Retaliate for... what? The whole point of "countries" is that other countries don't get to tell them how to run their country, unless there is a legally binding consortium agreement. If SK wants to make it hard or impossible for foreign countries to compete, that's literally up to them. It's their country. They make the rules.
If other countries give market access to Korean companies, then as a member of the WTO, South Korea's non-reciprocity comes with significant penalties.
This isn't about nimble startups being protected against Google and Apple. It's Naver, Samsung and Kakao Corporations. They sometimes exercise vast power over government and so such decisions are less democratic than oligarchic.
Let's remember that all this protectionism ends up costing consumers. UX design in Korean apps alone sucks so hard, and there's monopolistic pricing for lots of services. Consumers could benefit from alternatives.
You do realize that a country like the US does the same in many ways, just less on the nose - until 2 years ago that is, now doing even more blatantly than SK? And most of the countries in the world put up all sorts of visible and invisble barriers?
It was clickbait because the article, which I did read, did not support the contention that package managers are evil. Therefore "evil" seems to be used in a hyperbolic way to grab attention, which makes it clickbait, specifically ragebait.
I wouldn't class it as clickbait myself, but I will stand by the use of the word "evil". I am using evil in the very old fashioned sense: the privation of the good. Is the title provocative? Yes. But that's the point of the article in general. I am trying to argue that they are a net bad with virtually no good upsides to them for the programming world as a whole. They've automated something at scale which should not have been automated. And to be clear, there is no solution to the problems they are trying to solve, rather it's all about trade-offs.
I a little annoyed that HackerNews post renamed it to "A critique of package managers" because that implies very different connotations. I'd view an article written like that as if I have some criticisms that could be addressed, rather than the entire concept being bad from the start.
> I am trying to argue that they are a net bad with virtually no good upsides to them for the programming world as a whole.
What I'm saying is that you have failed in this argument. You hardly even attempt to make it. Thus clickbait.
You said "this is why I am saying it is evil, as it will send you to hell quicker."
Okay, so then it's up to you to prove this hell actually exists. But you don't. You just assert its existence -- "Dependency hell is a real thing which anyone who has worked on a large project has experienced." By framing it this way, you can dismiss anyone who claims to not have experienced this as not having sufficient experience. But reading the comments here, a lot of people have experienced a sort of "dependency hell" (the kind that's talked about in the wiki you link to) that is solved by package managers.
So that's why it's classed as clickbait -- you (admittedly) wrote a provocative headline that you don't even remotely back up.
FYI for the future since you're lamenting in many comments that people are misinterpreting you, this is why. Given that you don't really make an attempt to prove this dependency hell and package managers are evil, and you don't acknowledge anything good about them, it's reasonable to assume your bias is just that dependencies are evil at their core. It's actually the most charitable reading because otherwise you seem confused.
Then again, there is a trope going back to Knuth - "Premature optimization is the root of all evil" - which is an argument that it is not clickbait, but merely applying a pattern in discussions about computer programming.
Not sure how many civil project that takes hundred or thousands of engineers working on independent teams. I think sometimes people ignore and what scale we are building software.
Not sure if there are true heterogeneous server CPUs or not, but with Intel Speed Select Technology (SST) you can mimic some similar features. SST allows you to set some cores to be higher base frequency, turbo frequency, etc, in return for other cores having corresponding lower frequencies. Naturally, the cores are not inherently different here, it's just distributing the power/thermal loads differently, so it's not a true heterogeneous system.
Plenty of computing uses don't run third party potentially malicious code.
SMT, implemented well, can significantly increase execution unit usage in the face of memory latency.
Now, if it makes business sense to have cpus with such a major functionality that is only useful for render farms and other compute clusters is another question.
Funny enough clone and compile is easier now than ever before. You can ask a llm to create a docker to compile any random program and most of the time will be okay.
Recently I migrated some small bash scripts to Js thanks to bunjs, chose it because its simplicity, speed and nice out-of-the-box features (db scripts without external deps). Quite happy with it, now I dont to relearn bash syntax every few months when I need to do some minor changes. Also, Javascript is prob the language that llvms understand the best which is a cool thing on these days.
reply