3 ms·
There have been licenses that do this same kind of stuff before. There have been dozens of variations of FOSS licenses over the years that prohibit things lik
by kube-system 5y ago
There have been licenses that do this same kind of stuff before. There have been dozens of variations of FOSS licenses over the years that prohibit things like use for nuclear facilities, use in weapon systems, use in genocide, etc.
That's not to say I think any of this is a good idea. I think it's insanely naive to try to contractually block a dictator from using your software in the first place. Do these people think they're going to sue the dictator in the dictator's court? The Russian government is already considering legalizing piracy of proprietary software to evade sanctions; they DGAF about these licenses.
I mostly just think that the OSI and FSF's "all or nothing" stance on some topics is a bit extremist. A hypothetical license that said "You can do anything with this software except launching missiles at kittens" is pretty damn in line with the spirit of FOSS, even though the OSI or FSF would never approve it. I understand they take that ideological angle because they are thought leaders and need a clear message, not a slippery slope, but this is where I think it's reasonable to be reasonable.
I think it's better to criticize these types of "weaponized" FOSS licenses by pointing out that they're entirely ineffective wastes of time. There are plenty of real merits to undermine these licenses, which is why they haven't caught on previously. Saying "that's not open source" is just like a chocolate snob telling their friend that their Hershey's bar is "totally not real chocolate"
That being said, we're beyond the realm of software licenses in this conflict anyway. Sanctions have already started cutting people off from FOSS projects, regardless of license.
- the_af 5y ago> There have been dozens of variations of FOSS licenses over the years that prohibit things like use for nuclear facilities, use in weapon systems, use in genocide, etc. I'm not aware of any FOSS license that states things like that. I think you're wrong. > A hypothetical license that said "You can do anything with this software except launching missiles at kittens" is pretty damn in line with the spirit of FOSS No, it's not in line with the spirit of FOSS, but rather against it. I understand that you have this vision about a kind of license you would like for software. I cannot argue with that -- it's your vision and it's your right to have it. The problem is that it's not any kind of FOSS license that is recognized as such (i.e. as FOSS) by the wider community. You can call it FOSS but nobody else will.
- kube-system 5y ago> I think you're wrong. Here's examples: https://spdx.org/licenses/BSD-3-Clause-No-Nuclear-License.html https://spdx.org/licenses/BSD-3-Clause-No-Nuclear-License.ht... https://spdx.org/licenses/BSD-3-Clause-No-Military-License.html https://spdx.org/licenses/BSD-3-Clause-No-Military-License.h... https://spdx.org/licenses/Hippocratic-2.1.html https://spdx.org/licenses/Hippocratic-2.1.html > The problem is that it's not any kind of FOSS license that is recognized as such (i.e. as FOSS) by the wider community. You can call it FOSS but nobody else will. There is evidence to the contrary. There are popular community supported projects that are often colloquially described as open source which use licenses that are not OSI/FSF.
- the_af 5y ago> There is evidence to the contrary. Your evidence is incorrect. None of those are FOSS licenses. If you google their names, you'll see each sparked controversy, and ultimately the SPDX doesn't exclusively list FOSS licenses. Ultimately they infringe on the most basic right: to use the software for whatever you want. > There are popular community supported projects that are often colloquially described as open source which use licenses that are not OSI/FSF. Nobody cares about "colloquially". Colloquially, Microsoft in the past tried to pull a fast one and pass their Shared Source initiative as "open source", which it wasn't. FOSS means something. None of the examples you gave are FOSS.
- kube-system 5y ago> Your evidence is incorrect. None of those are FOSS licenses. I said “variations of FOSS licenses”, which is precisely what I provided examples of. > Ultimately they infringe on the most basic right: to use the software for whatever you want. Unless of course, “whatever you want” includes not publishing your changes. Copyleft does not permit people to do “whatever they want”. It places specific requirements on people who use that code and punishes them under penalty of copyright law if they do not. You’re touching on the reason that some people prefer public domain waivers instead of copyleft licenses. If you want an example of disagreements in FOSS: case in point. > FOSS means something. None of the examples you gave are FOSS. There are factually multiple definitions of what falls under the umbrella of FOSS. FSF and OSI are certainly prominent. There are other individuals and organizations who disagree. I think it’s kind of funny that people want to argue that the majority opinions on licensing are the only valid opinions on software licensing. That’s exactly how people dismissed RMS in the 80s and 90s.
- acoustics 5y ago> Saying "that's not open source" is just like a chocolate snob telling their friend that their Hershey's bar is "totally not real chocolate" I don't think this is a great comparison. Open source has a particular, commonly understood meaning, just like chocolate does. When I say "this is not an open source license", I am not saying that it doesn't meet some kind of quality standard, or that I just don't like it. I am saying that it is a deceptive label that contradicts the common understanding of the term. If an author would like to relicense their open source project to exclude certain groups of users, why must they insist on using the open source label? They ought to be proud that they reject open source norms. Why would they want to associate themselves with projects allow dictatorships to use them? I have written software with a license that allows for noncommercial use only. I label it as freeware (which clearly communicates "you need to check the license"). I have no desire to label it as open source, because I do not wish to mislead people who have a coherent and widely accepted mental model of what that entails. I am generally suspicious of the motivations of people who would want to muddy the waters with unclear labeling. There is a perfectly honest way to address this, as esr did: invent a different term. If you can come up with a short, snappy term for "source-available software that allows for open-source-like use for certain classes of people", then use that, and don't lie to your users.
- kube-system 5y ago> I don't think this is a great comparison. Open source has a particular, commonly understood meaning, just like chocolate does. I’m using that example because there is disagreement about the definition of chocolate. Here in the US there are some things called chocolate which are not called chocolate in other parts of the world. How much cacao is required? How much milk solids are required? You will get different answers from different organizations. All of which, claim to be an authority. The FOSS landscape is no less opinionated.