Real programmers will write in a way that user’s resources are not being wasted because you need a full browser, a JS runtime, and DOM juggling, to show even the simplest application.
It’s not rare for simple JS applications to consume over half a gigabyte of RAM on startup, and way more CPU than their native counterparts. That this was normalized and even defended is stupid.
I think you’re thinking of Electron apps, but that’s not really a criticism of JavaScript, that’s a criticism of Electron. There are plenty of JS platforms that don’t require a browser/DOM. React Native is the biggest example. Also, GJS if you want native Linux apps.
But why Node? Node cost five seconds just to start up back when I worked on my embedded ARM v7 platform, and on modern x86_64 computers, npx anything takes just as long. Why rely on another runtime? Why not native binaries instead?
Because you have to figure out how to build them. And with that I mean, how do you make sure that whatever you’re doing will work and work the same way not only on your “embebbed ARM v7” architecture and all the other CPU architectures, but also the operating system libraries included? How do you make it work the same way on Mac, Windows 7, Windows 11, Ubuntu, custom Arch installations, FreeBSD, etc etc?
If you build native binaries, you personally are the one who has to make sure it runs. This means (depending on how much you want to support) a lot of development or support time. (Or you make your users build it themselves and fix errors, which means a massively reduced userbase, good luck with adoption…)
If you use Node, (or other virtual machines) you literally don’t have to do anything, because it just works.
What you mentioned is compatibility across platforms. A program written in C is also guaranteed to run on all the systems you mentioned, given that the system has a C compiler and libc that stick to the standard. You, the programmer, does not have to anything to “make sure” your program works.
We’ve invented high-level programming languages like C 53 years ago, just to get away from assembly, and to avoid dealing with the “cross-platform” problem you mentioned, remember?
Real programmers will write in a way that user’s resources are not being wasted because you need a full browser, a JS runtime, and DOM juggling, to show even the simplest application.
It’s not rare for simple JS applications to consume over half a gigabyte of RAM on startup, and way more CPU than their native counterparts. That this was normalized and even defended is stupid.
I think you’re thinking of Electron apps, but that’s not really a criticism of JavaScript, that’s a criticism of Electron. There are plenty of JS platforms that don’t require a browser/DOM. React Native is the biggest example. Also, GJS if you want native Linux apps.
Node does not require an excessive amount of resources.
But why Node? Node cost five seconds just to start up back when I worked on my embedded ARM v7 platform, and on modern x86_64 computers,
npxanything takes just as long. Why rely on another runtime? Why not native binaries instead?Because you have to figure out how to build them. And with that I mean, how do you make sure that whatever you’re doing will work and work the same way not only on your “embebbed ARM v7” architecture and all the other CPU architectures, but also the operating system libraries included? How do you make it work the same way on Mac, Windows 7, Windows 11, Ubuntu, custom Arch installations, FreeBSD, etc etc?
If you build native binaries, you personally are the one who has to make sure it runs. This means (depending on how much you want to support) a lot of development or support time. (Or you make your users build it themselves and fix errors, which means a massively reduced userbase, good luck with adoption…)
If you use Node, (or other virtual machines) you literally don’t have to do anything, because it just works.
You really don’t see the value in that?
What you mentioned is compatibility across platforms. A program written in C is also guaranteed to run on all the systems you mentioned, given that the system has a C compiler and libc that stick to the standard. You, the programmer, does not have to anything to “make sure” your program works.
See this insane list of platforms GCC supports.
https://en.wikipedia.org/wiki/GNU_Compiler_Collection#Architectures
We’ve invented high-level programming languages like C 53 years ago, just to get away from assembly, and to avoid dealing with the “cross-platform” problem you mentioned, remember?