail.
Amazon can't "steal" something that is given freely. That's the whole point of free software: it's no longer up to you how others get to use it.
The core idea of copyrights and ownership of sequences of bytes is the problem. We need to stop thinking of information as something that it is possible to own.
Or at least a labourer may legitimately claim they wish to share "freely" only to some. And they can also choose to change their mind on any subsequent work.
This is especially the case when the entities involved are corporate, not individual humans.
But this is the deal you sign up for when you make free and open source software. If you want to make another kind of software, that's quite fine, but it's not free software nor open source.
Would the software have attained the userbase it did if they were the terms from the outset? A lot of these companies just assume that it would have been the exact same except for the absence of an amazon competitor if these licenses were used.
"I would feel a lot more comfortable with this if it didn't collect any data of mine"
I have pirated movies on my hard drive. Do I own them?
Sure. But until we reach that world, software licenses have a place.
However, don't all (or nearly all) licenses imply restrictions of some sort? Do you think code should just be released CC0/Public Domain or are certain restrictions (like copy-left) worth the cost?
As far as land goes, you're right. The same rules for objects probably shouldn't be used for land.
I'm not terribly convinced this legalistic nitpicking is really the right discussion though. I think the real questions here are normative. If I want to share my code freely with hobbyists and individuals, while ensuring that Amazon can't just steal it, pervert it into something I hate, and then borrow my own reputation to help them sell it, is it morally bad to add a license that stops this behavior?
From a developer's perspective: Why should code be 100% Certified Free™ instead of 95%, or even 50%? Is it a moral imperative to make code truly free, or an aesthetic ideal? What harms come from a CC-BY-NC license relative to MIT and why do those harms matter? I think answers to these questions are much harder and more interesting to come by than splitting hairs over the semantics of freedom.