Most complaint against Rust is fucking culture war, not technical, so people who actually have technical concerns with Rust are being lumped together with Brian Lunduke and others.
I guess it largely depends where one looks, to pick up that kind of perception.
There are technical discussions out there.
There’s much I admire about rust (viewed from afar, yet to even try coding anything in it yet), some of it tugs me away from my beloved Haskell, and then some of it repels me again. And regardless of either, there are aspects that cause chagrin about it being put in things like the linux kernel and replacements for coreutils, and other essential things, not least of which is how much further that takes us away from curt bytecode. Cant really do the likes of tinycore so easily with rustification, afaiu. But I understand many aspects are very alluring to coders. And then a lot of it gets overhyped… with over-eager premature implementation, causing harms, including to perceptions of rust. Gently does it. Tortoise wins the race.
I think it’s silly to be against Rust code in the kernel because it’s not C or whatever. Though I do agree with the criticisms of Rust projects shipping with the MIT license instead of GPL.
GPL code can also be used for commercial and military use. What are you smoking where you think that is even remotely true? Genuinely asking. It feels like people on your side of the argument have all learned what you have from the same, ill informed source.
GPL impose that derivative work is GPL, not MIT. Indeed, a commercial and a military use could embed work under GPL licence, but I suppose to point ou which part is under GPL. Which is more complicated than MIT, that you could use and label under any licence you want.
I agree that it suppose that those organisations give a shit about law (clearly, they don’t), however a GPL infraction could be used against those organisation, next to other infractions for a trial and type of of actions.
If I’m a military supplier of nukes to the government, I can freely use GPL and there’s no legal issue with that. You cannot request the nuclear launch software or the guidance control software even if they use GPL licensed code within it. Why? Because they don’t distribute said code to the public. If you develop something for private use, and it never gets a public release there’s no obligation or requirement to release the source! This is especially true for a government contractor that only makes software for a single customer (the government).
I think we’re agreeing that your claim was nonsense at this point, but I still don’t understand where people get these strange ideas about how GPL stops commercial or military use outside of very specific and frankly niche ways. If this is your reason for preferring GPL, it’s poorly thought out.
In purely private (or internal) use—with no sales and no distribution—the software code may be modified and parts reused without requiring the source code to be released. For sales or distribution, the entire source code needs to be made available to end users, including any code changes and additions—in that case, copyleft is applied to ensure that end users retain the freedoms defined above.
GPL does not prevent using code for commercial or military use. Commerce and military deserve freedom in their computing too. ;)
It’s perhaps more accurate to say GPL (and all copyleft licenses) be free and free alike, whereas MIT (and all permisive licenses) be free to unfree.
Copyleft free software licenses keep the FAIF (Free as in freedom).
Permissive free software licenses only assure the original avails the same freedom to the users.
So yeah, better to stay on the copyleft side of free software licensing, if wanting to ensure it stays assuring users of their 4 freedoms, to use, study, share and change the software, such that released changes are shared with the same freedoms, not with extra restrictions tacked-on like can more easily happen with permissive licenses.
… And it’s right to feel ones tin-foil hat get itchy, when there’s such a replacement plan coming from the corporations, like they’re gearing for a future total usurpation.
Rust was among the first more well known projects, which adopted a Code of Conduct, then grifters in the OSS community cried censorship, which made people flock to it to “own the right”. Even if I think it’s an overrated marriage of flesh between C and OCaml, Code of Conducts are generally a good thing, and the people who really like toxic callouts arre more of an anomaly, and likely were flown there due to the culture war stuff.
People hate rust because of its fences/training wheels, not because it’s “woke”
…actually I just saw someone in this very comment section ranting about “soydevs”, you’re not wrong. But there are valid complaints too! Some of us are just old and think our computers should do whatever we tell them up to and including “shit yourself and catch fire”
But there are valid complaints too! Some of us are just old and think our computers should do whatever we tell them up to and including “shit yourself and catch fire”
I agree with this, but I think rust is fine here. It has unsafe as a keyword that let’s you do the breaking stuff, it just makes sure you know you’re doing something dangerous and makes it stand out for code reviewers
That’s why I prefer D’s approach for memory safety. Sure, it’s not as well featured as Rust’s, however it has “fat pointers”, three levels of safety, and kind of optional garbage collector.
Oh no, I bet you there’s a fair few who think rust is woke and deeply connecting with programming socks --> it should be cancelled.
IMO it’s also an ideological fit for the current right wingers who connect Rust to progress and use progress as an insult. Since they can’t fight Rust on its merits, they have to make up whatever argument they can that fits their narrative.
Somehow yes. It started as reasonable criticism but people like Brian Lunduke managed to interpret C and X as “conservative” while rust and wayland as “progressive.”
He was even criticising rust projects for “having too many people with anime profile pictures” in one of my youtube recommendations.
Most complaint against Rust is fucking culture war, not technical, so people who actually have technical concerns with Rust are being lumped together with Brian Lunduke and others.
I dont know about “most”.
I guess it largely depends where one looks, to pick up that kind of perception.
There are technical discussions out there.
There’s much I admire about rust (viewed from afar, yet to even try coding anything in it yet), some of it tugs me away from my beloved Haskell, and then some of it repels me again. And regardless of either, there are aspects that cause chagrin about it being put in things like the linux kernel and replacements for coreutils, and other essential things, not least of which is how much further that takes us away from curt bytecode. Cant really do the likes of tinycore so easily with rustification, afaiu. But I understand many aspects are very alluring to coders. And then a lot of it gets overhyped… with over-eager premature implementation, causing harms, including to perceptions of rust. Gently does it. Tortoise wins the race.
I think it’s silly to be against Rust code in the kernel because it’s not C or whatever. Though I do agree with the criticisms of Rust projects shipping with the MIT license instead of GPL.
https://users.rust-lang.org/t/im-shocked-the-rust-community-is-pushing-an-mit-licensed-rust-rewrite-of-gnu-coreutils/126110
I hate the culture war stuff. I also hate that the Rust core utils rewrite was done under an MIT license instead of GPL.
A gain of memory safety with a poison pill of permissive licensing is no gain at all.
How is MIT poisonous? It’s still FOSS
Until it isn’t. MIT is permissive and allow to use code for commercial or military use. GPL work have to stay GPL
That free work for corpo
GPL code can also be used for commercial and military use. What are you smoking where you think that is even remotely true? Genuinely asking. It feels like people on your side of the argument have all learned what you have from the same, ill informed source.
GPL impose that derivative work is GPL, not MIT. Indeed, a commercial and a military use could embed work under GPL licence, but I suppose to point ou which part is under GPL. Which is more complicated than MIT, that you could use and label under any licence you want.
I agree that it suppose that those organisations give a shit about law (clearly, they don’t), however a GPL infraction could be used against those organisation, next to other infractions for a trial and type of of actions.
If I’m a military supplier of nukes to the government, I can freely use GPL and there’s no legal issue with that. You cannot request the nuclear launch software or the guidance control software even if they use GPL licensed code within it. Why? Because they don’t distribute said code to the public. If you develop something for private use, and it never gets a public release there’s no obligation or requirement to release the source! This is especially true for a government contractor that only makes software for a single customer (the government).
I think we’re agreeing that your claim was nonsense at this point, but I still don’t understand where people get these strange ideas about how GPL stops commercial or military use outside of very specific and frankly niche ways. If this is your reason for preferring GPL, it’s poorly thought out.
GPL does not prevent using code for commercial or military use. Commerce and military deserve freedom in their computing too. ;)
It’s perhaps more accurate to say GPL (and all copyleft licenses) be free and free alike, whereas MIT (and all permisive licenses) be free to unfree.
Copyleft free software licenses keep the FAIF (Free as in freedom).
Permissive free software licenses only assure the original avails the same freedom to the users.
So yeah, better to stay on the copyleft side of free software licensing, if wanting to ensure it stays assuring users of their 4 freedoms, to use, study, share and change the software, such that released changes are shared with the same freedoms, not with extra restrictions tacked-on like can more easily happen with permissive licenses.
… And it’s right to feel ones tin-foil hat get itchy, when there’s such a replacement plan coming from the corporations, like they’re gearing for a future total usurpation.
I’ve worked with a lot of devs. I’ve seen a group of devs invent a new language to keep from having to learn a new off-the-shelf language.
I’ve seen devops rip out entire working systems and work on replacement python for months rather than coming up to speed on existing stuff.
It honestly think a lot of it comes from the poor perception of starting over from scratch on someone else’s code vs on your own code.
As in Rust = gay kind of culture war? I’m so OOTL. Please help
Rust was among the first more well known projects, which adopted a Code of Conduct, then grifters in the OSS community cried censorship, which made people flock to it to “own the right”. Even if I think it’s an overrated marriage of flesh between C and OCaml, Code of Conducts are generally a good thing, and the people who really like toxic callouts arre more of an anomaly, and likely were flown there due to the culture war stuff.
People hate rust because of its fences/training wheels, not because it’s “woke”
…actually I just saw someone in this very comment section ranting about “soydevs”, you’re not wrong. But there are valid complaints too! Some of us are just old and think our computers should do whatever we tell them up to and including “shit yourself and catch fire”
I agree with this, but I think rust is fine here. It has
unsafeas a keyword that let’s you do the breaking stuff, it just makes sure you know you’re doing something dangerous and makes it stand out for code reviewersThat’s why I prefer D’s approach for memory safety. Sure, it’s not as well featured as Rust’s, however it has “fat pointers”, three levels of safety, and kind of optional garbage collector.
The whole reason I use Linux is to shit myself and catch fire.
Fucking up is part of the fun. Y’all remember when computers used to be fun?
Oh no, I bet you there’s a fair few who think rust is woke and deeply connecting with programming socks --> it should be cancelled.
IMO it’s also an ideological fit for the current right wingers who connect Rust to progress and use progress as an insult. Since they can’t fight Rust on its merits, they have to make up whatever argument they can that fits their narrative.
I feel like you either didn’t read my comment or have been smoking a little too much weed lol
Somehow yes. It started as reasonable criticism but people like Brian Lunduke managed to interpret C and X as “conservative” while rust and wayland as “progressive.”
He was even criticising rust projects for “having too many people with anime profile pictures” in one of my youtube recommendations.