- cross-posted to:
- linux@programming.dev
- cross-posted to:
- linux@programming.dev
Ubuntu 25.10’s transition to using Rust Coreutils in place of GNU Coreutils has uncovered a few performance issues so far with the Rust version being slower than the C-based GNU Coreutils. Fortunately there still are a few weeks to go until Ubuntu 25.10 releases as stable and upstream developers are working to address these performance gaps.
They articulated the reason and gave examples of precedence.
And you’re dismissing their voice as irrelevant, but as the consumer of the product, their voice is most critical, and more people should be aware of how corporations use their massive wealth to choke and starve open source competition out of existence despite building their products on open source work in the first place.
I continue to fail to see the issue with the author, the person whose actual labour goes into the software, not your labour, deciding that they are fine with their source code being used in any way the general public sees fit provided they simply credit the author and provide a copy of the MIT licence. If I give you something, you’re not stealing by accepting my gift. They’re choosing voluntarily to make their source code available under such a licence. If they weren’t okay with that, they would’ve chosen a copyleft licence.
That seems insanely entitled, but you’re allowed to not use non-copyleft software if you really care that much. The authors of permissively licensed software aren’t forcing you to use their software.
There are plenty of valid reasons to license a work as MIT or BSD or similar. Firstly, libraries are almost always going to be permissively licensed, not just because it allows proprietary software to use those libraries, but also because it allows permissively licensed FOSS to use those libraries. If I want to use a GPL library, it’s not just that I have to make my software FOSS, it’s that I have to make my software GPL specifically. If I want to make a FOSS MIT program, I can’t use any GPL libraries.
Secondly, sometimes it’s because, well, as the licence text provides, I don’t give a shit what you do with the code. I write lots of little tools that are just for myself and I share them in case they’re of use to someone else. If some big corpo uses it in their proprietary money-making machine it’s no shit off my back. It was just a little tool I wrote for myself and it doesn’t affect me if other people use it to make money.
I think GPL is reasonable if a lot of labour goes into a project, and you’d be discouraged from working on it if someone was leeching off of it for their proprietary software. But my MIT/BSD code requires 0 maintenance labour from me, and I don’t care to control how other people use it. That’s the whole point of MIT/BSD/Apache/etc. It’s the “don’t give a shit” licence.
You’re arguing with a strawman you created, no one made any statements about the author. They simply said no one should use the software.
The author can choose to use the MIT license, and we can choose not to use their software.
What do you think we’re saying here? We’re saying we’re choosing not to use the author’s software, what are you taking issue with exactly?
And we’ve articulated valid reasons not to replace GPL core libraries with MIT ones…
Good for you? This isn’t about you (the author)… It’s about us not wanting to use your work, which you seem to take offense to, as if you did us a favor. Talk about entitlement.
Solid justification for using it for coreutils you got there…
The original comment called it stealing. There’s nothing morally wrong with stealing, but regardless it’s not even stealing. It’s a stupid argument.
I’m obviously talking about not giving a shit about how people use it. Which makes sense for coreutils. Loads of people use it for loads of different purposes. The author shouldn’t care how people use it.
We’re obviously talking about corporations intentionally using open source software with the intention of eliminating it as competition, we aren’t talking about the literal definition of the word “stealing”, which you seem smart enough to be able to recognize, but you’re insisting on being pedantic.
You twisted that into an argument about what the author of the code has the right to do… No one gives a shit, and we aren’t obligated to support it.
That’s not what corporations do when they use MIT/BSD code. They rely on that code; it’s not their “competition”. Unless you are talking about stuff like WhatsApp using libsignal, where they do use code from a direct competitor, but that’s far less common, and also not going to have a negative effect on Signal. I can’t speak for Signal of course but they are probably quite happy with WhatsApp using libsignal, as it both spreads Signal’s beliefs about communication being E2EE, and it makes WhatsApp reliant upon Signal. FOSS projects like ffmpeg, curl, etc, are (reasonably!) happy that the entire industry depends on the tool they wrote. And they are kept alive because they are so widely depended upon. Corporations donate to FOSS projects because they need them.
wtf is a non-literal definition of “stealing”? The idea of stealing is stupid enough already, I can’t play your games to figure out how you extrapolate something sillier from it. I’m a communist. I don’t believe in private property and I don’t give a shit about stealing.
They steal labor. You would think someone with the username communism would instinctively understand that.
And no, it isn’t “far less common”, and you have already been provided with examples of what we mean, so I’m not playing your games of providing examples (that don’t even use the MIT license, but now I’m the one being pedantic).