GNU Hurd 0.9, GNU Mach 1.8, GNU MIG 1.8 Released (opens in new tab)

Basically Linux and then BSD took all the wind out of Hurds sails. A lot of that was timing, and also because those projects were much more attractive to big business who put a tremendous amount of development resources behind them.

As far as I know QNX is the only usable OS based on a microkernel.

> I'm also noting that Hurd is still only 32-bit.

I wonder if they'll make it to 64-bit before we start upgrading to 128-bit computers?

They also still publish SHA1 sums, definitely stuck in the past.

That's great. You make an excellent comparison between its promises vs success vs DNF's. disordinary inadvertently points out that, unlike Hurd, DNF eventually delivered and at least made some money selling 376,300 units. It's kind of a cutdown on DNF to compare it to Hurd which wasn't finished, had little adoption, and sold 0 units. Even MULTICS and SCOMP probably had more adoption in significant use cases despite them practically being case studies on things with low adoption due to expenses or tough tradeoffs.

Meanwhile, Minix 3 and Genode are going strong despite originally being made by only a handful of people for a tiny fraction of the time. Hurd is truly a dead end in marketing if the likes of DNF surpasses it in deliverables and adoption.

Hopefully it will run Project Xanadu

https://en.wikipedia.org/wiki/Project_Xanadu

(check the start date and first working version)

Unlike DNF it's not a commercial product, it relies on volunteers and corporate benefactors. For multiple reasons Linux got the brunt of OSS development resources behind it which wouldn't have helped Hurds schedule.

Most device drivers are more or less independent of CPU architecture. A PCI card is going to be the same PCI card no matter whether it's running on x86, Power, RISC-V or anything else.

How usable is it outside of hardware issues?

It is unlikely that this will ever take off.

As I understand it, the idea of this project is to build a microkernel and associated systems that might compete as an alternative to the Linux kernel.

This would allow things like, more elegant hot patching of the kernel, formal verification of core parts of the kernel, better hardware enforced isolation of different systems for security. I guess also some people like the idea of a GPL3 kernel against tivoisation.

I can see why someone might adopt a mature version of this for say, a safety or security critical system. Trouble is, until it is mature and battle tested, its not a good bet for those things. And without adoption, it's not gonna mature or get used.

It's technically interesting and competition is never bad.

A fringe operating system can drive innovation by taking risks that a mainstream OS would never. Ultimately any improvements will appear on main stream operating systems so even if it's never used we're going to see some benefits.

I agree. Given the massive success of BSD opensource L4, I think the only reason for Hurd's continued claim is its association with Stallman. Are there any technical merits here?

The niceness of the Hurd design, conceptually, is that the microkernel orchestrates system integrity, while even an unprivileged user has a lot of freedom as to what to run on top of that. They can launch drivers, file systems, run sandboxed kernels, and so forth.

we don't want til 2060 https://xkcd.com/1508/

Andrew Tannenbaum's Minix [1] was pretty much made for this purpose. The current version is no longer a "teaching OS", but there is a lot of documentation on earlier versions around, including the original 1987 book [2] that Tannenbaum wrote on operating systems (which comes with the Minix source code).

[1] https://en.wikipedia.org/wiki/MINIX

[2] https://en.m.wikipedia.org/wiki/Operating_Systems:_Design_an...

This is the best source of OS development info/tutorials: http://wiki.osdev.org/Main_Page

A lot of applications already work on the Hurd. There's GNU Guix for the Hurd as well. Support for applications is pretty great. It's hardware support that's lacking, but it's getting better with NetBSD's rump kernel.

Beware of using dang here, as it might confuse people with HN admin ;)