My impression is that it is like wearing a face mask when you are sick. You should wear a mask to protect the others, but people generally don't give a shit.

There is no extra burden in choosing a copyleft license for your project (you still have the copyright, you can do whatever you want with it). What creates extra burden is when you use copyleft dependencies. But that's not your choice.

And as a user, I'm pretty sure you are happier if code that you don't own is open source. All devs should push for copyleft licenses because it benefits them as users. But for some reason many are pushing for whatever is best for companies who employ them.

The problem is you can't get the benefits without the cost. And all of the derivative works problems are political -- the issue is that companies don't want to (or can't for various reasons) give you their changes. Personally I view this as antisocial and shortsighted behaviour (when a company fails, the work in their fork is lost).

Though to be honest, most proprietary changes aren't very good. The GPL is just one more (political) hiccup that gives an incentive for companies to contribute upstream -- in addition to general out-of-tree problems.

I think this was probably more of a critical mass sort of thing. Historically, it seems like copyleft licenses (and perhaps even GCC's horribly byzantine internal architecture) probably played a key role in getting free software off the ground. Once you have sufficient mind share though it no longer seems to be as necessary.

All technical discussions will end in license arguments. Put simply, I don't agree (obviously). What I don't understand is this viceral hatred of copyleft to the point where you will make even weak forms of copyleft unfriendly in a language ecosystem. Aren't languages meant to be agnostic to license arguments...

At the end of the day, I don't care either way. I make all of my standalone Go and Rust code GPLv3. It's just a shame that people are willing to stunt their own languages just to make a point about licensing.

The obvious solution to this is to use the AGPLv3 license for all or most software one writes, which is what I prefer to do, as it's the strongest copyleft license available currently.

We see in many circles that it's not the hip thing to do, to write copyleft software. No, the hip thing to do is write permissively-licensed software and either not care about companies using it as free labor or complaining about a ''social contract'' that wants to be implied, but never explicitly stated as copyleft does.

I believe the trend towards permissive licensing has been pushed mostly by corporate interests and now we see the result. I believe things would be better if every company had to either share its code as copyleft, write everything itself, or actually pay for other proprietary software it could incorporate into its own. That is, I'd rather Grammarly, as an example, have no real option but to purchase an Allegro or Lispworks license, rather than being able to use SBCL and contribute little or nothing.

No, it seems horrible. We need more free software, not less.

That's not the reason why someone would choose a copyleft license of the kind the FSF advocates, like the GPL or one of its variants.

You choose a copyleft license because you primarily care about freedom.

I rarely bother to use copyleft for my personal open source work, because I'm not typically interested in hiring lawyers to try to enforce compliance. And if I wouldn't pay a lawyer to go after a violator, then I don't personally see much point in using anything more complicated than CC0 or a boring permissive license.

Licensing adds overhead. If I use the GPL, then somebody has to stop and ask themselves, "Hey, can I use reuse 30 lines of this GPLed code in this MIT-licensed project? Do I need to change my project's license? Do I need to ask permission?" Usually I don't care.

However, for certain projects, I am perfectly willing to enforce a copyleft license. In that case, I'm happy to use the GPL 3. It's a good license, and I'm very glad it exists.

I don't feel like anybody else should feel obligated to take my approach. Part of my philosophy here comes from painful personal experience that successfully commercializing code requires years of sacrifice, and 99% of what I write in my free time will never be worth that effort.

I want the source code, so I want developers to push for copyleft. I am a developer, so I push for copyleft on my end, for the others who want the source code like me.

> even the best programmers would rather just pay money to not have to fix someone else's bugs

I regularly need to patch some (open source) software I use. I probably couldn't pay for it (not sure if anyone would fix it for me), so I don't have a choice. The proprietary alternative is that I just can't have the damn software fixed.

> If we're being pragmatic about what is good "for the user" then the best thing for the user is that there are as few obstacles as possible for developers to create products for the users to use.

When you make it as easy as possible for developers, you end up with ElectronJS crap and similar. With proprietary protocols, nobody can write a different client (e.g. to make it more accessible). I don't consider this good for the user.

> If a company doesn't like GPL, they'll simply not use GPL code, and maybe not using GPL means their product is not viable anymore

First, there is not only GPL. The weakest copyleft I know is MPLv2, which requires to share the modified files. If a company's product is not viable under those conditions, then that product is simply worthless.

Second, as an open source developer, I don't really give a shit if another company that is not paying me is not viable with my copyleft license. I am not working to help them make money while making the world less convenient for me.

> I have to add this feels so awkward because every time there's a thread about AI you'll find someone saying that the fact that data is copy-able renders copyright a thing of the past

It's not weird: I don't want AI to train from my code without permission. BigTech can abuse my licenses because they are too powerful for me to do anything about it. It's just a shameful workaround, and I am hoping that new licenses will come out that explicitly forbid ML training, so that I can update my licenses.

> everyone's like "if you didn't want to get scrapped, shouldn't have posted it on the internet."

I surely don't say that. People who say that have no idea how copyright works, there is nothing worth discussing there.

I don't want to make a political message here, but this is essentially the risk you take with a more permissive license. Someone absolutely can come along and build a product that subsumes yours. They can create proprietary extensions that you can not use, essentially gaining the benefit of your work without having to return the same favour. Basically, you then have to compete against yourself.

With a copyleft style license, it is much harder (though not completely impossible) for someone to do that. You will most likely be able to use their extensions and will be able to compete on a basis of who is executing better.

There are serious downsides to copyleft style licenses for things like containerisation, though. The barrier to entry is really large and you are going to cut out a lot of players who could help you. So it's really a matter of strategy. But if you are going for a permissive license, you'd better have all your ducks in a row because you should expect something like this to happen.

Free and open source business models are still pretty naive these days and I think it's going to take a few more decades before we have a really good idea of the best way to proceed.

Yes, the GPL is very important. The BSD license _is_ a problem. Those of you that were around when FreeBSD was maybe going to 'win' remember why Linux(GPL) ended up capturing everyone's attention.

You can not count on your code staying free if it's under a BSD license. If there is a way to make money on it, someone will fork it privately and force you to pay for it. If they don't succeed, you will still be under continual threats that you will have to pay for it.

He always wanted a copyleft license because it enabled kernel development to really take off by enabling him to make use of other people's contributions, and he's always been consistent about that.

I bet that what he really regrets, though, is not requiring copyright assignment for inclusion into mainline, so rogue developers can't just join up with the SFC and sue people against his wishes.

> This is why we have "Copyleft".

This. Exactly. It's suprising how many developers have strong anti-copyleft/anti-GPL opinions while being completely uninformed on what they're talking about (but hey, I guess "uninformed but strongly opinionated" is HN in a nutshell). The purpose of GPL and other copyleft licenses is exactly to combat the insanity of intellectual slavery.

Is there a current need for the FSF or alternatives? There's a wide variety of licenses already developed and between GPL2 and 3, LGPLs, AGPL, BSD/MIT, Apache, Mozilla, and normal commercial licenses, plus or minus PATENTS files, I'm not sure we need an organization that is developing another license, plus advocacy, plus developing a collection of fairly unrelated software at this point. I guess you could say the Apache foundation almost does the same thing, although I don't think they did much advocacy to use their license.

Yes, FSF was important, but when it winds down, it will not be replaced by one organization.

That's the entire point of copyleft licenses. Businesses want to benefit from the community without giving back.

The main point of copyleft is to pass down freedoms to use/modify/distribute all the way to the end user.

Instead we got locked-down privacy-breaching smartphones, IoT devices, SaaS, where only the manufacturer benefits from OSS.

> GPL license attempts to dictate to me what and how I am to do with my own code I write ("you must license all derived work under the GPL").

> I do not accept that, even if I firmly believe that open sourcing software is the right thing to do, and believe that open source, free software is the way to go.

The CDDL has the same requirement, except it's file-based. What are you even talking about FFS.

Linux, GNOME, VLC, WordPress... I hear this a lot but I really think it's overblown. There's plenty of GPL software with lots of users.

And so many of the ecosystem problems people talk about with FOSS would be much less of a problem with copyleft licenses.

I promise I won't debate this further, I just wanted to say that I also dislike the concept of intellectual property. The reason I like copyleft is because it takes the draconian machinery of copyright and puts it to work protecting users.

Yes, it inconveniences developers, but I think users are more important. You wouldn't prioritise the "right to jerry-rig" of bridge architects over the people who use their bridges.

> You say "stunt," I say "free."

I'm not sure how making it harder to change out parts of a compiled program makes the language "more free". It's a feature that's missing.

> Languages are permissively licensed to drive adoption.

I don't disagree with that (though I think compiler _implementations_ should be copylefted). But that's not the point, my point was that if you can't replace parts of a compiled program then LGPL loses its appeal.