Dusk OS Demo | Hacker News Reader

Dusk OS Demo

vimeo.com

23 pointsakkartik8 comments

Once I've read "Its primary purpose is to be maximally useful during the first stage of civilizational collapse," my mind started to wonder away from this blend of Forth and C. It is quite weird to think about such future. Is the novel simple and efficient OS going to be useful in this scenario?

Had some Terry Davis vibes.

Indeed, TempleOS had a dynamic runtime C (HolyC) compiler too. Perhaps when people see the world differently, they try to compile C code on fly.

Honestly, following total civilization collapse, I think computers are going to be the least of our problems. Finding a power source for your computer, being able to find and use devices to work with it, etc. seem like wastes of time unless you can secure long term access to water, shelter and food. Nice to have something around I guess, but I don't see that computers would be terribly useful unless you manage some stability beforehand

I don't believe this is meant as an OS for personal use.

This is designed to reduce the impacts of a possible collapse in the semiconductor supply chain, by enabling older and scavenged parts to be recycled and reused for critial applications.

You'd be using Dusk OS in some essential piece of infrastructure that has a microcontroller for which no spare parts are available anymore.

I don't exactly agree with the project author on the timeline for all of this, but I don't consider it useless or misguided.

Projects like this might enable reusing that old 8086 forgotten in a drawer for something useful instead of just post-apocalyptic jewelry. Sure, you can use a 8086-era OS as well, but if something far more sophisticated is available, you would prefer that.

I am skeptical of the utility on an PC-class machine of an OS that uses 1013KB sitting idle.

CollapseOS, from the same author, can self-host on Z80s with 8KB: http://collapseos.org/hardware.html

I was referring to it when talking about jewelry 8086s.

As far as I understand, Dusk is not finished yet and is meant fore larger machines (maybe 386 to Pentium I range, I dunno).

For these apocalypse scenarios (which, again, I don't exactly agree on the same timeline as the author), I can see how both of them could be very useful.

I am extremely excited about and invested in this project, but I don't necessarily agree with the author's predictions. Rather, I see the way Virgil's design constraints have led to a very cool, novel and interesting approach to operating systems -- something so simple that you could reasonably comprehend every aspect of it starting from bare metal, yet powerful enough that you could use it to do useful computing work. I think there is value in having a system that you can comprehend fully -- something that is totally impossible with something like modern Linux. It doesn't have the same advantages as a modern operating system, but it has advantages that those systems lack: the ability to do a more independent, self-sufficient, sustainable kind of computing.

Reply on news.ycombinator.com