1. 14
  1.  

  2. 24

    Been waiting for someone to solve ethics in a single legally binding paragraph. Nice.

    1. 8

      FWIW. This reads like sarcasm, and good use of sarcasm at that.

      It is thought provoking instead of pushing a preprocessed view on the reader.

    2. 21

      This license is, verbatim, the MIT license plus this clause:

      The software may not be used by individuals, corporations, governments, or other groups for systems or activities that actively and knowingly endanger, harm, or otherwise threaten the physical, mental, economic, or general well-being of individuals or groups in violation of the United Nations Universal Declaration of Human Rights.

      Setting aside argument/discussion about the human rights bit, I think that it’s noteworthy that “no warranty” clause is still there.

      I’m kinda curious what train of thought allows developers to claim moral authority about how their software may be used while at the same time denying any responsibility for the quality/reliability of that same software.

      1. 25

        “I made this thing. I can’t promise it’s perfect. If you want, you can use it as long as you’re not hurting anyone.”

        I don’t see the problem (from a moral standpoint). If you don’t like the terms, don’t use the software.

        From a legal standpoint, I’m unclear on whether most jurisdictions allow you to do disclaim responsibility while dictating use. AFAIK that’s largely untested waters.

        1. 8

          Copyright licenses are legal documents. The legal enforceability of the terms is a valid question to ask.

          1. 1

            It’s valid, but in practice, if a court is involved in picking over a license then the license has already failed at its task (preemptively avoiding certain behaviors by making the kinds of people who want to engage in them turn up their noses at the software on the grounds of license terms).

            This is already sufficiently more specific than the JSON license that the people the license is supposed to scare off probably genuinely will be scared off. I’d like it if they struck the ‘underprivileged’ part so that there wasn’t the glaring loophole of having to prove privilege – just make it “no physical, mental, or economic harm to persons” (i.e., it can’t be used commercially or in a military or law enforcement context, because regardless of the ‘greater good’ arguments justifying these things existing, there’s no doubt that individuals are routinely harmed by them).

            1. 1

              I’m 100% sure tons of companies are unwittingly using free software with licenses they are technically not allowed to use. Their devs are just concerned with shipping, they don’t really care about the terms. And when they find out, they put migration away from them on the relevant teams’ backlogs and forget about it. If they’re taken to court over it, what do you think their lawyers will do, just say ‘mea culpa’ and incur penalties? No way. They’re going to incur billable hours and fight the licenses.

              1. 2

                In any kind of licensing issues I’ve been in, they said “mea culpa” and incurred penalties. I know people have a bad view of lawyers, but companies that have their house in order wrt. to lawyers just agree on mistakes and pay up. Lawsuits also incur organisational costs and there’s not a huge interest there.

                To be quite frank, if your lawyer never tells you “don’t pick that fight”, he’s a bad lawyer.

                1. 1

                  Sure. At the same time, a lot of businesses have over-broad policies about what licenses are acceptable (based, pretty much, on the same metric: generalizations about licenses that don’t really involve reading them).

                  Where I work, permissive licenses are fair game, but using anything GPL’d or under a GPL-like theoretically involves getting permission from the legal dept on a per-package basis (more effort than even licensing a proprietary thing for millions) simply because they’re afraid that we might have to make modifications & be forced to distribute those modifications (thus giving away work the company paid for), and any non-EULA license not on their list is supposed to be subject to serious vetting (which functionally means: either we steer clear of it entirely or we don’t tell anybody, because if legal gets involved we won’t be able to touch the project for at least 18 months).

                  In other words: litigation is not where licenses have power (and even pretty well-established licenses may not stand up to serious litigation). Adoption is the point at which licenses exert their power – by scaring people off. This has nothing to do with whether or not they’d be legally effective, and everything to do with whether or not non-lawyers think they could justify the worst case scenario.

          2. 3

            So, a more verbose JSON license.

          3. 10

            I’m a bit tired, so forgive if this will end up more of a stream of thought.

            This license might be practical or not, but there’s some interesting aspects to this:

            Compliance with the license is the users concern. That might still lead to them decide that it doesn’t apply to them and you end in court, but they have to check. If this term is confusing/ambiguous to them, they can always ask the project for clarification or a written statement that their usage is okay. Essentially, it’s the Lamson license by a different phrasing: https://github.com/zedshaw/lamson/blob/master/LICENSE#L21-L22

            What we are seeing here (and in other things getting somewhat popular lately, like the parity licenses or commercial use licensing), and this is not to be understated, is a fundamental shift. People don’t feel like OSS or FOSS are categories matching their wishes anymore. That might be monetary (someone else is making money of this), moral or both. It’s not unsurprising that these pushes come from a cluster of people that have grown into the community around 2010 and have built structures. They literally don’t agree with the status quo that (F)OSS licensing has.

            Also, the surroundings have changed: while the discussion 15 years ago was “some router manufacturer is using free software and doesn’t give access to the firmware image”, nowadays, we’re talking about “people are using FOSS software to build aiming software for auto-aiming sniper rifles”. I think it’s easy to agree that these are questions from different moral categories.

            I don’t want to take a stance on the statements above, because I think there’s a second problem here: the FOSS community values uniformity over everything. That plays out in weird ways: either, everyone has to agree that it’s fair to use FOSS for everything (the libertarian stance) or that it is a very moral thing (the ethical stance). Those can be rather easily unified, once the question “my software is used to kill people” (and lesser forms of it) doesn’t come up. They are mutually incompatible if it is. The unification of the FOSS communities was needed in the early stage of the movements to make sure you have power. Nowadays, even small community orgs have the size and support networks of the early FSF. This truce is cancelled.

            The second part is that we don’t talk about the stances of community leaders as much as we should. Because that makes the debate very interesting. “FOSS software is used as firmware for Sniper rifles” is my favourite example here, as people like ESR, outspoken gun advocates, would clearly have a positive reaction that - probably even be proud of their work. They will actively work on making this possible. Even without taking a personal stance here, it is clear that this view is incompatible with at least a substantial subset of contributors to e.g. Linux. The same goes for views on surveillance, privacy, monetization etc.

            The hack around this was to consider “the tech” neutral. I predict that this doesn’t work anymore. We’re going to have more and more clusters. FOSS vs. OSS will become an uninteresting distinction there.

            Interestingly, I assume that we will see more and more “source available” licences with rules attached (I would consider this one one). We might also end up in a situation where license handling will become more and more messy again. I also hold the belief that this is a failure of both the OSI and FSF, having ignored fundamental shifts in opinion.

            1. 3

              having ignored fundamental shifts in opinion.

              But, what do you expect them to do? The goal of the FSF is to endorse and advocate for Free Software. They can’t change the definition of freedom 30 years later, to remove freedom to use the software in certain situations.

              1. 4

                Yeah, you can’t blame the FSF for the existence of people who think that creating free software is less important than creating non-free software to try to advance their own political goals, which is exactly what this non-free license is trying to do.

                1. 3

                  You can blame both the FSF and the OSI though to be sleeping on those subjects. Or at least, not cut them any slack for it. The FSF has multiple times amended their license and introduced co-licenses, so it’s not like they carry on only the ideas of the 30s.

                  Sure, they are free to not pick up new movements. But that’s the best way to irrelevance. And considering that the FSF and especially Stallman do not have a good track record for communicating, they’ve got a problem at hand.

                  I mean, they are at the level where the FSFE keeps distance in public communication.

                2. 2

                  Recognize shifts in user demand, identify potential solutions with licensing and/or community management, and keep trying them out. They also do this in a way that tries to achieve as many goals of free software as possible. Individuals and small groups were steadily doing this on their own. So, FSF could’ve but didn’t.

                3. 1

                  Good write-up and analysis. I do have a question about your auto-aiming, sniper rifles. The basic moral issue is usually whether what you’re involved in does harm or not. There’s another area of discussion of doing work that minimizes harm of inevitable actions. Are we supporting harm or minimizing harm?

                  Snipers are usually, versus average shooter, world-class shots using custom, high-accuracy hardware. They usually have to fire a shot or two to access environmental factors. Then, they kill almost every target with no automation (exceptions exist). Auto-aiming is mostly about getting the job done faster with less risk to the shooter. However, both reducing the number of set-up shots and increasing accuracy of target shots will also likely reduce collateral hits that maim or murder non-targets. The person that builds this tech might save lives on top of whoever would’ve already be shot by these people that rarely miss. Is the auto-aiming sniper rifle still a harmful tech to build if it nets an increase in lives and health over standard practice?

                  There’s another risk that comes with it which makes me not want to work on it. I’ll save that one for after folks answer this one. I think this category of moral dilemma is itself interesting, necessary, to discuss.

                  1. 4

                    I think it’s safe to say that a sniper rifle is without question tech intended to inflict harm and be covered by this. Which lowers the thing to the classic non-software debate about ethical killing. I’m the wrong person to have that debate with. (though, in the context of archery and hunting, I regularly have similar debates)

                    If we’re talking about military, we’re coming around to more interesting grey areas. And the discussion around non-military licenses is old. For example, military uses tons of stock tech nowadays, especially hardware. (cars and such) So, can civil helicopter using software in a non-mil version be sold to the military?

                    This isn’t an absurd question, SQLite for example is used on guided missile carriers. I pick SQLite as an example, as it was intentionally built for that, so those authors certainly don’t have a problem there. And that’s valid. Do you want to use software that was built for that?

                    There’s a huge gamut there. People working living and working close to military locations regularly deal with that. I’m from a region in Germany where there’s both US and German military (Ramstein), so these discussions are not even remote from people.

                    I think it’s important for the free software movement to center back on what it was built for (moving power towards the users) while still making sure that they are a space for people answering all of these questions differently.

                    This unadressed problem has practical repercussions: it has just bitten Chef. Many companies in that field currently believe they have a uniform community and rely on that. And suddenly, Seth Vargo goes and deletes his (privately supported!) projects and retracts his willingness to do a service to his old employer. He’s did not even pull licensing issues, he’s just shutting down his service! And he finds support! So I do see these licenses even as a possibility to communicate better! I found Chefs responses very bad, but they also tell a consistent story about the company finally realising that their “community” PR term and their actual community don’t overlap.

                    To be quite open about the above, I have no fundamental problem with the concept of military and would not even consider myself a pacifist, I still wouldn’t like to have my work used to make development of weapon tech easier. Now, I do promote a programming language… I just don’t think “Freedom 0” it it’s purest form cuts it any more.

                    1. 1

                      Thanks for the detailed, nuanced response. Very interesting. Ramstein and SQLite are good examples, too. Ramstein is where the drone operators are located due to latency requirements on their software that takes out innocent people. I didn’t know SQLite was designed for missile carriers. Wth!? I’ve been alive too long to be surprised that something achieving such pervasive use and benefit was originally designed for something else more morally questionable. The Internet and GPS seem like they can be lumped into that category to some degree, too.

                      I think the Seth Vargo situation is even more interesting than that. Reading between the lines, you’re already saying it in your post. It’s that such actions will cause both community members and employers to reconsider the power relationships in any of their dependencies. That might affect both licensing (this thread) and how companies approach communities (or if they do at all).

                      Although I can’t predict much, I think they might complement their preference for permissive licenses with assigning copyright to foundations or something chartered to not revoke licenses or always expand uses. Also, they might try to acquire the copyrights or launch more internal projects they own like we see (random example) Cloudfare doing with Wireguard. I mean, some companies might already be doing it for that reason covering known unknowns that come with the (known part) of lack of full control over a dependency. Focusing on copyright control pushing for expansive use and no revocations (to them) is my only prediction.

                4. 9

                  There was an argument on Twitter about whether this actually protects people: if I decide an oil company that’s harming the environment is using my software, can I yank it from them under this license? I’d be harming their economic well-being.

                  1. 12

                    What happens when this gets weaponizes by someone who doesn’t want to support homosexuality or companies supporting it. Same for gun control or health insurance or the issue of the day. I don’t like putting kids in cages but I worry that this carve out idea has major flaws.

                    What if someone decides that my Judaism and its Zionism should exclude me? I feel personally threatened by this.

                    1. 1

                      Sadly I think this is more likely than say, a human rights violating military giving a shit about it.

                      Strangely enough complying with licences is not high on the agenda for torturers.

                      I know from The Bad Old days of Apartheid the sanctions only applied to innocent governmental activities like water supply… the military honestly didn’t give them a seconds thought.. if they needed software or a machine or whatever they’d lie, steal, put up fronts conceal whatever.

                      1. 2

                        Many kinds of entity are able to ignore these clauses.

                        However, there’s at least one growing group that can’t: companies who are selling to foreign governments (eg US companies selling tech to the Chinese government).

                        The US government isn’t going out of its way protect them, and the Chinese government can’t offer much protection from US courts.

                        1. 1

                          US companies selling tech to the Chinese government

                          You mean US companies openly selling tech to the evil entity that is openly using it.

                          In the Bad Old Days, US arms dealers quite liked sanctions…

                          US government regulations don’t mean “you cant do a deal”.

                          They mean the deals have to be done “at arms length” with a bit of forgery in the middle.

                          Meaning the salesdroids get to clip the ticket several times instead of just once.

                          Unsurprisingly, arms dealers are not very nice people.

                          1. 1

                            It sounds like you are saying (in a very roundabout way) “Yes, using this licence will increase the price paid by Bad Guys for technology, since they’ll have to do their deals at arms length”.

                            1. 1

                              That is an optimistic interpretation of what I’m saying, yes.

                              In practice since they are breaking (or bending to near breaking) some laws…. they tend to break any that will cost money. (eg. Getting export permits, paying tax… etc.)

                    2. 3

                      The most relevant part of the license for that question would be this:

                      The software may not be used by individuals … for … activities … that actively and knowingly … harm … the … economic … well-being of groups in violation of the United Nations Universal Declaration of Human Rights.

                      But revoking a license is not using the software. The restriction doesn’t apply to the license holder’s activities of licensing. So my answer is yes, you could revoke the license in that situation.

                      (I am not a lawyer.)

                    3. 8

                      There are arguments made that social media causes harm to others (e.g. polarization, isolation, and anxiety). Futhermore, many believe they are harmed by a loss of privacy from the advertising strategies of Google, Facebook, and others. The EU has taken these concerns very seriously and the US has been in a heated debate on the role of social media companies in society. A fair amount of open-source contributors are employed by such companies.

                      The point I’m making is not whether social media and advertising causes harm, but that what causes harm is highly politicized and subjective. Software development is not the same as practicing medicine.

                      1. 0

                        It is true that harm is subjective, but that doesn’t seem to be a problem for this license as currently written. Subjectivity isn’t a problem for the license because it implicitly uses the software user’s definition of harm (not the licensor’s or anyone else’s).

                        How I came to that conclusion: the restriction on use of the software only applies to “systems or activities that actively and knowingly” harm someone’s well-being. In my interpretation, “knowingly” means that the user of the software must think they are causing harm for that rule to apply. So even if most people think that, for example, social media software is harmful, the license wouldn’t forbid a Facebook employee from using the licensed software as long as that employee personally doesn’t “know” that social media is harmful.

                        Of course, enforcing that could be difficult, because if you accuse the software user of causing harm and they respond that they didn’t think they were causing harm, there is no way to know whether they are telling the truth about their thoughts. But that’s a separate problem.

                        1. 2

                          In my interpretation, “knowingly” means that the user of the software must think they are causing harm for that rule to apply.

                          What you said may be textually correct, but it is easily weaponized. It can be argued that Google and Facebook are lying, like Big Tobacco. Sean Parker, a Facebook founder, said they knowingly created an addictive platform. Other executives and developers have made similar statements about ‘knowing’. There are also studies that can be cited. Therefore, it could be argued that such employees knowingly cause harm.

                          Again, I’m not making this argument. I’m just making a point.

                          If the open-source community advocates this license, many could argue that is not Hippocratic but instead hypocritical. The open-source community is sponsored to a great extent by social media and advertisers.

                          EDIT: ‘many could argue that this is not Hippocratic but instead hypocritical.’

                      2. 13

                        This software may not be used by […]

                        Doesn’t that make it not open source software anymore? The OSI website says that open source licenses don’t discriminate against people, groups, or fields of endeavor. I also think it goes against the GNU definition of free software.

                        1. 11

                          There’s also considerable controversy over whether the OSI gets to define anything like that, or whether the OSD is a usable definition.

                          1. 7

                            Personally this feels like botanists telling you that a strawberry isn’t a berry but a watermelon is. Berry has a working definition that humans have been using for longer than we’ve had botanists and it doesn’t have to correspond 1:1 with some unrelated biological concept.

                            I know the analogy isn’t perfect because the history of the terms “open source” and “free” and “libre” are more tied up in each other, but I’m supremely uncomfortable with someone claiming authority to decide what somebody else means when they use a word

                            1. 4

                              You’re not wrong, the discussion whether OSI can lay claim on the name is also not new.

                              1. 3

                                “a strawberry isn’t a berry but a watermelon is”

                                I’ll be damned… Well, educated. I’m still calling them strawberries and watermelons, though. Call it inertia.

                                1. 2

                                  Bananas too!

                            2. 10

                              A lot of people have made that point–and for what it’s worth, the license page itself uses Google Analytics and hence is not libre software either.

                              This is kind of one of the root philosophical rifts in the community of folks who want open source software right now: hardcore FSF types that would argue that any discrimination is an infringement of freedoms, and those who believe that it is permissible to allow discrimination and limitation of rights so long as it protects underprivileged groups.

                              One of those positions is a lot more straightforward to both define and enforce, the other one is a lot better in terms of trying not to reinforce power hierarchies.

                              1. 3

                                I’m not sure even the first position is straightforward. My personal preference for the ISC (MIT/BSD-style) license is because I choose to give as much freedom as possible directly to those who follow me. GPL infringes on those people’s freedom for the greater benefit of my followers’ followers (and on and on).

                                I expect most of those “hardcore FSF types” consider GPL-style more “freedom-loving” framed this way, but the same facts lead me to the opposite conclusion.

                            3. 13

                              There’s something very Silicon Valley-about a license that let’s a megacorp turn it into effectively propitiatory software while playing the moral good guy.

                              Btw. the people behind this license were among the same ones who were pushing for RMS to be deposed. I had already seen this kind of anti-Freedom 0 talk on the Fediverse before, where people placed more value on stopping some other people from using code (such as Gab, in that instance) than sticking with the principles of free software/the OSI definition. It had really disturbed me, to see a growing group of people who didn’t even really know what it is they are talking about (claiming stories like Gab had “destroyed the name of Free Software”, because that’s the only instance where they had heard of it), and then pushing for not only legally unstable licenses, and limiting user freedom, but also undermining an entire movement for temporary symbolic victories. As evil Stallman says:

                              A condition against torture would not work, because enforcement of any free software license is done through the state. A state that wants to carry out torture will ignore the license. When victims of US torture try suing the US government, courts dismiss the cases on the grounds that their treatment is a national security secret. If a software developer tried to sue the US government for using a program for torture against the conditions of its license, that suit would be dismissed too. In general, states are clever at making legal excuses for whatever terrible things they want to do. Businesses with powerful lobbies can do it too.

                              1. 11

                                Just name the people. Coraline Ada Ehmke is the person writing this language, and also author of Contributor Covenant, maintainer of a multitude of rubygems. Sarah Mei, who was central to bringing the complaints about RMS to public (again!) is chair of Ruby Central, RubyConf and central to the formation of the Railsbridges. Both received awards for their community work, among them the Ruby Heroes award. Both have a huge track record as community builders.

                                Claiming that they don’t know what they are talking about is… something. Yes, they don’t feel a strong attachment to “Freedom 0” and the principles of the FSF/OSI. That can be treated as a fact. Yet, they have promoted and succeeded with many initiatives in the last 10 years. There’s a new political block, deal with it.

                                1. 4

                                  I’m not quite sure where you’re reading that Ehmke and Mei don’t know what they are talking about, I would fix it if I would see where. The people I claim are not entirely informed, are much smaller individuals that I have had discussions with, have read their comments and thoughts. The two you named might be spear-heading such initiatives, and I don’t even care that much how conscious or unconscious it is, but it’s the followers and those they manage to push along with them that make the change, and it is these I think have for the most part not seen both sides of the debate. Again, just my impression.

                                2. 9

                                  Btw. the people behind this license were among the same ones who were pushing for RMS to be deposed.

                                  Of course they are, tarnishing the name of Stallman was their first step in this attempted coup on the definition of software freedom.

                                  I believe this license is just testing the waters. This group’s future licensing will be more restrictive, more anti-freedom.

                                  1. 4

                                    The FSF’s view of software freedom is just one of many. Why else is the MIT license so popular?

                                    1. 4

                                      I don’t know what that has to do with my post. The MIT license has also been under attack lately (and will continue to be), as part of the same plot.

                                      1. 3

                                        You wrote

                                        […] tarnishing the name of Stallman was their first step in this attempted coup on the definition of software freedom.

                                        Ask a proponent of a BSD or MIT license if they believe that the GPL doesn’t restrict the freedom of the end user. They do. It’s the reason these licences exist in the first place (or rather, that there’s a plethora of FLOSS licenses).

                                        That’s what I meant by views of software freedom.

                                        The MIT license has also been under attack lately

                                        I am not aware of this. Can you give a term I can search for?

                                        1. 5

                                          The BSD and MIT licenses are not responses to the GPL license. They arose independently at roughly the same time, but for a different reason: academics trying to cut their projects loose from restrictive licenses imposed by commercial interests (AT&T and IBM).

                                          (Edit) It’s easy today to forget that it was frequently the case in CS academia in the late 80’s for your work to be somehow derived from code provided by a huge corporation, and thus the default license was very far from free.

                                          1. 1

                                            You are correct, I got the timeline wrong.

                                        2. 4

                                          It’s absurd to call this plot, sorry. It’s standard political action.

                                          Also, Stallman has tarnished his name himself. I mean, he’s the person that called people with mental disabilities “pets”. People have just called him out the final time.

                                          1. 2

                                            What’s wrong with the word “plot”?

                                            1. 3

                                              For one thing, it implies that it’s being done in secret, which it clearly is not.

                                              For another, it has connotations of “evil”. But because it’s just a connotation, it leaves open the chance to deny wanting to call the CCCoC and the Hippocratic license evil if someone calls you out on that.

                                              Coup d’éat is also, typically, illegal (different word, but also in your original post). Going to the media and convincing the board to vote out their chief executive is not only perfectly legal, it’s how it has worked with everyone from Richard Nixon to Ellen Pao.

                                              1. 1

                                                I suppose it does have connotations of evil, and I think that’s accurate, because I believe the actions being taken to destroy Free Software and Open Source are evil.

                                                The way I’ve used coup is completely reasonable. It’s used this way all the time, there is no reason to argue pedantics. Something being legal and previously successful does not mean that thing is ethically good.

                                            2. 1

                                              I’m almost afraid to ask for a cite to his views on mentally disabled people, because everything I’ve read by Stallman in his public posts has just made me go “WTF dude” even more.

                                      2. 1

                                        The stupid part of all this is that, while I have been arguing against your use of charged language, I agree with Stallman’s objection to anti-torture clauses. You wind up going to a lot of trouble to define torture, and then the State turns around and involves imminent domain of you.

                                        The UN Declaration of Human Rights isn’t a very good choice anyway. It includes positive rights like “having food” (article 25); if a State affirms the Declaration, then they’re supposed to give food to those who need it. Anybody who starves to death anywhere in the world is a violation of the Declaration of Human Rights, and, as a result, is a violation of this copyright license. (Unlike copyright licenses, the UN doesn’t kick members out for every violation of human rights.)

                                        If distributing your software requires me to solve world hunger, then I’m going to have to pass, because compliance is currently impossible.

                                      3. 5

                                        Admittedly I’m trying to generalize to the maximum degree possible, but this license seems like “Yup, this is open source, just don’t use it to do anything that might hurt people”

                                        This is the problem. There’s an implied “that I think” in there. The user of the license has to figure out what the author might feel about various people and what “harm” means. The author themselves may change their mind as time passes. Since there are no rights of use transferred in a predictable way, I don’t see how this qualifies as a license. It feels much more like a political statement. I would not want my software used to hurt people either. It’s just I have a different idea of what hurting people means than everybody else in the world I’ve met. We all develop our own standards as we grow up. (We are also all terribly influenced by our social cohorts)

                                        The Hippocratic Oath works for doctors because it addresses a human-to-human interaction with a fixed start and stop period, a set of observations and actions, and a measurable result. Software doesn’t fit into any of that.

                                        I think it’s fine not to want your software being used in ways you find horrible. It’s also true that you need to provide a good enough definition so that folks downstream can reliably parse that.

                                        There’s also a difference here between licensing and professional ethics. Professional ethics are the way we as professionals agree to act. We bind ourselves as part of whatever standards we think our profession should have. Licensing is a different matter entirely. It’s not an ethical matter, unless you feel that programmers should only make/use certain kinds of licenses because of the dangers of what we do.

                                        Definitions are slippery. I think we should reserve ethics for things professionals do, not legal contracts.

                                        1. 5

                                          The Hippocratic Oath works for doctors because it addresses a human-to-human interaction with a fixed start and stop period, a set of observations and actions, and a measurable result. Software doesn’t fit into any of that.

                                          As someone pointed out, the original hippocratic oath does not work as well in medicine. For example, surgeons inflict harm and therefore never use the hippocratic oath.

                                          But also, not everyone has to use this license.

                                          Definitions are slippery. I think we should reserve ethics for things professionals do, not legal contracts.

                                          Professionally speaking, I signed a number of legally biding ethical contracts in a B2B context. It’s not unusual.

                                          1. 2

                                            Ethics and morals clauses are quite common in business, or at least they used to be. Much of them boil down to “Don’t do something that would embarrass us” which is a reasonable request to make of a business partner. The implication here is that we trust you enough to know whether or not something would be embarrassing, i.e., there is a relationship.

                                            As far as surgeons hurting people, this is a long-standing problem with utilitarianism: at what scale do we define harm or benefit? Various scales give you different answers.

                                            Perhaps top-down, abstract-concept-to-rule is the wrong way to go here. Perhaps a more useful approach might be “find small things that work from other professions and steal them” I don’t know.

                                          2. 7

                                            “This is the problem. There’s an implied “that I think” in there.”

                                            “The Hippocratic License was created by Coraline Ada Ehmke”

                                            That’s intentional. Ehmke is a far-left, radical activist who tries to get projects and companies to strictly enforce their brand of inclusive, P.C. politics. The Codes of Conduct were designed for this. Any interpretation will be theirs. That’s why they both push for these kinds of documents and get like-minded people into community/moderation positions to enforce that interpretation. They’re usually subtle about it since it’s more effective when they pretend to be oppressed victims representing majority of minority members instead of aggressive activists backing a fringe belief likely created by and popular among privileged, University-educated people. This slip showed their true colors. Always assume anything involving these people will be about that or leveraged for it if possible. And that they’ll always lie about that since they consider that a necessary evil supporting the[ir] greater good.

                                            That said, other groups using it will likely have their interpretations and agendas. The safest thing is avoiding those license wherever possible (most will). If using one, skade’s recommendation to “ask the group whether X would constitute harm to them” before using it makes sense. Also, the licenses should be irrevocable to counter the fact that they could change their mind sometime after permission was granted and it was a dependency. Some might even set users up intentionally with that being leverage. The user might also change their behavior to be harmful later. High likelihood if for-profit companies.

                                            So, a mutually-protecting provision would say it’s irrevocable so long as their usage continues to avoid harms they promised to avoid. What you think of that politically-neutral strategy?

                                            1. 4

                                              I wonder how much we can expect to control and be responsible for the eventual result of our actions, assuming ethics is about results and not standards-of-practice, which was my supposition.

                                              Before the American Revolution, there was a famous case in Boston where British soldiers were accused of massacring colonists. Defending them, a hugely unpopular position, was future president Samuel Adams. He felt they deserved a good defense. Or rather, he felt an ethical obligation as a lawyer to make sure the system functioned as it should, regardless of the outcome.

                                              This outcome-based licensing and stress on “correct” use of code is problematic at best. We do work with small pieces of tiny things at time where we have no idea where they will eventually be used. This is one of the big reasons nobody ever wants to talk about this stuff. I think if you’re going to talk about the effects of the code we write as being an ethical matter, and I am doubtful this is a good course, then it has to be at the point of feature integration, not at some far earlier point.

                                              There are people who don’t have levels. That is, what’s moral for them is also the ethical thing and should be enforced by law (or licenses). They don’t distinguish between the personal, the professional, and the public. Many of these folks are activists. Since they mix everything up no matter what their inputs are, they make for a great anti-pattern. We need these folks to show us where the guardrails are, the places where conversations on ethics drift over into morals or public policy. (Note that I’m not saying I agree or disagree with any of them. That’s not even on the table here.)

                                              I think ethics has to be something like “how I act with this work directly in front of me and how those actions impact other humans” I don’t think it’s “stuff people do with my code” and I know it’s not “stuff I feel is wrong or right”

                                          3. 4

                                            Will this be used to oppose abortion as well?

                                            If people want a morality clause license this is much better -

                                            https://twitter.com/SarahJamieLewis/status/1176347681871912960

                                            1. 4

                                              Here’s a response from Karl Fogel, a former board member of the Open Source Initiative and author of the book Producing Open Source.

                                              1. 4

                                                Even better, a version of this with copyleft to force contributions back in.

                                                1. 7

                                                  Politics and software are so tangled that they cannot be reasonably separated.

                                                  I reject this, my test running framework is not political, nor are basically any of the tools I’ve written in my free time.

                                                  1. 1

                                                    I think this notion that “everything is political” is fascinating. I’d love someone to explain it to me in a way that I understand.

                                                    1. 1

                                                      If someone thinks that the OP’s test running framework is actually harmful and needs to be altered or shut down - maybe because it doesn’t check if ICE is running it and is therefore complicit with enforcing American immigration law, maybe because it is closed source software and supporting companies that produce closed-source software is immoral, maybe it is in opposition to the text of the Quran and is therefore heretical, any number of reasons that someone might have - then continuing to use it to test software in spite of vocal opposition is political. “Political” doesn’t mean “bad”, it means that some people somewhere think it’s bad when you don’t.

                                                      1. 1

                                                        “Political” doesn’t mean “bad”, it means that some people somewhere think it’s bad when you don’t.

                                                        Aha, then I agree that everything is “political” by that definition. But is that what they really mean?

                                                        1. 1

                                                          I find this idea to be completely absurd, it’s like saying that the authors GNU coreutils are guilty of murder because some targetting program on a guided missile uses ls at some point. The fact that a guided missile uses GNU coreutils doesn’t make listing the contents of a directory political.

                                                          1. 2

                                                            No, but the people behind the license in this article do think that the GPL itself is bad, precisely because it doesn’t allow you to legally prevent someone from using GNU coreutils on a guided missile (or in this case, doesn’t allow you to legally prevent your software from being used by a law enforcement body that enforces laws in a way you don’t like - it’s pretty clear that this license is aimed at ICE specifically). That is political, just as the GPL itself is political.

                                                            1. 1

                                                              You’re right that the GPL can be considered political, this isn’t the point I’m rejecting. The point I’m rejecting is this:

                                                              Politics and software are so tangled that they cannot be reasonably separated.

                                                              As I have already shown, this point is demonstrably false.

                                                    2. 6

                                                      economic well-being

                                                      Good luck selling software under capitalism that doesn’t impoverish somebody somewhere.

                                                      For me this license is naive crap that only a culturally-sheltered STEM person could have come up with. I mean, I can see it’s done in good faith, but really… There are just so many wrong things that I don’t know where to begin to criticize it.

                                                      1. 0

                                                        For me this license is naive crap that only a culturally-sheltered STEM person could have come up with. I mean, I can see it’s done in good faith, but really… There are just so many wrong things that I don’t know where to begin to criticize it.

                                                        I’ll begin by criticising you for criticising people by their identity and nothing else. Your assumption of good faith is not needed.

                                                        The person you are criticising has multiple times brought social projects to succession, so I’m surprised where the “culturally-sheltered STEM” assumption comes from. (FWIW, Coraline is a FOSS programmer with long track record, nothing could be further from the truth)

                                                        1. 9

                                                          Obviously I know her, she’s so controversial that it’s impossible not to know her.

                                                          Being involved in the FOSS community doesn’t say much. This is still a very ideological (and an outdated one) and naive piece. Being a prolific coder and contributor doesn’t say anything about your knowledge of philosophy, ethics or sociology.

                                                          Now “ethics” is a buzzword that lot of people swing around without knowing the implications of claiming an ethical position and this shows in the naivete of this license.

                                                          1. 2

                                                            More to the point, the license is literally just the MIT license plus one clause. You could start at the clause.

                                                            1. 3

                                                              so one problem is that it equates the United nation universal Declaration of human rights to ethics. But this is just in the title. Sticking to the clause, the main problem is that it still frames responsibility as something that emerges from the interaction between the company and the individual or groups. These is a very outdated and limited definition of responsibility. It doesn’t account for any systemic externality and is this definition is the source of so many problems of the modern world such as ecological collapse and poverty. These are hard to pin down, but are nevertheless critical to improve our world. While in the 19th century such a limited and traceable definition of responsibility was still acceptable, today it just sounds like a way to maintain the status quo and the “business as usual” attitude of our society in a time where radical change is needed.

                                                              Also, it’s not at all clear how this license should be enforced. It’s clear to me that she just wanted to make a political statement and not actually use the license in a real setting, but as a provocation, it is just too weak. Especially because it appeals to the declaration of human rights that historically serve more the interest of the (often colonial) oppressor than the interest of the oppressed.

                                                              So the license sits in a very weird spot where it is not good provocation and it is not the good license, with intention to achieve both.

                                                      2. 2

                                                        I wonder whether a Hippocratic Oath might be useful for developers as well. It seems that removing one’s code when it is depended upon for the day to day operation of many businesses violates the spirit of “first, do no harm.”