Multiplix, operating system kernel for RISC-V and AArch64 SBCs | Hacker News Reader

Multiplix, operating system kernel for RISC-V and AArch64 SBCs

github.com

70 pointsyawniek26 comments

Walter Bright would be delighted. He's the author of the D language.

He often posts here

That's true. And I liked his Zorland C compiler when it first came out and didn't cost a fortune to buy.

Datalight C, later acquired by Zortech.

Zortech C++ was the first C++ compiler available for PCs. I bought it even though I had a Mac, and ran it in emulation, to learn C++. Apple had their AT&T Cfront-based MPW C++ compiler about half a year later I think.

Then Symantec acquired Zortech, and totally destroyed the compiler in version 6, immediately driving everyone to the new Metrowerks just as PowerPC came out.

Originally there was Multix.

Then Unix was named as a reference to it.

And now Multiplix is named as a reference to Unix.

We've come full circle, just about.

I had the idea of making a Unix clone called "Mimix" a few years ago. Never got around to it, but thought the name was clever.

> Originally there was Multix.

Multics.

this is really exciting stuff. looking forward to seeing future progress on this project

Great to see a D based effort.

I wonder why the developer picked the D language?

I wonder why that's something worth wondering about.

Maybe they like it? It's a perfectly reasonable systems language, even if it's not my personal favorite. It might be something questionable if they used Fortran or Algol-68, sure.

I think it's a fair question to ask; what particular qualities of D drew them off the beaten path for this kind of task?

Now, I know the answer as a former D acolyte - it's an interesting language with powerful abstractions, albeit hobbled in some ways - but most people don't, so I think it a reasonable question to ask.

I guess the way I read it was my like "why would you choose D when Rust and Zig exist"...but you're right, it could have been a perfectly curious question.

Fortran and Algol 68 were used to write operating systems before C was invented, hardly questionable.

Starting a new OS in those languages today would be heavily questionable.

Same applies to C, 10 years younger than them, less capable in language security, in an highly connected world, yet people keep doing it.

You're really pettyfogging simply to be contrarian. C is a language with a large and active community, and is still the most common/popular language in its fields. Algol-68 and Fortran are not. It has nothing to do with age of the language and everything to do with modern-day usefilledness.

The fact that you're unable to distinguish the difference means you're being purposefully obtuse and annoying or are simply naive.

A usefulness out of historical accidents and sunken cost.

Which has absolutely nothing to do with the original point.

It definitely has everything to do with sensible decisions.

Ok.

> The fact that you're unable to distinguish the difference means you're being purposefully obtuse and annoying or are simply naive.

This part is really unnecessary. Parent is most likely not "unable to distinguish" and name calling poisons the air and cheapens your other point which I think is good.

Fair point, I apologize for the petty comments. They were unnecessary. I sometimes have a low bar for intentional contrarianism/obtuseness at the expense of useful discourse.

Author's talk at dconf (from today, actually) may be of interest: https://www.youtube.com/live/wXTlafzlJVY?feature=shared&t=24...

(Timestamp is 6:45:08)

The author elaborates more in this blog post…

https://zyedidia.github.io/blog/posts/1-d-baremetal/

Cant speak for the author, but I would pick D for anything where C++ is appropriate, but I want to have less hassle (and dont need any dependencies).

D is very, very fun.

Reply on news.ycombinator.com