Ah, yet another "open-source if you follow my principles" license. It should be an uninteresting fact at this point that yes, you can in fact put whatever clauses you want in your software license.
Note that this is a license restriction addon, not a standalone license. To use, ammend to and reference by another license. Note that this removes "open-source" status from the original license if it is open-source, according to OSI's "No Discrimination Against Persons or Groups" definition.
Don't create new accounts to work around bans. However, since your question is one that comes up a lot (though, perhaps, only from your accounts), I'll answer it as if it's in good faith and not downvote / flag your comment.
The Open Source Definition is not at all the opinion of OSI. It was taken word-for-word from the Debian Free Software Guidelines, with some minor editorial changes to make it more generic than just Debian. It has influenced the free software definition of other projects (the Ubuntu Licensing Policy, Fedora and Red Hat's list of approved licenses, etc.) The practical reason to care about the Open Source Definition is that you want your software to be in Debian, Ubuntu, Fedora, RHEL, etc.
Being able to install software as part of a distro, and having distro maintainers work with you on getting visibility for the software, is valuable, and you shouldn't give it up lightly. If you are going to give it up—which is a fine thing to do, there's lots of good non-open-source software in the world—why restrict yourself to a half-open-source license? Why not preserve your right to commercialize your software so you can profit from your labor as much as you want? Presumably you've decided that you will make more profit (or other benefit) from association with the concept of "open source," but that association only helps you if you actually satisfy the Open Source Definition.
Well, it might also be that you actually want many people to be able to use, learn and understand your software freely but you also don't want your software to be installed on a military drone or used to enable a company that is working its employees to death. Not everything is profiteering and marketing.
That sounds great! But it's not open source. Debian, Fedora, etc. by policy do not want to prevent use of their software on military drones or by companies that work their employees to death. You will not convince them to change that policy. Therefore, you need to decide which of those two desires is more important to you: is it more valuable to have people be able to apt install / dnf install your software at the cost of letting it be usable by the military and exploitative employers?
It's totally legitimate to decide no, I care more about restricting use of my software, so I'll make packages (or a curl | sh script, or whatever) available on my website and guide people to do that. By all means, go do that, if that's what you want to do.
(If your'e actually considering doing this, I would like to try to convince you that a software license is the wrong tool for the job, here. But reasonable people can disagree on whether that's true.)
Still open source, but I guess you think Debian, Fedora, etc. are willing to ignore a lot of new open source software with more restrictive licenses and sacrifice some or a lot of their user base in favor of more full featured distros. I don't think this is going to happen. They will change policies and accept the new world.
I do strongly think that will not happen. Debian, for instance, constitutionally requires a 3-to-1 supermajority for a change to its "foundation documents," which include the Debian free software guidelines. The people who would have to vote on this are project members who have already pledged agreement with the existing free software guidelines. There are a lot of people who work on Debian precisely because they personally value the DFSG as it exists today.
There have been a lot of attempts at sort-of-free-software-but-not-really licenses over time: see, for instance, https://www.gnu.org/licenses/license-list.en.html#NonFreeSof... . There has been a lot of important software previously released under such licenses, from JSON to code drops from Sun (Java), Microsoft (.NET), and Apple (Darwin). The Linux distros have not included them. Remember that Debian doesn't even ship the gcc man page in the default distribution because it doesn't believe the FSF's own license for that man page is free software. I don't see what is different about today that would make things change. There isn't actually that much new "open source" software under non-OSD-compliant/non-DFSG-compliant licenses. And most of it comes from software that was previously under a compliant license, so for instance Debian and Fedora maintainers have forked the most recent OSD-compliant release of the Redis modules: https://goodformcode.com
Remember that every new license - even a new free software license - has an uphill battle because if it's incompatible with existing licenses you can't link software under those licenses or incorporate code under those licenses. There is a strong natural pressure to move to a standard license (or set of mutually-compatible licenses like MIT/BSD/etc.), and we've seen OSD-compliant licenses that are fine but uncommon be dropped in favor of more common ones. That pressure applies much more strongly to code that tries to set itself out as "open source" and is incompatible with every open source license.
Their opinion is not relevant. What is relevant is whether their definition of open-source (https://opensource.org/osd) is close to the ideals of developers. And it suffices for me and many others.
Well of course, because open source is about ignoring the ability of software developers to do good in the world, their abilities to change things, so instead companies can make money off the backs of software developers who are almost always working for free in their spare time. All in the hope that a software developer might get hired to maintain a project that they thought up.
The point of open source is to not be political, but that is an inane statement. Choosing to 'not be political', is itself a choice for the status quo. Choosing to 'not discriminate against any particular group, either positively or negatively' in a world where there is implicit, mostly-unspoken discrimination, is an implicit choice to continue that forward.
So why should those values be something to aspire towards? The status quo is a world in which the majority of people are not being payed what they are worth, so the people doing the paying can further inflate their ridiculously large bank balances.
I am probably the last person to quote MLK in common conversation, because a lot of his quotes have been misused or mispurposed to argue for something that he personally did not support. But, right now this one feels incredible pertinent:
""
I have almost reached the regrettable conclusion that the Negro’s great stumbling block in his stride toward freedom is not the White Citizen’s Counciler or the Ku Klux Klanner, but the white moderate, who is more devoted to “order” than to justice; who prefers a negative peace which is the absence of tension to a positive peace which is the presence of justice; who constantly says: “I agree with you in the goal you seek, but I cannot agree with your methods of direct action”;
""
Right, but MLK would not have identified himself as a "white moderate except ___," nor said "I like order, but also ___". His point was he had a different worldview from the white moderate, and tacking on to the white moderate would not have accomplished his goals.
Open source is definitely political, but it's also a very specific political movement. There can be other political movements with other goals. The reasons that open source advocates dislike licenses with restrictions on use are not just "we don't want to be political" or "we don't want to rock the boat." Some have a moral/ethical opinion that you should not restrict use via copyright, usually that you should not be able to restrict use via copyright. Some have a pragmatic opinion that attempts to restrict use will lead to worse outcomes.
The 996.ICU folks seem like they would be better served with their own movement that doesn't go by the name of "open source." We don't need to add that label to them.
> make money off the backs of software developers who are almost always working for free in their spare time.
Part of open sourcing something is accepting that very thing. It's volunteer work and a gift to the commons. Open source software directly helps people who can't afford expensive licenses of proprietary software, either because they are poor in a rich country, or because they aren't rich in a poor country. So don't say that open source software doesn't do good in the world!
> Choosing to 'not be political', is itself a choice for the status quo.
With this argument, most activism could be seen as "choice" for the status quo because you can always spend more of your time and money with activism. Barely anyone is spending all their time and money with it, so per your argument that makes most of them support the status quo.
The mere act of open sourcing something can already be seen as political action about the question of proprietary vs libre software. But that political action has limits, and I think those boundaries should be accepted.
Being non-political is analogous to abstaining in an election. Abstaining is the same as supporting whoever is the stronger force in that election. So it's not supporting the status quo, but whomever is strongest currently. There's a difference between the two.
Also note that activism, especially in non-democracies, often has negative consequences, onto your career, life, health, and of your entire family. Activism in democracies is dangerous as well, due to political discrimination by non government entities (which seems to be allowed in the US) as well as lone haters who resort to violence. And it's always possible that the democracy becomes totalitarian and the people in power take revenge on their prior opponents. This has happened in nazi germany as well as in russia and in other places. People shouldn't be required to ignore their own personal safety or career opportunities to fight for a political cause. Yes, people who do should be admired, but we shouldn't shame people who don't.
Last but not least, a society where people are forced to be activists is a totalitarian one and one where I don't want to live in.
It removes the status of OSI approved license if it used such license, because it becomes a different license, but it's still what people usually call open source. And it doesn't matter if you don't like the common meaning of open source, which has nothing to do with OSI's own definition.
"People"? Most people I know do actually mean the OSD. Most of my friends do, and certainly all my employers that have had open-source processes do—I can't necessarily use that process to release code to a weird non-OSD-compliant license just because the source is visible under some conditions.
(Also, "people" use the word "copyright" when they mean "trademark", and post "No copyright intended" on their YouTube video when they mean "no copyright infringement intended" and absolutely unequivocally are infringing copyright. The fact that some people don't know the meanings of words doesn't change whether the words have meanings.)
You are assuming they do, but they likely don't. Counter example for this is SQLite. Almost everyone considers it open source, presumably even you. Lots of projects that embed it are considered open source. SQLite, however, is not OSI open source and none of the projects that embed it can be OSI open source either. And it's ok, because people generally don't think of open source as something with OSI approved licenses.
I disagree because the OSI definition explicitly states that the software must not limit groups of people of any kind from using the software, regardless of how evil they may be or whether they comply to whatever monoculture you speak of. That's the point. By writing open-source software, you are explicitly allowing people with completely opposite ideals to use, review, and derive the software. And that's the beauty of it.
This license excludes any company that practices a 10x4 (10 hours for 4 day a week) work schedule which positive things have been said about previously. Also excludes companies with hourly jobs that work more than 45 hours but get paid 2/3x for extra hours. A lot of government agencies with unions have employees that love doing that since it gives them a lot of money.
edit: Hospitals as well since employees work odd shifts such as 2 weeks on and 2 weeks off for ER doctors, or 6 days in a row for lab technicians (with more days off every other week), etc.
It's a hard problem. To cover all such cases, the license would essentially have to be as complex as a developed country's labour laws. But adding complexity adds loopholes that could be exploited, for instance if a rule just allowed "working more than 45 hours for 2/3x overtime pay", a company could just reduce the main pay by two thirds for non-overtime work. Countering this might require some reference to local minimum wages or average wages, which makes the license significantly more complex to apply across jurisdictions.
You have no clue why someone is working more than 10 hours consecutively or more than 45 hours a week. Maybe that's something they're being paid quite well to do, and maybe they want to do so. This type of restriction just screams "I want to feel like a good person."
You're within your rights to use this restriction, but it's equally as ridiculous as a restriction forcing employees not to eat toast on Tuesdays, Wednesdays, or Thursdays.
I hope this doesn't catch on.
Edit: If the main target here is China, how much do you really believe Chinese companies care about Western copyright laws?
Right, this is perhaps the fundamental objection to social-good software licenses: how many people who don't engage in social good do respect copyright?
The GPL works because it's pretty unambiguous about what exactly your obligations are, it's pretty unambiguous when there's a violation of those obligations, and because there are entities that do care deeply about copyright (like, oh, Microsoft and Oracle and IBM) that you'd like to affect.
Any license of the form "You must not use this code to wage war." is not going to get respected. The military doesn't care if you sue them.
Any license of the form "You must obey laws" (like this one, "The above license is only granted to entities that act in concordance with local labor laws") is not going to get respected. Either the entity was going to obey laws, in which case it's redundant, or they weren't, in which case what do they care about copyright.
Any license that's five lines long and tells a US court they have to decide whether a Chinese entity is complaint with Chinese labor law is just ... not going to work the way you want it to.
Any license that restricts use and not distribution has the practical problem of being difficult to enforce. If your code runs inside Palantir's systems or on a Patriot missile or as part of the PRISM backend, how will you ever know? The AGPL skirts this line by retaining the "distribution" concept and just defining networked use as distribution (and I think even that is questionable) but anything where you're not actually seeing the infringement is not going to be feasible to enforce.
> Any license of the form "You must obey laws" (like this one, "The above license is only granted to entities that act in concordance with local labor laws") is not going to get respected. Either the entity was going to obey laws, in which case it's redundant, or they weren't, in which case what do they care about copyright.
What about companies that see respecting laws as a tradeoff between having less profits and getting fined?
If breaking the law implies an additional fine due to this license, it may tip the balance the other way.
>You have no clue why someone is working more than 10 hours consecutively or more than 45 hours a week.
I have quite a few mainland Chinese friends in tech (I used to work there), and most work extremely long hours not because they like it, but because management and company culture enforces it and they're powerless to change it. They can't even conduct collective action, as that's essentially illegal. Chinese courts have however recognised the GPL, so it's conceivable that a license like this could have a real effect.
I'm also led to believe that many people in the western game industry don't like crunch time (sometimes 100hour+ weeks), but have no choice in the matter. Hence their frequent calls for unionisation, such as https://www.reddit.com/r/gamedev/comments/86hac3/its_time_fo.... If such a license became widely used and could be legally enforced, it could potentially contribute to better working conditions in the game industry.
Moreover, note the words "request or schedule". Seems like it wouldn't apply to companies whose employees willingly work extra hours without this being implicitly requested or scheduled by the company.
>but it's equally as ridiculous as a restriction forcing employees not to eat toast on Tuesdays, Wednesdays, or Thursdays.
The difference is that it's not forcing anyone to do anything; nobody has to use software with this license, but if they choose to use the software they must accept the terms.
>Edit: If the main target here is China, how much do you really believe Chinese companies care about Western copyright laws?
I think 'Kaveren isn't saying "You don't know why these people are working 10 hours consecutively" but rather "You don't know why every person who works 10 hours consecutively is doing so." There is certainly a problem with the culture of mainland Chinese tech work (as well as with game dev work), but a license that legally bans use from any company that ever asks anyone to work more than a 10-hour day has too much splash damage, no matter how well-intentioned it is, and is thus a license that perfectly reasonable employers will avoid, and is thus a license that will only protect code that gets reimplemented by perfectly reasonable employers, and thus won't work.
>and is thus a license that perfectly reasonable employers will avoid
There are reasons a reasonable employer might adopt this. Maybe they never require employees to work more than 40 hour weeks, so they can advertise they follow this strict license to give them an advantage in hiring employees who want some guarantee they won't have to work overtime. Maybe they want to "open source" something to have the community contribute to their product for free, but don't want their competitors to be able to use it. Maybe a lone open source developer builds something with this license, then starts a business.
I think the idea is that if it catches on, eventually somebody will write e.g. a really important library using the license, then companies would have to consider using it.
Are there employers that never require employees to work more than 40 hour weeks? Ever? Even "implicitly"? Even if at 5:58 PM the elevator breaks and you can't leave until 6:01 PM? At the risk of losing what you say is a really important library forever? It seems too risky.
And even if there is one employer that is willing to make that guarantee, what library is important enough that it must be used and cannot be reimplemented, even given the combined resources of every employer not willing to make that guarantee?
If a lone developer wants to write software, great, but again it's not an open source license and they should have no expectation of others contributing to it, the way they would for open source. There's been lots of freeware / shareware in the world. It's mostly been out-competed by actual open source.
>Are there employers that never require employees to work more than 40 hour weeks? Ever? Even "implicitly"?
I've seen enough Europeans commenting on this site to say that yes, this is the case as it's illegal for companies in some countries to require employees to work more than that. Even the amount of paid overtime is strictly limited.
>Even if at 5:58 PM the elevator breaks and you can't leave until 6:01 PM?
I don't think any court (contracts are interpreted by courts) would reasonably consider this as the company requiring anybody to work to 6:01 PM, unless the company sabotaged the elevator.
What's a use case at a software company for requiring someone to work over 40 hours that can't be solved by having an extra employee with overlapping shifts?
>And even if there is one employer that is willing to make that guarantee, what library is important enough that it must be used and cannot be reimplemented, even given the combined resources of every employer not willing to make that guarantee?
Maybe one library is not enough, but if an ecosystem grows, it's possible. Companies don't exactly love the GPL, but adopted software using it where there was no other option.
>If a lone developer wants to write software, great, but again it's not an open source license and they should have no expectation of others contributing to it, the way they would for open source. There's been lots of freeware / shareware in the world. It's mostly been out-competed by actual open source.
Personally I agree that this license has a low chance of having any effect, but that doesn't mean it's not worth trying.
> What's a use case at a software company for requiring someone to work over 40 hours that can't be solved by having an extra employee with overlapping shifts?
Except labour laws treat overlapping shifts not as a shift work, but as a scheduled work. This means different overtime provisions etc.
>You have no clue why someone is working more than 10 hours consecutively or more than 45 hours a week. Maybe that's something they're being paid quite well to do, and maybe they want to do so.
Well, stopping them from being able to, will also help the other 90% that don't want to do so, doesn't get any monetary advantage, but still is pressured to. Small sacrifice to make.
Before this comment section gets any more toxic, please understand the context.
The org on GitHub that developed this license is called 996ICU (https://996.icu/#/en_US). It is a (very large) group of Chinese developers that are upset with the common practice of working 9am to 9pm, 6 days a week.
Please be mindful of the fact that lots of developers around the world work under genuinely awful conditions, and that your (x times y) work week is a privilege earned by a long history of labour activism in the Western world.
(P.S. This is a protest, not a legally binding license)
> Before this comment section gets any more toxic, please understand the context.
Comment section is not "toxic" just because it expresses the notion that this license is useless, nonsensical, stupid or any other thing you disagree with. Even if commenters might have missed the point.
your (x times y) work week is a privilege earned by a long history of labour activism in the Western world.
With most of "us" being some kind of a software engineer, I'm sincerely curious: was 80-hours workweek ever a mandatory thing for an engineer (teacher, doctor)?
In terms of a software development, laws/licences requiring maximum workweek hours is kind of pointless. If any company forces their employees to work in those kinds of conditions, they're going to get poor results, and their customers will notice. Software development requires a clear thinking mind, you can only get away with so much before productivity tanks. Manufacturing work is mindless, so they can get away with extra hours without as much decrease in productivity.
Well, it sort of depends on how / where you see yourself and whether you work for yourself. Doctors today work long hours, and teachers are sort of limited by class schedules so their labor organizing has generally focused on better pay / benefits rather than fewer hours, but even for engineers, there's a spectrum from locomotive engineer to Nikola Tesla.
- In 1791, carpenters in Philadelphia went on strike against "the prevailing sun-to-sun system, and the 75 hour work week" and fought for a 10-hour day. http://www.red-coral.net/WorkCoops.html
- In 1907, the US passed the "Hours of Service Act," which limited railroad telegraphers to a 9-hour day. In committee the act was almost going to be a limit to a 12-hour day (i.e., the normal day was longer than that!), until the telegraphers spammed their congresspeople using their telegraphs. https://en.wikipedia.org/wiki/Order_of_Railroad_Telegraphers...
I would say that software engineers, SREs, etc. working IC jobs maintaining large codebases/services at the direction of their bosses in large companies are far closer to the professions that historically have made up the labor movement—carpenters and typesetters and electricians and actors and journalists and composers and telegraph operators and so forth—than they are to the doctor running their own private clinic as their own business.
A quick search for “doctor work week” is returning numbers that start around 50 and go up to 108 for an emergency room surgeon. There is a horrible cultural expectation of crazy amounts of overwork in the US hospital world.
Seems dangerous in a medical profession. I certainly start making all kinds of obvious mistakes when tired or overworked, but my web application doesn't hurt anyone.
I read somewhere that the rationale is essentially that more mistakes are made when handing a patient off to another doctor on shift change. Communicating all the necessary information is difficult, context or details can be left out, resulting in larger mistakes being made by the incoming doctor. They think the tradeoff is worthwhile
Its not great for the doctors, but if you're a patient with an acute problem its much better to have the same few doctors working keeping watch over you than a brand new doctor every 8 hours coming on board that doesn't know anything about why you're there.
There's probably a more optimal solution then either a current 8h handoff, or a zombie caring for you.
However, doctoring, as profession, is a conservative one, and is highly resistant to process improvements. It took decades of effort, and decades of trivial mistakes (wrong organs removed, surgical instruments left inside patients when they are sewn up) to get the profession to even consider that maybe, maybe, checklists are a good idea during surgeries.
If someone came up with a process improvement that made patient handoff better, many doctors would resist it out of principle, vanity, or plain-old "I was hazed with 90 hr/week residency, so should every new doctor."
It's not necessarily a question of tech, or 'let's all jump to adopt this new scrum/agile/extreme doctoring new hotness'.
But I, as a consumer of medicine, would like the profession to spend more time in reviewing its processes, their outcomes, and to not assume that doctors are infallible geniuses, who don't need any of that checklist nonsense. (Oh, hey, the patient just woke up, and told us that we removed the wrong lung...)
2 hours shared, 4 hours alone, then 2 hours shared?
The extreme case actually seems good; pair doctoring where shifts start four hours apart, so you spend the first half of your shift with one doctor, and the second half with another.
Maybe instead do 2 x three hours, and give doctors really reasonable work hours or time to keep up on current research.
I fear the conclusion to that would be managers saying "what are all these doctors doing, we need to lay 3/4ths of them off and make the rest work twice as much for 50% more pay!"
You get license to use the software if you abide by the terms of the license agreement. A company that asks its employees to work 60 hour weeks breaks the license agreement, and therefore cannot use the software.
The only LAW that could be written about this would be "Contracts cannot include limits on the number of hours employers ask people to work".
So maybe, yes, the megacorps get that law passed. Until then this license is binding...
Now whether the any particular legal system cares to enforce the license terms is another matter
By using the software, you agree to the terms of the license, and can in principle suffer a civil (non-criminal) lawsuit against you for breaching them.
>Good working conditions for highly-skilled in-demand work arise naturally through supply and demand.
But they don't necessarily arise if the demand from workers wanting to do that work is sufficient. For instance, game developers have relatively terrible work conditions by the standards of the software industry, because so many people want to work on games. You could reasonably argue that they get enough satisfaction from being able to work on games that they're willing to put up with worse conditions, but that doesn't change that their work conditions are relatively bad. The nice thing about somehow bringing about a reduction in developer working hours is that it wouldn't necessarily result in any loss in economic output or employer profit, as there's a wealth of research showing that working beyond a certain amount of hours daily is counterproductive, so it's a win-win, and could even increase productivity.
I would love to have examples of good working conditions being arrived at naturally. Given that there are many programmers of various talent levels throughout the industry I have noted a number of dark patterns in programming work: long hours, pay that refuses to follow value of work (Apple makes 400k profit per employee), reduction in workspace quality, etc.
We can all pretend that we have more economic power than our highly paid blue collar peers that worked in highly unionized work forces in the past, but the programmers that make the kind of money for that to be true are very rare. Nearly 30 years ago my father made 70k as a union welder. It is highly unlikely that I will be able to match that salary street inflation, and the world is more expensive: education and health care could eat us alive at any moment.
I don't know how much of our effort the capital owning class needs to strip from us before we take advantage of the massive productive capacities we have before we make it fair again, but I assure you, it will be organized, not natural.
This is easy to say. Can you give examples? I'm no expert on economic history, just a run of the mill history buff, but my impression is that the divide is not between people who are unskilled and in low demand vs. those skilled and in high demand. I would say it is between owners of capital, who are not bound to work for a salary, and everyone else. Typically if you work for a salary you're worried about and dependent on the arrival of the next payment. Regardless of how easy you think it will be to find another job the loss of the current income stream is disruptive. I just finished a book that related the very long working hours of clerks performing accounting tasks for Canton-based American merchant houses during the middle 19th century. So there you had skilled clerks, who had to be imported by sailing vessel from the other side of the pacific, and still felt compelled to work essentially dawn to dusk 6 days a week.
While I'm not defending the schedules that hospital employees work, one thing to factor in is that every hand off can increase errors. (I wish I could find a paper, but I'm on mobile and not having luck.)
However, yes, I do believe that multiple 12+ hour days in a row is ultimately counterproductive. Also, I wonder if overlapping shifts would reduce the chance of errors.
Those long shifts are there to protect patient safety - they reduce the number of dangerous handovers. Part of being a doctor or in something like the military is putting other people first rather than having a comfortable short day.
Take this for example - from the Guardian, a very left-wing newspaper that is normally about as far from arguing for longer hours for works as you can possibly get.
> The report suggests the shorter working week is responsible for a lack of continuity in out-of-hours care
> shortening rota hours had led to more shift changes between medical teams and greater chances for mistakes to occur
> the early effects of the 48-hour European working time limit on surgeons and found that these new rotas had almost entirely removed adequate time for handover of sick patients
Also think about this - those doctors who don't do handovers, such as a dermatologist doing out-patient clinics, don't work long hours. Hospitals only do it where clinically required.
Handovers are dangerous, but making decisions that need your full attention after hard working 11:30 hours is also dangerous. And sometime doctors have even longer shifts.
It's an insane culture.
My guess is that it is easy to spot error in handovers (X didn't say/write this, Y didn't hear/read that). But errors due that you have worked 11:30 hours are more difficult to isolate.
Why do you think the hospitals do long shifts then? Just for fun?
They don't do it for people who aren't doing handovers, like general practitioners who can work half-days just a couple of times a week if they want, or administrations. So we know it isn't inherent for doctors or for the staffing of the hospital. Only for those doing handovers...
Personally I don’t find “everyone is doing this so it must be rational” a very convincing argument. After all, maybe everyone else followed that same weak argument.
But conversely how do you think everyone arrived at an unsafe practice particular to just these doctors who do handovers and no others? And why do you think they’re campaigning themselves to bring them back for safety in the UK?
All of these countries didn't independently discover medicine, and they do speak with each other i.e. they share a culture. Why do they all dress the same, why are so many addressed with a title in front of their names, and why do they all seem to refer to a lot of things in Latin?
We can choose to think that they independently arrived at an ideal state, but that's not reasonable.
Hospitals have some of the most toxic work cultures around. So much of that shit is perpetuated by senior level employees justifying hours that are harmful to patient care with “I went through it so why shouldn’t you?”
I support the general idea of not having to work 12 hours a day for 6 days. That's crazy. But there are other venues for labour activism than changing the license. And in other Chinese industries that aren't related to software at all, labour activism is quite successful as well. So there is no need to resort to license changes. As for your PS, there is no indication that it's a protest and not just a license.
> The purpose of this license is to prevent 996-style companies from using the software or codes under the license, and force those companies to balance their work schedule.
Doesn't sound like protest to me, but like extortion. Either you do what we want, or you can't use our software.
Offering a benefit (software license) you have the rights to and are under no obligation to give in exchange for something you want (adopting certain labor practices) isn't extortion, it's just selling.
> But there are other venues for labour activism than changing the license
The thing is, this kind of online protest is the safest for both initiators and participants. If by "other venues for labour activism" you mean go to the street, you're basically risk sending people to jail (And after the thing, nobody want to risk to get milled down by tanks).
Also, did you know that 996 is by law illegal in China? You can technically report your company if they force you to work 996. But: 1) After that, you will probably get fired for any number of excuses. 2) Your company is clever, they know many ways to get around the law.
Now, after all that dilemma, how could you protected yourself and help other developers just like you? The license idea come out naturally.
> The thing is, this kind of online protest is the safest for both initiators and participants. If by "other venues for labour activism" you mean go to the street, you're basically risk sending people to jail (And after the thing, nobody want to risk to get milled down by tanks).
Why will the government avoid to use repressive measures against this? Otherwise they don't do their job of totalitarian government well enough. In a totalitarian state, I'd certainly be afraid to support this.
> Also, did you know that 996 is by law illegal in China? You can technically report your company if they force you to work 996. But: 1) After that, you will probably get fired for any number of excuses. 2) Your company is clever, they know many ways to get around the law.
Yeah, I've read the website. So technically, the license isn't needed.
> how to you protected yourself and help other developers just like you? The license idea come out naturally.
If you can't get the law enforced, how will you be able to enforce the license? I've heard that chinese courts generally don't meet western standards.
> Why will the government avoid to use repressive measures against this?
This is how the labor law been structured in China. In many cases you need to file the case by yourself in order to get it investigated.
> So technically, the license isn't needed.
> If you can't get the law enforced, how will you be able to enforce the license
I don't think you've understood the situation. In China, we play it safe.
You can get administration involved, but you will likely need to file the case by yourself, which may cause you to get fired. Now, if a third-party can help you by putting some pressure on your company, then both of you will be safe.
The license adds another reason to convince your boss to not implement 996.
> there are other venues for labour activism than changing the license.
This argument would work just as well for all of these other venues. That there's something else you could be doing is not a valid argument against doing the thing that you are doing.
"Why strike when you could change software licenses?"
I live in Brazil. Under Brazilian labor laws, a work week is defined as 44 hours, with each work day having up to 8 hours. So you just excluded companies that have ever had anyone do more than one hour of (paid) overtime any week. Which is basically every company in my country, considering (occasional, paid) overtime is normal and predicted in the relevant laws.
Here, 12 hour shifts with 36 hour rest periods is also the norm for security guards, etc. (So that 4 can provide 24h security) That results in more than 45 hours every other week.
The license also doesn't specify how to remedy the situation in case it's ever violated, which feels very draconian and not very pragmatic.
Maybe you can try to file a pull request to suggest a change, remember to explain your situation and concerns inside the pull request so it won't be discarded.
The issue with licenses is that you can't just change them around like you want. The GPLv2 in the Linux kernel for example is almost impossible to get exchanged due to the giant amount people who possibly hold copyright. You need agreement of every single one of them.
Personally, I don't even believe they can receive that many stars on GitHub. The fact they did amazed me.
Of course I understand ask some projects to suddenly change their license to support anti-996 is hard and unreal, but hey, the show of force and well there is already a win.
For me, the step they've taken to reach out for support (by ask others to change their code license etc) is already something positively progressive. Regardless whether or not they did successfully convinced anyone.
It's a bit worse than it seems, because lunch breaks are not included, so whereas an American would call his 9-to-5 job a 40 hour work week, it would be equivalent to a 35 hour Brazilian work week. A Brazilian work day is usually 8h30 to 18h18 so that 44h will fit into 5 days. You can see how adding a commute of at least and hour (could easily be 2h or 3h in São Paulo or Rio) will simply eat up your whole day.
I think a reduction of the 44h week is long due in Brazil and I don't understand why there isn't more pressure from workers. They probably don't have the time to protest.
> I think a reduction of the 44h week is long due in Brazil and I don't understand why there isn't more pressure from workers.
I found the expectations around what you're supposed to do during work hours to be very different between Brazil and US.
Extreme conditions obviously exist, (e.g. telemarketers, which are usually _continuously_ taking calls) but from my observation a significant portion of those 44 hours is spent idling or socializing.
I think proposing any change in work hours under the current political and socioeconomic climate (12% unemployment) wouldn't be feasible, and without the needed shift in work culture our productivity would probably fall further behind.
I don't know, I think 9 to 6 has been becoming more of the norm here as well. All the places I have worked so far (maybe I've been unlucky) have expected more of a 9 to 6 (give or take an hour shift in either direction) than a 9 to 5 cadence.
Also, I don't know many people who work a strict 9 to 5 in office jobs, many take work home with them or are expected to always be available even on weekends should something come up. Not saying this is good, but I don't think strict 9-5 is the norm anymore.
It seems fairly self-explanatory to me, but the second result of a Google search for "time h notation" gave me this (the first was documentation for the time.h header):
> So you just excluded companies that have ever had anyone do more than one hour of (paid) overtime any week.
I think this was entirely the point of the license. The authors presumably believe that companies shouldn't make their workers work this long, and they don't want those companies to do to benefit from their work. Thus providing an (admittedly small) incentive for those companies to reduce their working hours.
In the best job I ever had in my life, my boss basically told me that she didn't care about the hours I kept or how long I worked every day as long as I got my job done. I usually showed up to work at 12pm and left around 4pm. It ended up being one of the most productive times of my life. I created products that had hundreds of millions of dollars worth of impact on the company bottom line. In my immediate next job I had to keep long hours and for some reasons I wasn't nearly half as productive.
Atleast in my life, shorter work hours tend to heavily inversely correlate with my work output.
I am currently working in a ROWE (results only work environment). I have a similar schedule. I get way more done and have way more ideas. It's almost as if "gasp" letting people design their own work process makes them more effective!
Does this work best if your work is well-planned by someone else? E.g., you have a list of prioritized tickets with requirements complete, and can just code down the list?
I work 32h/wk (mon-thu), and I feel happy with my productivity, but also recognize that I spend tons of time in meetings, clarifying requirements, helping other devs, etc. These aren't all really my role, so I could theoretically optimize them away and hope the invisible tasks get picked up successfully by someone else, but I know we'd have worse outcomes if I wasn't involved and "just" did my explicitly assigned tasks.
Can someone with team lead/management responsibilities optimize for 'job done' this aggressively?
> In the best job I ever had in my life, my boss basically told me that she didn't care about the hours I kept or how long I worked every day as long as I got my job done. I usually showed up to work at 12pm and left around 4pm.
Did you have to interact with anyone?
I can imagine this can be challenging if another person you need to interact with follows the same principle with a 7am to 11am schedule.
Correct. A large part of this arrangement is an implicit be an adult understanding. So if such a scenario came up, I'd work out an arrangement with the other person where we could overlap some time as frequently as we need to.
That's a pretty cool arrangement. How was "getting job done" defined? My experience as a developer is that there is always something to do. So where is the line drawn?
There will always be something to do and you just have to draw a line somewhere. In my experience once you have successfully delivered a few projects with high efficiency and people trust/respect you, they stop asking questions.
Overworking is a problem on a national scale, because it leads to the decay of every aspect of society that is not work. The most critical aspect is that it leads to the decay of families. It can lead to a big problem where the population is much lower and the people that do exist have had little invested into them as children. It also leads to a decay of the civil duty to shepherd your government, although the Chinese would consider that a good thing. It is tempting to eat the cake today but the price will be paid in full in one generation.
I know. I'm just arguing that asking for more work is OK, if you're willing to pay for it. Many years ago, I worked a union job (chemical plant) and we got 1.5x wages for overtime, and 2x wages for working legal holidays.
Copyright license only covers copying the software or source code. This seems like a "terms of use" that they tried to shove into a copyright license which I don't think would hold up in court. This license would cover the package distribution site/provider/github, not the person/entity using the package. The entity/person using software would only trigger the copyright license if they distributed the package(eg: putting the package into a mobile app or selling the software).
139 comments
[ 3.5 ms ] story [ 199 ms ] threadNote that this is a license restriction addon, not a standalone license. To use, ammend to and reference by another license. Note that this removes "open-source" status from the original license if it is open-source, according to OSI's "No Discrimination Against Persons or Groups" definition.
The people who talk loudest about Open Source always manage to get a position that profits from the labor of the actual authors.
The people who talk loudest about Open Source always manage to get a position that profits from the labor of the actual authors.
The Open Source Definition is not at all the opinion of OSI. It was taken word-for-word from the Debian Free Software Guidelines, with some minor editorial changes to make it more generic than just Debian. It has influenced the free software definition of other projects (the Ubuntu Licensing Policy, Fedora and Red Hat's list of approved licenses, etc.) The practical reason to care about the Open Source Definition is that you want your software to be in Debian, Ubuntu, Fedora, RHEL, etc.
Being able to install software as part of a distro, and having distro maintainers work with you on getting visibility for the software, is valuable, and you shouldn't give it up lightly. If you are going to give it up—which is a fine thing to do, there's lots of good non-open-source software in the world—why restrict yourself to a half-open-source license? Why not preserve your right to commercialize your software so you can profit from your labor as much as you want? Presumably you've decided that you will make more profit (or other benefit) from association with the concept of "open source," but that association only helps you if you actually satisfy the Open Source Definition.
It's totally legitimate to decide no, I care more about restricting use of my software, so I'll make packages (or a curl | sh script, or whatever) available on my website and guide people to do that. By all means, go do that, if that's what you want to do.
(If your'e actually considering doing this, I would like to try to convince you that a software license is the wrong tool for the job, here. But reasonable people can disagree on whether that's true.)
There have been a lot of attempts at sort-of-free-software-but-not-really licenses over time: see, for instance, https://www.gnu.org/licenses/license-list.en.html#NonFreeSof... . There has been a lot of important software previously released under such licenses, from JSON to code drops from Sun (Java), Microsoft (.NET), and Apple (Darwin). The Linux distros have not included them. Remember that Debian doesn't even ship the gcc man page in the default distribution because it doesn't believe the FSF's own license for that man page is free software. I don't see what is different about today that would make things change. There isn't actually that much new "open source" software under non-OSD-compliant/non-DFSG-compliant licenses. And most of it comes from software that was previously under a compliant license, so for instance Debian and Fedora maintainers have forked the most recent OSD-compliant release of the Redis modules: https://goodformcode.com
Remember that every new license - even a new free software license - has an uphill battle because if it's incompatible with existing licenses you can't link software under those licenses or incorporate code under those licenses. There is a strong natural pressure to move to a standard license (or set of mutually-compatible licenses like MIT/BSD/etc.), and we've seen OSD-compliant licenses that are fine but uncommon be dropped in favor of more common ones. That pressure applies much more strongly to code that tries to set itself out as "open source" and is incompatible with every open source license.
Their opinion is not relevant. What is relevant is whether their definition of open-source (https://opensource.org/osd) is close to the ideals of developers. And it suffices for me and many others.
Your second sentence is nonsense.
No it is not. But I see that you are a "founder" and thus profiting from my work. Your opinion is self-serving and meaningless.
The point of open source is to not be political, but that is an inane statement. Choosing to 'not be political', is itself a choice for the status quo. Choosing to 'not discriminate against any particular group, either positively or negatively' in a world where there is implicit, mostly-unspoken discrimination, is an implicit choice to continue that forward.
So why should those values be something to aspire towards? The status quo is a world in which the majority of people are not being payed what they are worth, so the people doing the paying can further inflate their ridiculously large bank balances.
I am probably the last person to quote MLK in common conversation, because a lot of his quotes have been misused or mispurposed to argue for something that he personally did not support. But, right now this one feels incredible pertinent:
"" I have almost reached the regrettable conclusion that the Negro’s great stumbling block in his stride toward freedom is not the White Citizen’s Counciler or the Ku Klux Klanner, but the white moderate, who is more devoted to “order” than to justice; who prefers a negative peace which is the absence of tension to a positive peace which is the presence of justice; who constantly says: “I agree with you in the goal you seek, but I cannot agree with your methods of direct action”; ""
Open source is definitely political, but it's also a very specific political movement. There can be other political movements with other goals. The reasons that open source advocates dislike licenses with restrictions on use are not just "we don't want to be political" or "we don't want to rock the boat." Some have a moral/ethical opinion that you should not restrict use via copyright, usually that you should not be able to restrict use via copyright. Some have a pragmatic opinion that attempts to restrict use will lead to worse outcomes.
The 996.ICU folks seem like they would be better served with their own movement that doesn't go by the name of "open source." We don't need to add that label to them.
Part of open sourcing something is accepting that very thing. It's volunteer work and a gift to the commons. Open source software directly helps people who can't afford expensive licenses of proprietary software, either because they are poor in a rich country, or because they aren't rich in a poor country. So don't say that open source software doesn't do good in the world!
> Choosing to 'not be political', is itself a choice for the status quo.
With this argument, most activism could be seen as "choice" for the status quo because you can always spend more of your time and money with activism. Barely anyone is spending all their time and money with it, so per your argument that makes most of them support the status quo.
The mere act of open sourcing something can already be seen as political action about the question of proprietary vs libre software. But that political action has limits, and I think those boundaries should be accepted.
Being non-political is analogous to abstaining in an election. Abstaining is the same as supporting whoever is the stronger force in that election. So it's not supporting the status quo, but whomever is strongest currently. There's a difference between the two.
Also note that activism, especially in non-democracies, often has negative consequences, onto your career, life, health, and of your entire family. Activism in democracies is dangerous as well, due to political discrimination by non government entities (which seems to be allowed in the US) as well as lone haters who resort to violence. And it's always possible that the democracy becomes totalitarian and the people in power take revenge on their prior opponents. This has happened in nazi germany as well as in russia and in other places. People shouldn't be required to ignore their own personal safety or career opportunities to fight for a political cause. Yes, people who do should be admired, but we shouldn't shame people who don't.
Last but not least, a society where people are forced to be activists is a totalitarian one and one where I don't want to live in.
(Also, "people" use the word "copyright" when they mean "trademark", and post "No copyright intended" on their YouTube video when they mean "no copyright infringement intended" and absolutely unequivocally are infringing copyright. The fact that some people don't know the meanings of words doesn't change whether the words have meanings.)
Why is it the case that projects that embed SQLite cannot be open source in the OSI sense? (I don't think this is true.)
From what I can read, that's the plan. After all "open-source" or not is pointless when developers are on the bed inside of ICU :D
edit: Hospitals as well since employees work odd shifts such as 2 weeks on and 2 weeks off for ER doctors, or 6 days in a row for lab technicians (with more days off every other week), etc.
You're within your rights to use this restriction, but it's equally as ridiculous as a restriction forcing employees not to eat toast on Tuesdays, Wednesdays, or Thursdays.
I hope this doesn't catch on.
Edit: If the main target here is China, how much do you really believe Chinese companies care about Western copyright laws?
The GPL works because it's pretty unambiguous about what exactly your obligations are, it's pretty unambiguous when there's a violation of those obligations, and because there are entities that do care deeply about copyright (like, oh, Microsoft and Oracle and IBM) that you'd like to affect.
Any license of the form "You must not use this code to wage war." is not going to get respected. The military doesn't care if you sue them.
Any license of the form "You must obey laws" (like this one, "The above license is only granted to entities that act in concordance with local labor laws") is not going to get respected. Either the entity was going to obey laws, in which case it's redundant, or they weren't, in which case what do they care about copyright.
Any license that's five lines long and tells a US court they have to decide whether a Chinese entity is complaint with Chinese labor law is just ... not going to work the way you want it to.
Any license that restricts use and not distribution has the practical problem of being difficult to enforce. If your code runs inside Palantir's systems or on a Patriot missile or as part of the PRISM backend, how will you ever know? The AGPL skirts this line by retaining the "distribution" concept and just defining networked use as distribution (and I think even that is questionable) but anything where you're not actually seeing the infringement is not going to be feasible to enforce.
What about companies that see respecting laws as a tradeoff between having less profits and getting fined? If breaking the law implies an additional fine due to this license, it may tip the balance the other way.
I have quite a few mainland Chinese friends in tech (I used to work there), and most work extremely long hours not because they like it, but because management and company culture enforces it and they're powerless to change it. They can't even conduct collective action, as that's essentially illegal. Chinese courts have however recognised the GPL, so it's conceivable that a license like this could have a real effect.
I'm also led to believe that many people in the western game industry don't like crunch time (sometimes 100hour+ weeks), but have no choice in the matter. Hence their frequent calls for unionisation, such as https://www.reddit.com/r/gamedev/comments/86hac3/its_time_fo.... If such a license became widely used and could be legally enforced, it could potentially contribute to better working conditions in the game industry.
Moreover, note the words "request or schedule". Seems like it wouldn't apply to companies whose employees willingly work extra hours without this being implicitly requested or scheduled by the company.
>but it's equally as ridiculous as a restriction forcing employees not to eat toast on Tuesdays, Wednesdays, or Thursdays.
The difference is that it's not forcing anyone to do anything; nobody has to use software with this license, but if they choose to use the software they must accept the terms.
>Edit: If the main target here is China, how much do you really believe Chinese companies care about Western copyright laws?
Chinese courts have in fact shown willingness to enforce the GPL in some cases: https://heathermeeker.com/2018/04/30/first-gpl-case-in-china...
There are reasons a reasonable employer might adopt this. Maybe they never require employees to work more than 40 hour weeks, so they can advertise they follow this strict license to give them an advantage in hiring employees who want some guarantee they won't have to work overtime. Maybe they want to "open source" something to have the community contribute to their product for free, but don't want their competitors to be able to use it. Maybe a lone open source developer builds something with this license, then starts a business.
I think the idea is that if it catches on, eventually somebody will write e.g. a really important library using the license, then companies would have to consider using it.
And even if there is one employer that is willing to make that guarantee, what library is important enough that it must be used and cannot be reimplemented, even given the combined resources of every employer not willing to make that guarantee?
If a lone developer wants to write software, great, but again it's not an open source license and they should have no expectation of others contributing to it, the way they would for open source. There's been lots of freeware / shareware in the world. It's mostly been out-competed by actual open source.
I've seen enough Europeans commenting on this site to say that yes, this is the case as it's illegal for companies in some countries to require employees to work more than that. Even the amount of paid overtime is strictly limited.
>Even if at 5:58 PM the elevator breaks and you can't leave until 6:01 PM?
I don't think any court (contracts are interpreted by courts) would reasonably consider this as the company requiring anybody to work to 6:01 PM, unless the company sabotaged the elevator.
What's a use case at a software company for requiring someone to work over 40 hours that can't be solved by having an extra employee with overlapping shifts?
>And even if there is one employer that is willing to make that guarantee, what library is important enough that it must be used and cannot be reimplemented, even given the combined resources of every employer not willing to make that guarantee?
Maybe one library is not enough, but if an ecosystem grows, it's possible. Companies don't exactly love the GPL, but adopted software using it where there was no other option.
>If a lone developer wants to write software, great, but again it's not an open source license and they should have no expectation of others contributing to it, the way they would for open source. There's been lots of freeware / shareware in the world. It's mostly been out-competed by actual open source.
Personally I agree that this license has a low chance of having any effect, but that doesn't mean it's not worth trying.
Except labour laws treat overlapping shifts not as a shift work, but as a scheduled work. This means different overtime provisions etc.
Well, stopping them from being able to, will also help the other 90% that don't want to do so, doesn't get any monetary advantage, but still is pressured to. Small sacrifice to make.
The org on GitHub that developed this license is called 996ICU (https://996.icu/#/en_US). It is a (very large) group of Chinese developers that are upset with the common practice of working 9am to 9pm, 6 days a week.
Please be mindful of the fact that lots of developers around the world work under genuinely awful conditions, and that your (x times y) work week is a privilege earned by a long history of labour activism in the Western world.
(P.S. This is a protest, not a legally binding license)
Comment section is not "toxic" just because it expresses the notion that this license is useless, nonsensical, stupid or any other thing you disagree with. Even if commenters might have missed the point.
With most of "us" being some kind of a software engineer, I'm sincerely curious: was 80-hours workweek ever a mandatory thing for an engineer (teacher, doctor)?
Here are some things I've learned from https://en.wikipedia.org/wiki/Eight-hour_day#United_States and linked pages :
- In 1791, carpenters in Philadelphia went on strike against "the prevailing sun-to-sun system, and the 75 hour work week" and fought for a 10-hour day. http://www.red-coral.net/WorkCoops.html
- The International Typographical Union, which had relatively well-off members, went on strike around 1900-ish initially for a 48-hour work week and then for a 40-hour work week. https://en.wikipedia.org/wiki/International_Typographical_Un...
- In 1907, the US passed the "Hours of Service Act," which limited railroad telegraphers to a 9-hour day. In committee the act was almost going to be a limit to a 12-hour day (i.e., the normal day was longer than that!), until the telegraphers spammed their congresspeople using their telegraphs. https://en.wikipedia.org/wiki/Order_of_Railroad_Telegraphers...
I would say that software engineers, SREs, etc. working IC jobs maintaining large codebases/services at the direction of their bosses in large companies are far closer to the professions that historically have made up the labor movement—carpenters and typesetters and electricians and actors and journalists and composers and telegraph operators and so forth—than they are to the doctor running their own private clinic as their own business.
Its not great for the doctors, but if you're a patient with an acute problem its much better to have the same few doctors working keeping watch over you than a brand new doctor every 8 hours coming on board that doesn't know anything about why you're there.
However, doctoring, as profession, is a conservative one, and is highly resistant to process improvements. It took decades of effort, and decades of trivial mistakes (wrong organs removed, surgical instruments left inside patients when they are sewn up) to get the profession to even consider that maybe, maybe, checklists are a good idea during surgeries.
If someone came up with a process improvement that made patient handoff better, many doctors would resist it out of principle, vanity, or plain-old "I was hazed with 90 hr/week residency, so should every new doctor."
But I, as a consumer of medicine, would like the profession to spend more time in reviewing its processes, their outcomes, and to not assume that doctors are infallible geniuses, who don't need any of that checklist nonsense. (Oh, hey, the patient just woke up, and told us that we removed the wrong lung...)
I would imagine that it would involve overlapping shifts so the handoff process can be extended over time.
The extreme case actually seems good; pair doctoring where shifts start four hours apart, so you spend the first half of your shift with one doctor, and the second half with another.
Maybe instead do 2 x three hours, and give doctors really reasonable work hours or time to keep up on current research.
I fear the conclusion to that would be managers saying "what are all these doctors doing, we need to lay 3/4ths of them off and make the rest work twice as much for 50% more pay!"
Which country is going to vote a LAW that writes: nobody works more than 45 hours/week?
The next day healthcare will collapse. Big corporations will lobby VERY hard against that.
You get license to use the software if you abide by the terms of the license agreement. A company that asks its employees to work 60 hour weeks breaks the license agreement, and therefore cannot use the software.
The only LAW that could be written about this would be "Contracts cannot include limits on the number of hours employers ask people to work".
So maybe, yes, the megacorps get that law passed. Until then this license is binding...
Now whether the any particular legal system cares to enforce the license terms is another matter
The limit is 48 hours, and it's calculated with some averaging over multiple days, but I hope that's close enough.
https://en.wikipedia.org/wiki/Working_Time_Directive_2003
By using the software, you agree to the terms of the license, and can in principle suffer a civil (non-criminal) lawsuit against you for breaching them.
You have no proof of this. The (x times y) work week for manual labourers was earned by a long history of labour activism.
Good working conditions for highly-skilled in-demand work arise naturally through supply and demand.
But they don't necessarily arise if the demand from workers wanting to do that work is sufficient. For instance, game developers have relatively terrible work conditions by the standards of the software industry, because so many people want to work on games. You could reasonably argue that they get enough satisfaction from being able to work on games that they're willing to put up with worse conditions, but that doesn't change that their work conditions are relatively bad. The nice thing about somehow bringing about a reduction in developer working hours is that it wouldn't necessarily result in any loss in economic output or employer profit, as there's a wealth of research showing that working beyond a certain amount of hours daily is counterproductive, so it's a win-win, and could even increase productivity.
So collective bargaining is unnatural? Except when a corporation does it on behalf of its shareholders, of course.
We can all pretend that we have more economic power than our highly paid blue collar peers that worked in highly unionized work forces in the past, but the programmers that make the kind of money for that to be true are very rare. Nearly 30 years ago my father made 70k as a union welder. It is highly unlikely that I will be able to match that salary street inflation, and the world is more expensive: education and health care could eat us alive at any moment.
I don't know how much of our effort the capital owning class needs to strip from us before we take advantage of the massive productive capacities we have before we make it fair again, but I assure you, it will be organized, not natural.
And you don't need to put double quotes around employees. They are employees and deserve to work in a good environment like everyone else
However, yes, I do believe that multiple 12+ hour days in a row is ultimately counterproductive. Also, I wonder if overlapping shifts would reduce the chance of errors.
https://www.theguardian.com/society/2009/nov/05/nhs-confiden...
> The report suggests the shorter working week is responsible for a lack of continuity in out-of-hours care
> shortening rota hours had led to more shift changes between medical teams and greater chances for mistakes to occur
> the early effects of the 48-hour European working time limit on surgeons and found that these new rotas had almost entirely removed adequate time for handover of sick patients
Also think about this - those doctors who don't do handovers, such as a dermatologist doing out-patient clinics, don't work long hours. Hospitals only do it where clinically required.
It's an insane culture.
My guess is that it is easy to spot error in handovers (X didn't say/write this, Y didn't hear/read that). But errors due that you have worked 11:30 hours are more difficult to isolate.
They don't do it for people who aren't doing handovers, like general practitioners who can work half-days just a couple of times a week if they want, or administrations. So we know it isn't inherent for doctors or for the staffing of the hospital. Only for those doing handovers...
Because the culture of US medical training was initiated by a coke addict, and it's extremely conservative and slow-moving.
https://en.wikipedia.org/wiki/William_Stewart_Halsted
https://en.wikipedia.org/wiki/Medical_resident_work_hours#EU...
https://www.theguardian.com/society/2009/nov/05/nhs-confiden...
In the UK they always worked long shifts because it’s safer. They’re trying to stop them, and it’s less safe.
Why do you think the reason the UK has always worked long hours is due to a drug addiction in the US rather than experience in safety?
It looks like they are not hiring enough senior staff, not handover problems.
We can choose to think that they independently arrived at an ideal state, but that's not reasonable.
https://github.com/996icu/996.ICU
https://996.icu/#/en_US
Even on HN: https://news.ycombinator.com/item?id=19507620
The link just shows you one particular file in the repo, but that should be obvious to anyone who ever used GitHub.
> The purpose of this license is to prevent 996-style companies from using the software or codes under the license, and force those companies to balance their work schedule.
Doesn't sound like protest to me, but like extortion. Either you do what we want, or you can't use our software.
> (P.S. This is a protest, not a legally binding license)
It is a legally binding license and not just mere protest. That's all I claimed.
So... just like any other license?
Not everything is sold for money.
The thing is, this kind of online protest is the safest for both initiators and participants. If by "other venues for labour activism" you mean go to the street, you're basically risk sending people to jail (And after the thing, nobody want to risk to get milled down by tanks).
Also, did you know that 996 is by law illegal in China? You can technically report your company if they force you to work 996. But: 1) After that, you will probably get fired for any number of excuses. 2) Your company is clever, they know many ways to get around the law.
Now, after all that dilemma, how could you protected yourself and help other developers just like you? The license idea come out naturally.
Why will the government avoid to use repressive measures against this? Otherwise they don't do their job of totalitarian government well enough. In a totalitarian state, I'd certainly be afraid to support this.
> Also, did you know that 996 is by law illegal in China? You can technically report your company if they force you to work 996. But: 1) After that, you will probably get fired for any number of excuses. 2) Your company is clever, they know many ways to get around the law.
Yeah, I've read the website. So technically, the license isn't needed.
> how to you protected yourself and help other developers just like you? The license idea come out naturally.
If you can't get the law enforced, how will you be able to enforce the license? I've heard that chinese courts generally don't meet western standards.
This is how the labor law been structured in China. In many cases you need to file the case by yourself in order to get it investigated.
> So technically, the license isn't needed.
> If you can't get the law enforced, how will you be able to enforce the license
I don't think you've understood the situation. In China, we play it safe.
You can get administration involved, but you will likely need to file the case by yourself, which may cause you to get fired. Now, if a third-party can help you by putting some pressure on your company, then both of you will be safe.
The license adds another reason to convince your boss to not implement 996.
This argument would work just as well for all of these other venues. That there's something else you could be doing is not a valid argument against doing the thing that you are doing.
"Why strike when you could change software licenses?"
Here, 12 hour shifts with 36 hour rest periods is also the norm for security guards, etc. (So that 4 can provide 24h security) That results in more than 45 hours every other week.
The license also doesn't specify how to remedy the situation in case it's ever violated, which feels very draconian and not very pragmatic.
Maybe you can try to file a pull request to suggest a change, remember to explain your situation and concerns inside the pull request so it won't be discarded.
Of course I understand ask some projects to suddenly change their license to support anti-996 is hard and unreal, but hey, the show of force and well there is already a win.
For me, the step they've taken to reach out for support (by ask others to change their code license etc) is already something positively progressive. Regardless whether or not they did successfully convinced anyone.
They got the message out, that's important.
I think a reduction of the 44h week is long due in Brazil and I don't understand why there isn't more pressure from workers. They probably don't have the time to protest.
I found the expectations around what you're supposed to do during work hours to be very different between Brazil and US.
Extreme conditions obviously exist, (e.g. telemarketers, which are usually _continuously_ taking calls) but from my observation a significant portion of those 44 hours is spent idling or socializing.
I think proposing any change in work hours under the current political and socioeconomic climate (12% unemployment) wouldn't be feasible, and without the needed shift in work culture our productivity would probably fall further behind.
Also, I don't know many people who work a strict 9 to 5 in office jobs, many take work home with them or are expected to always be available even on weekends should something come up. Not saying this is good, but I don't think strict 9-5 is the norm anymore.
Can you explain this notation?
https://en.m.wikipedia.org/wiki/Date_and_time_notation_in_Fr...
I think this was entirely the point of the license. The authors presumably believe that companies shouldn't make their workers work this long, and they don't want those companies to do to benefit from their work. Thus providing an (admittedly small) incentive for those companies to reduce their working hours.
_This_ license? Probably not.
Making your own license is hard. The best advice is probably don't try unless you're a licensing expert or have unlimited access to one.
Atleast in my life, shorter work hours tend to heavily inversely correlate with my work output.
I work 32h/wk (mon-thu), and I feel happy with my productivity, but also recognize that I spend tons of time in meetings, clarifying requirements, helping other devs, etc. These aren't all really my role, so I could theoretically optimize them away and hope the invisible tasks get picked up successfully by someone else, but I know we'd have worse outcomes if I wasn't involved and "just" did my explicitly assigned tasks.
Can someone with team lead/management responsibilities optimize for 'job done' this aggressively?
I will have an answer for you in a month because I am going to start trying this strategy out in my current job starting this week. :)
Did you have to interact with anyone?
I can imagine this can be challenging if another person you need to interact with follows the same principle with a 7am to 11am schedule.