Oh hey, we've reached the "Metal" stage! <a href="https://www.destroyallsoftware.com/talks/the-birth-and-death-of-javascript" rel="nofollow">https://www.destroyallsoftware.com/talks/the-birth-and-death...</a>
Interesting to see how many things Gary got wrong.
Funny, I've always found it interesting how "on point" it was...<p>Granted, yeah, we never (or haven't yet) really transitioned to running "full legacy software" inside the browser, or at least it's not common place. That said, I've seen people compile Wine to wasm, Linux to wasm, and lots of other things to wasm, and run em in a browser. Many of the "fake" demos could be done for real now.<p>The one aspect that remains thoroughly farcical is an equivalent of Wine for OS X/Cocoa good enough to run a web browser. :-(<p>[edit] And asm.js kind of died on the vine. Not sure how to feel about that one. Wasm could he described as an evolution of the same idea, but in a lot of ways it's something entirely different.
I'm doing my part ∠(‘-‘)
I don't know if it's only me, but did this guy... Did this guy make a huge mistake?<p>I think he was trying to bend reality with words. I can see many apps that are running in electron on my laptop, each consuming 300MB+ (e.g. Spotify), while many other apps are written in native Swift for example, especially with the help of AI, giving the best performance possible...<p>Edit.<p>And prices of RAM nowadays...
Thank you for the reminder to do my yearly viewing of that video lol
Whoa, I haven not been following ASM.js stuff in any detail.<p>Seeing that Metal replaces kernel/userspace boundaries with VM protections for memory, meaning that system call overhead is eliminated, at the price of ASM/VM overhead.<p>What a fascinating idea. Kidding on the square...
Love this. An example of complete and total dominion over the machine. Great quote here too lol<p>> Prometheus stole fire from the gods and gave it to man. For this he was chained to a rock and tortured for eternity.
Can someone break this down for me? Looks like it's using... C? to load a js interpreter which bootstraps an API around all UEFI features? Do I have that right?<p>And, if so, does that mean that once the API has been bootstrapped, one could actually write an OS in js? Or are there other abstractions that would need to be migrated first?
You don't need a JS bootloader to write an OS in JS. The bootloader just drops the machine into some memory address for it to start executing your OS init script. that bit could be a Javascript interpreter. You can't do much with the architecture in Javascript though, because it doesn't allow you to map memory directly to your types (unless there's some ungodly nonesense I'm not aware of) so you'll have to drop into C/asm to e.g. interact with the ports/registers/tables to set up userspace.
> And, if so, does that mean that once the API has been bootstrapped, one could actually write an OS in js?<p>I bet somebody has done that.<p><a href="https://www.google.com/search?q=os+kernel+in+javascript" rel="nofollow">https://www.google.com/search?q=os+kernel+in+javascript</a><p>Seems like a small number of hobbyists have attempted.<p>I've heard of people doing this with other high level languages. Basically you need enough low level code to bootstrap a VM. Once you have that, you <i>can</i> make the high level language decide some logic that traditionally would be in C code, like manipulating page tables or whatever.
Automatic Garbage Collection in a kernel probably won't work:<p>I vaguely remember hearing about someone trying to use .Net in the Windows kernel.<p>The big problem is garbage collection: If I remember correctly, the fact that "any" operation can fail with an out of memory exception was a huge problem. Another problem was that random pauses for garbage collections in the kernel had major stability issues.<p>In short, I hope that the js kernel is for amusement and education; otherwise it would need a much more advanced garbage collector then earl 2000's .Net.
> I vaguely remember hearing about someone trying to use .Net in the Windows kernel.<p>Microsoft did that, it was called Longhorn. That release cycle was long delayed and they abandoned most of its ambitious projects, especially C# in the kernel, and the result was Windows Vista.<p>GC was not the only reason for the failure of that project. Someone could write a book about it. A lot of it was actually more about the organization of people. I also had heard from insiders that lack of ahead of time compilation was an issue. The other issue I remember hearing about was a complaint that Windows components were not layered cleanly and they ended up with circular dependencies when they tried to rewrite them.<p>I think it's possible to write a kernel with GC, and to still be judicious about memory usage with a GC language. And I say that as someone who happens to think that a big issue with modern software is that too many programmers are spending their whole education and career to depend on GC without thinking about it carefully. That is to say I'm already a skeptic of high-level languages and GC, but I will still afford that it is technically possible.
> I think it's possible to write a kernel with GC, and to still be judicious about memory usage with a GC language. ... but I will still afford that it is technically possible.<p>I need to split some hairs for a bit:<p>Do you mean what is colloquially referred to as "GC", as in the dotnet / Java / Javascript / golang "mark-and-sweep", fully-automatic style?<p>Or do you mean other automatic memory management systems, which some people technically define as GC, like automatic reference counting? (IE, they clean up memory immediately, and except for requiring some manual form of breaking cyclic loops, generally are fully automatic?)
Given that we are talking about JavaScript, I meant the former.<p>The latter is pretty much the status quo in terms of the state of the art. Most kernels aren't built with <i>automatic</i> reference counting, reference counted objects are plentiful even if more manual, but the automatic part is trivial to add to a C++ code base using smart pointers for example, and things like rust or apple ARC can also do it.<p>I think the former (a "true" GC that can resolve cycles) could be done in a kernel, it's just not common.<p>I'm not well versed in this historical example, did the "lisp machine" not have a GC in its OS?
Another part of it was, IIRC, that Longhorn was based off of the Windows XP core, i.e. the non-server stuff. While the Windows Server development continued apace, with lots of security and hardening to make for a reliable OS to build upon, the Windows Non-Server team continued with the existing mess of a codebase, not prioritizing security features or stability in favor of trying to manage feature creep. Longhorn was meant as a stopgap between XP and 'Blackcomb', but a lot of Blackcomb stuff started creeping backwards, bogging them down.<p>When security and reliability were suddenly key issues for Microsoft (to the extent that they ever were), it was obvious that what the Longhorn team had built was never going to meet that bar so they started over building off the Windows Server codebase instead.<p>Most of this story I remember from a video on YouTube of that old guy who worked at Microsoft since forever and left around the time of the Longhorn debacle, but a lot of it is corroborated in the Wikipedia article as well. <a href="https://en.wikipedia.org/wiki/Development_of_Windows_Vista" rel="nofollow">https://en.wikipedia.org/wiki/Development_of_Windows_Vista</a>
> trying to use .Net in the Windows kernel.<p>> Microsoft did that, it was called Longhorn<p>Do you have any reference for that? Or are you confusing Longhorn with Singularity (<a href="https://en.wikipedia.org/wiki/Singularity_(operating_system)" rel="nofollow">https://en.wikipedia.org/wiki/Singularity_(operating_system)</a>) / Midori (<a href="https://en.wikipedia.org/wiki/Midori_(operating_system)" rel="nofollow">https://en.wikipedia.org/wiki/Midori_(operating_system)</a>)?<p>I suspect you're referring to the shell/internals, though, not the kernel (<a href="https://longhorn.ms/the-reset/#:~:text=Why%20start%20over,reliability%20and%20security." rel="nofollow">https://longhorn.ms/the-reset/#:~:text=Why%20start%20over,re...</a>)
You'd need to write an entire hardware abstraction layer to do anything useful. There's projects that do this for microcontrollers - eg MicroPython and Espruino.
Yes, it would need support from lower level code. But then, so does C -- many things that an OS needs to do, such as installing interrupt handlers, changing the current page table pointer, jumping into a target process already in progress, etc., are not part of the C standard.
Should be able to do similar with MicroQuickJS or maybe just QuickJS...
I'm pretty sure someone already compiled Linux to asm.js a few years ago. As asm.js is/was a subset of JS, you could say it's already been done. In theory, you could continue work from there in JS.<p><a href="https://medium.com/@retrage/lkl-js-running-linux-kernel-on-javascript-directly-9bc8c7abfc5d" rel="nofollow">https://medium.com/@retrage/lkl-js-running-linux-kernel-on-j...</a>
Depending on your definition of OS, yeah you could do that :)
Hey, when Apple transitioned from m68k to PowerPC, it took them a hell of a long time to rewrite massive parts of their OS. It's a low bar, though...
OS in JS, ok I am interested now...
I presume you'll add the network stack next, so that I can use my favourite, most useful packages?<p><pre><code> import isOdd from "https://unpkg.com/is-odd";</code></pre>
Well, there's a network stack already there, including HTTP and HTTPS on newer firmwares.
We are getting isOdd in the bootloader before GTA-IV
> If this makes you grin, you are probably holding the torch.<p>What if it makes me recoil in horror? <i>screams into the void</i>
next step is to create a UEFI TUI using react (please don't)
Considering how bad some of the Gamer-ish firmware UIs are now, it might be an improvement.
you may just have casted a curse on our future motherboards, damn you
OMG we can use ink for that
This project will go places. Like every silly project not intended for production. :)
Does it manage to support floats? I am not sure if those can be safely used in the UEFI environment. (I recall GRUB’s build of Lua being integer-only, and Linux avoiding the use of floating-point arithmetic in kernel mode, but I don’t remember the reason.)
Wow, this is cursed.
"The Birth and Death of JavaScript" is coming true after all.
I think there are two philosophies here:<p>1) JavaScript must stay in the box (aka in the browser).<p>2) JavaScript as a general purpose programming language.<p>While I can absolutely understand 1), I have had wanted to access
the filesystem via JavaScript, just as I do via ruby or python, for
local use only. After I googled for a while, they would say that
this is not possible unless one uses npm/node. I think this shows
that there are use cases here and the "default" JavaScript, aka 1),
does not cover these. I do not like JavaScript, but based on my
own use cases, I actually favour 2) far more than 1). So from that
point of view, being able to access UEFI can also be useful. So
why not.
> I have had wanted to access the filesystem via JavaScript, just as I do via ruby or python<p>There are some (limited) ways to do so now: <a href="https://developer.mozilla.org/en-US/docs/Web/API/File_System_API" rel="nofollow">https://developer.mozilla.org/en-US/docs/Web/API/File_System...</a><p>As for (1) vs (2), it’s not really an issue of JavaScript at all. The main question is, do you want to build something that runs in a browser? If you’re building a web app, you’ll have to use the sandboxed APIs (and probably JavaScript). If you don’t care about the runtime, yeah, you can use Node or Bun or Deno (or use another language altogether).
> I googled for a while, they would say that this is not possible unless one uses npm/node<p>Gnome Shell and Firefox/SeaMonkey/Mozilla Application Suite/Netscape 6+ (and Zotero[1]) are implemented on top of SpiderMonkey.<p>1. <<a href="https://news.ycombinator.com/item?id=46735616">https://news.ycombinator.com/item?id=46735616</a>>
I've been using Deno a <i>LOT</i> for general shell scripting... it's been pretty nice in general. FWIW, Node, Bun and Deno have FS interfaces in the box, so yes, you can do it without npm modules. Though Deno allows you to directly reference the modules/repos from the script without needing a separate install step, package.json or node_modules directory.<p>It's also a single, self-updating executable and includes a lot in the box. Including SQLite3.
You are missing one option:<p>0) JavaScript must be abolished from the browser
Try webkitdirectory file attribute for browser access to the file system.
This is both so impressive and cursed that I'm not sure how to feel.
This is hilarious lol, it’ll be any day now before we get a full JS kernel. Garbage collection could be an obstacle, but I know there have been some kernels written in Go/Java before
It could be even better!<p><a href="https://en.wikipedia.org/wiki/Java_processor" rel="nofollow">https://en.wikipedia.org/wiki/Java_processor</a>
Who needs to garbage collect? Just leak memory until the system dies! That strategy seems to be good enough for claude code, anyway.
Turning in the widening gyre, the falcon cannot hear the falconer. The center cannot hold.. The old prophecy is coming true.
Finally I will be able to have graphical bootup logo like in The Hackers?
The choice of Duktape here is smart — it's one of the few JS engines that can actually run freestanding with minimal libc stubs, since it was designed for embedding in constrained environments. V8 or SpiderMonkey would be a nightmare to get running pre-boot.<p>What I find most interesting is the UEFI services binding approach. Rather than trying to abstract away the hardware, it exposes the raw EFI protocols (GraphicsOutput, SimpleFileSystem, etc.) directly to JS. That's a much more pragmatic design than trying to build a full HAL — you get to prototype UEFI applications rapidly while keeping the escape hatch to C for anything performance-critical.<p>Would love to see if anyone tries hooking this into UEFI's built-in network stack for PXE boot scripting. That could actually be useful beyond the novelty factor.
I’m always amazed and slightly envious of what programming languages with large developer bases can do. I mean if a language is Turing complete it can do anything, but JavaScript takes this to the extreme.<p>Mind you I never said anything about quality or performance, obviously doing everything in JavaScript comes with it’s own issues but if you were to say that someone got JavaScript running in the Linux kernel as a POC I wouldn’t even be surprised
Could this be used as a learning tool? Rebooting the computer takes so much more time compared to reloading the browser tab. And you probably can't brick your computer.
Can't wait for browser support for this... ;-)
Yeah, but your [developers] were so preoccupied with whether or not they could, they didn't stop to think if they should.
This is incredible.
> If this makes you grin you are probably holding a torch<p>Hilarious
>boot sector<p>>looks inside<p>>node modules
"Your scientists were so preoccupied with whether or not they could, they didn't stop to think if they should."<p>Pretty neat, though.
Your scientists were so preoccupied with whether or not they could, they didn't stop to think if they should!
Awesome! Everything will be rewritten in JS
Beautiful
Cursed, but fun
It begins!
Finally!
I love it.
But why?
Javascript is a horrible choice but I think having a scripting language for this is actually quite a good idea. If only there was a popular scripting language that didn't totally suck balls.
good. god.
Cursed