From: torvalds@klaava.Helsinki.FI (Linus Benedict Torvalds)
Newsgroups: comp.os.minix
Subject: What would you like to see most in minix?
Date: 25 Aug 91 20:57:08 GMT
Hello everybody out there using minix -
I'm doing a (free) operating system (just a hobby, won't be big and
professional like gnu) for 386(486) AT clones...
...PS. Yes - it's free of any minix code, and it has a multi-threaded fs.
It is NOT protable (uses 386 task switching etc), and it probably never
will support anything other than AT-harddisks...
100% of the user space programs were from gnu. Compiler was from gnu
The initial filesystem was minix
The kernel was Linux.
So I would say Linus had 100s of man years of ecosystem he could benefit from and that is the oss spirit.
But yeah he did not write a shell or an editor or Compiler etc and the kernel was of course oriented towards an unix like os.imagine you would have had to start from scratch, if you had just had a PC and a bios.
This entire site is a propaganda piece for one of the largest incubators in tech. Just because it has “hacker” in the name doesn’t mean it represents tech.
Like, the point is tech making money. You’re in the wrong place if you’re looking for something other than that.
I'd just like to interject for a moment. What you're referring to as Linux, is in fact systemd/Linux, or as I've recently taken to calling it, systemd plus Linux.
Linux is not an operating system unto itself, but rather one more free component of a fully functioning systemd system, made useful by the systemd unit files, target definitions and vital service managers comprising a full OS as defined by Lennart Poettering.
Many computer users run a modified version of the systemd system every day without realizing it. Through a peculiar turn of events, the version of systemd which is widely used today is often called "Linux," and many of its users are not aware that it is basically the systemd system, developed by the systemd project.
There really is a Linux, and these people are using it, but it is just a part of the system they use. Linux is the kernel: the program in the system that allocates the machine's resources to the other programs that you run. Chiefly systemd-udevd, systemd-journald, systemd-resolved, systemd-networkd, systemd-timesyncd, systemd-logind, systemd-homed, systemd-boot, and systemd-oomd, which decides which of your remaining programs deserve to live The kernel is an essential part of an operating system, but useless by itself; it can only function in the context of a complete set of systemd unit files.
I'd just like to interject for a moment. What you're referring to as systemd, is in fact a component of Gnome, or as I've recently taken to calling it, GNOME.
systemd is not a project unto itself, but rather one more component of a fully functioning GNOME system, made useful by the systemd unit files, target definitions and vital service managers comprising a full OS as defined by Lennart Poettering and the Red Hat vendor lock-in team.
GCC, GDB, CoreUtils, etc - A COLLECTION of useful programs alongside a kernel such as Linux.
Barely relevant?
We owe our thanks to the GNU project. If Linus did not create Linux, something else would have taken it's place. "Maybe" HURD... or maybe something else.
If you read the historical archives, Gnu was kinda late in the game with Linux. They were distributing the Joliets 386-BSD and NetBSD and had some drama with trying to do Linux. Debian is a historical spinoff of this.
I gave a conference talk about this about 4 years ago. There may be slight errors in my recollection
Ok, I'll bite. Which parts of the kernel should not be there, given that Linux is not a microkernel? I maintain that most of the code is there for hardware support, for security, or to support user-space application models.
I'll grant you the in-kernel nfs server and ipsec implementations, but those are not tens of millions of lines of code. The only way I can reach that number is by lobbing off a huge part of device drivers into userspace, but that's not really an interesting discussion as it seems purely subjective.
Hot take: filesystem drivers. Final stage boot should load two images: a kernel which surfaces block devices via HAL, and a one or more filesystem drivers which function as external backends for the in-kernel vfs system.
This would facilitate more innovation and ability to quickly resolve reliability issues with filesystems, make the whole “can you boot off of it or do you need a kernel module” distinction obsolete, and would massively reduce the temptation for the kernel to support certain APIs only for certain filesystems (let me inotify on procfs, and use nonblocking IO on any FS I like, dammit!)
> You're welcome to implement inotify for procfs right now.
I more meant that if filesystem drivers had always been external to the kernel, then it would have been harder for so many syscalls to develop filesystem-specific behavior (e.g. things that work on block FSes but not nfs/tmpfs/9pfs/procfs and so on--inotify, atimes, inode stability, sparsity, stuff like that). Being monolithic made it easier to get lots of FS-specific exceptions, which results in some things requiring you to carefully parse manpages or debug EINVALs based on where files live. The benefit of that approach is that it enabled fast development of new features, but looking back I'm not sure if it was worth it.
> Nonblocking I/O only makes sense for IPC.
I want O_NONBLOCK to not fail for disk files. It could lie and block, or only report ready once things were in the buffer cache, or only report ready once part of the VFS response came out of the hardware interrupt/command queue, or any one of several options. I don't know which if any of those count as "async I/O", but the core problem here is ubiquitous syscalls (read/open/inotify_add_watch) whose failure-or-not is contingent on the type of object you give them (blockdev file vs. non-blockdev file vs. semaphore vs. socket vs. pipe vs. timerfd and so on).
> You want async I/O, which Linux doesn't support but Windows does.
Linux kind of has async block device I/O with io_uring now, but they repeated what is basically the Linux original sin mentioned above and made it contingent on whether a device/driver/filesystem support asynchronous operation or not.
Windows overlapped I/O definitely did it first and better though.
I know this is in jest, but because of the "hurr durr" connotations, only the absolute most bearded of free software zealot neckbeards are going to ever run GNU Hurd beyond simplytest driving it.
Open source continues its unbroken streak of showing us why naming things is important.
Hardly any Americans would recognize “git” as an offensive word. That would not have been a barrier to adoption in the largest single software market in the world.
I grew up in one of the other ex-British colonies and we didn’t use “git” either. It’s a pretty localized word. Apparently it only came to prominence after the empire was already collapsing, so didn’t spread much.
Monty Python's Flying Circus spread it far and wide in the English-speaking world.
Was there any insult that Monty Python didn't use? Some of their gags literally involved opening up the thesaurus and reciting an entry. Most of their sketches involved hurling insults at one another. It was a Petri Dish of British slang!
I have no idea why you think this. You might be terminally online. Hurd is just like "herd" because it's a microkernel, and hurr durr is some social media/gamer shit a tiny amount of people started saying decades after Hurd started being developed.
I'll be honest by saying I for one have never heard hurr durr before. It might be a bit more localized than you think.
Frankly though it's unnecessary to be so personal regarding the people involved, or even the name.
Yes, at this point Hurd is no more than a curiosity. That has nothing to do with the name and everything to do with that we're not really looking for another OS. There are already a bunch of OS's to choose from.
For production machines you can already choose between Windows, Mac, Linux (plus ChromeBook et al), and even BSD. Within each of those there are several mainstream flavors, and other minor ones. No one is realistically punting for Hurd in any kind of "real" environment.
That said very minor OS's do provide interest to academics and hobbiests. New ideas and concepts can be tried before they get folded into main-stream OS's.
I wish that there was one day where I could log in and just feel like I was on the HN that existed when I first got my account. A day with a moratorium on all posting about LLMs, LLM startups, generative AI, just actual Hacker News, i.e. news which is interesting to hackers, and not a compendium of every terrible antisocial Bay Area echo chamber idea being sold as humanity's savior by narcissists with God complexes.
Does anyone more familiar with Hurd know how they addressed performance issues encountered by other Mach-based kernels? I remember reading that one of Mach's bottlenecks was validating that each IPC message had proper permissions/access when sending them between processes due to the large number and frequency of messages.
Mach's problem wasn't that it had slow IPC, but rather that in terms of microkernels, Mach was pretty damn big.
There are some very fast microkernels out there, like the L4 family, which negate the IPC overhead of microkernels by being small enough to fit entirely in the L2 cache of most processors. Linux may only have the single IPC call per round trip, but it's a fucking huge kernel and there is typically a ton of cache thrashing going on.
Are you saying there is cache thrashing because callers often sercice rheid own requests themselves? If you don't want to service requests in the same thread, doesnt it mean you have to spend entire core(s) for running the kernel?
What even is a 20% delta in performance? Like a five year old difference in hardware design.
If I could get a micro kernel with a better security profile than Linux, that doesn’t sound so bad. I could easily be using a ten year old chip for my day to day browsing and probably wouldn’t notice.
If anything, Mach style IPC has gotten worse because of the greater overhead if kernel/user transitions. Spectre really did a number on context switching overhead.
Not true in a data center - 20% less work for same CPU is like 25% more cores needed for any applications where the scaling is driven by CPU utilization. That is $ - and people have made things so far from efficient many medium size companies have 100k cores to pay for.
I'm not a Hurd contributor ( or any OS kernel / library ).
I'm not sure this holds today. One example I can see is related to crypto. We used to have specific hardware for computing cryptography functions but it's now handled directly in standard hardware and the software has not evolved ( but it's been faster and faster to compute checksum functions )
Yes, you want to do more IPC, its basically linear with the increase in performance. So the amount of overhead stays the same basically. At least in IPC benchmarks and those tend to be important.
So for an I/O heavy program like a web browser, what percentage of CPU time is spent in syscalls? Someone must have that data for Firefox. 1%? 10%? And then that moves to 1.1% (or 11%)? And then every other program is even less affected? Curious if anyone is in the know.
Firefox isn’t io bound it is spending a ton of CPU parsing network buffers and rendering UI.
Nginx is IO bound in general - for my headless go stateful server, i could get syscall6 up to 20% of the profile time. It was a fight between GC in the gRPC code allocating for headers and the syscall6 for top single profile footprint.
Most RTOS for high integrity computing deployments are microkernel based, and the large majority of cloud deployments are running on top of microkernels, which is basically what a type 1 hypervisor ends up being, even more so, if it they are serverless workloads running directly on top of the hypervisor.
This is what I like about Haiku OS, they have focused a lot functionality and have that fairly solid, now optimization and stability is being worked on more closely.
> 9P is perhaps not as robust or as fault tolerant as NFS
How is it not robust? What metric did they use? Fault tolerance is not explained here so I don't know what they mean by this. You can use aan(8) on 9front which runs on each side of the connection and holds it open if the network breaks, then resume on reconnect.
I hold out hope for (human) work to coalesce on creating a stable/secure microkernel (SeL4 anybody???) and getting it mature enough to build an OS on.
Once that is done, then you build out the absolutely massive collection of userspace drivers that would be needed to replace linux by leveraging LLMs and other automated techniques. The good part is that by keeping the drivers in userspace, the security risk of generative techniques for driver development is pretty low.
While I upvoted this, as being true for the general population, for me Linux has made it on the desktop (and laptop) much more than 2 decades ago, around 2003, when the remaining problems that had forced me to use dual-booting with Windows had been solved (i.e. when it became possible in Linux to play movies, read DVDs and use TV tuners). When that happened, I wiped out Windows from my computers.
If you have a pro to help you, already for a long time Linux has been fine on the desktop for anyone. My parents, over 80 years old, have used Linux on their PCs for many years, and they did not even know what "Linux" is. That was possible because I installed and configured for them the operating system and all the applications they needed for reading and writing documents, Internet browsing, e-mail, music listening and movie watching etc.
Windows is easy only because it comes preinstalled. I have installed Windows at work on many kinds of computers and during installation I have encountered much more problems impossible to solve for a non-expert than when installing Linux.
I've managed to run Linux just about exclusively since probably just before I was a teen in the early 2000's. I think I bought Red Hat 8 or 9 or something like that at CompUSA on a whim and proceeded to call customer support to get help doing basic tasks on the desktop which they very kindly helped me with. Man I was feeling like a seriously 1337 h4x0r.
I didn’t have a pro to help me. But I used Linux a lot on the cloud/cluster. I didn’t want to futz. So I bought a computer with Linux preinstalled. Then I got one for work. Then a desktop from a different vendor.
There is an extra cost associated with it (I’m paying more for the same hardware), but for me totally worth it to have a functioning Linux install on delivery.
Pretty much. I have been using Linux since 1998 but it became my only system back in about 2009 I think. It just isn't that geeky blued together unix-y thing any more. So long as your hardware is supported, thats it.
Yes, the games support was rough but that is almost a non issue nowadays. Thanks to Proton and Steamdeck support, I don't even check compatibility any more on games.
However the recent success of gaming-related distros such as SteamOS and Bazzite seem to be succeeding where Usability-focused distros such as Mint, Pop!, and Zorin couldn't get deep traction
This should be a lesson about providing a product versus providing a technology.
Ubuntu is "like Windows but not" which is mostly relevant to people boycotting Windows. Steam Deck is "a gamebox" which people who already like games might want to buy. Bazzite is "a thing that turns your computer into a gamebox." Of course, people don't actually want gameboxes either - they want to play games - but it's a few steps closer.
It just...has made it. Plenty of power users just use it as their primary OS. I think official figures under-estimate it because many people daily-driving will spoof user agents of other OSes.
Ah, Android, with a userspace based on Java, Kotlin, C and C++ standard libraries, Android native APIs, without anything related to GNU/Linux, unless it is a rooted device, or one gets to package a full Linux userspace as an APK, which isn't fully compatible due to security enforcement mechanisms.
I would look into those statistics regarding desktop world wide adoption, distribution per OS.
It gives a Linux experience within the security sandbox from Android, which is why there are limitations when using it on regular devices, via PlayStore.
Also it requires working around the NDK limitations on available POSIX APIs.
As a sign in the window of a tiny shop in my quiet suburb says "Laptop and phone repairs - Windows, Linux and MacOS". If that doesn't count, I don't know what else could.
Eric S. Raymond opines that now that Microsoft makes its profits mainly from Azure, some product owner in Microsoft will be allowed to enshittify Windows for collection short-term profit, thus paving way for Linux on the desktop.
First there will be the Desktop Linux of the year
breakthrough. Evil folks say it will coincide with
the end of the world via a total global catastrophy
but these people don't understand beauty - such as
TempleOS, the most perfect OS of all times.
I can't help but feel disappointed that after 35 years the first thing they're talking about is an SVG logo. I mean, there's a certain novelty that it was made on a laptop running Hurd itself, but still I was left with a sense that progress is meandering and whimsical.
I remember being excited by Hurd's ambitions after studying Tanenbaum's OS book in university and drinking the microkernel kool-aid; now I'm too old and jaded to believe it'll be a viable alternative.
There was a brief period about 10 years back where there was a sudden burst of activity on Hurd, but that started to dwindle shortly afterwards. Alas, it is like Fusion power, always moving forwards but under funded to get it across the line into something more.
For every year development Hurd moves forward, so does everyone else's, only the they have much more resources.
My opinion is Sol or Fable could absolutely "finish" Hurd (whatever the definition of "finished" is), but it would be difficult to review and understand everything it did, nobody would trust it, people would raise copyright concerns, doing so would violate all sorts of GNU cultural norms, and it would sit in a repo unused.
Next year will be the year of GNU Hurd, as well as
The Desktop Linux Year and people really using
HaikuOS on real hardware for years. You read it here
first!
Ha I just booted Haiku on my desktop yesterday and it ran surprisingly well. Kind of took me back to 2001. If Firefox is stable on it I actually probably could switch from Windows (my needs are pretty simple at home).
Cool to finally see some small silly tech stuff here that isn't trying to make money.
100% of the user space programs were from gnu. Compiler was from gnu
The initial filesystem was minix
The kernel was Linux.
So I would say Linus had 100s of man years of ecosystem he could benefit from and that is the oss spirit.
But yeah he did not write a shell or an editor or Compiler etc and the kernel was of course oriented towards an unix like os.imagine you would have had to start from scratch, if you had just had a PC and a bios.
This is all true. It’s also all in a thread about GNU Hurd, which also builds on all of that GNU software plus a Mach-based kernel.
Were. How many are now? 1%? GNU's Not Vim, GNU's Not IntelliJ, GNU's Not Google Chrome.
> imagine you would have had to start from scratch, if you had just had a PC and a bios.
Like templeOS I guess
TempleOS blows my mind, honestly.
This entire site is a propaganda piece for one of the largest incubators in tech. Just because it has “hacker” in the name doesn’t mean it represents tech.
Like, the point is tech making money. You’re in the wrong place if you’re looking for something other than that.
Where is the real hacker community then?
Real
You are calling a Linux (sorry, GNU/Linux) killer small silly stuff?
I'd just like to interject for a moment. What you're referring to as GNU/Linux, is in fact, just Linux.
I'd just like to interject for a moment. What you're referring to as Linux, is in fact systemd/Linux, or as I've recently taken to calling it, systemd plus Linux.
Linux is not an operating system unto itself, but rather one more free component of a fully functioning systemd system, made useful by the systemd unit files, target definitions and vital service managers comprising a full OS as defined by Lennart Poettering.
Many computer users run a modified version of the systemd system every day without realizing it. Through a peculiar turn of events, the version of systemd which is widely used today is often called "Linux," and many of its users are not aware that it is basically the systemd system, developed by the systemd project.
There really is a Linux, and these people are using it, but it is just a part of the system they use. Linux is the kernel: the program in the system that allocates the machine's resources to the other programs that you run. Chiefly systemd-udevd, systemd-journald, systemd-resolved, systemd-networkd, systemd-timesyncd, systemd-logind, systemd-homed, systemd-boot, and systemd-oomd, which decides which of your remaining programs deserve to live The kernel is an essential part of an operating system, but useless by itself; it can only function in the context of a complete set of systemd unit files.
I'd just like to interject for a moment. What you're referring to as systemd, is in fact a component of Gnome, or as I've recently taken to calling it, GNOME.
systemd is not a project unto itself, but rather one more component of a fully functioning GNOME system, made useful by the systemd unit files, target definitions and vital service managers comprising a full OS as defined by Lennart Poettering and the Red Hat vendor lock-in team.
Whats in a name?
Personally I use Google/Windows at work, and Steam/Windows at home. I also have a Meta/Android OS for my phone and a Garmin/Auto OS for my car.
My websites are hosted on AWS/Linux, and I browse on Comcast/Internet. My entertainment comes from Warner/Netflix. Or Disney/Disney)
I call it Linux, when I call it at all.
Yep. Stallman was just a bit salty that GNU turned out to be barely relevant. Most of the code in a so-called "GNU/Linux" system is not GNU.
There really was a GNU project, but it was just an attempt to harmonize system tools across all open and proprietary Unices.
> GNU turned out to be barely relevant
GCC, GDB, CoreUtils, etc - A COLLECTION of useful programs alongside a kernel such as Linux.
Barely relevant?
We owe our thanks to the GNU project. If Linus did not create Linux, something else would have taken it's place. "Maybe" HURD... or maybe something else.
Are you not using vendor-lockin-licensed LLVM instead?
NetBSD or one of that ilk.
I’ve recently taken to calling it GNU minus Linux.
Yeah GNU hasn't really been involved for a long time. It's Mozilla/Linux in reality. Not to mention the totally GNU-free distributions, like Alpine.
If you read the historical archives, Gnu was kinda late in the game with Linux. They were distributing the Joliets 386-BSD and NetBSD and had some drama with trying to do Linux. Debian is a historical spinoff of this.
I gave a conference talk about this about 4 years ago. There may be slight errors in my recollection
Where would someone read more about that history? Sounds up my alley:)
The efforts people will go to in order to avoid GPL3. Makes you wonder why?
Because large corporations want to take without giving.
Yep, look at Apples 12 year journey of rebuilding everything that was moved up to GPL3.
Are you saying that Hurd will also kill Google/Linux?
What I'm saying is that Hurd is just a kernel in the same way that Linux is also just a kernel
The snark artist in me would say that Linux is a kernel plus tens of millions of lines of code that absolutely should not be in the kernel but are.
Ok, I'll bite. Which parts of the kernel should not be there, given that Linux is not a microkernel? I maintain that most of the code is there for hardware support, for security, or to support user-space application models.
I'll grant you the in-kernel nfs server and ipsec implementations, but those are not tens of millions of lines of code. The only way I can reach that number is by lobbing off a huge part of device drivers into userspace, but that's not really an interesting discussion as it seems purely subjective.
Hot take: filesystem drivers. Final stage boot should load two images: a kernel which surfaces block devices via HAL, and a one or more filesystem drivers which function as external backends for the in-kernel vfs system.
This would facilitate more innovation and ability to quickly resolve reliability issues with filesystems, make the whole “can you boot off of it or do you need a kernel module” distinction obsolete, and would massively reduce the temptation for the kernel to support certain APIs only for certain filesystems (let me inotify on procfs, and use nonblocking IO on any FS I like, dammit!)
You're welcome to implement inotify for procfs right now.
Nonblocking I/O only makes sense for IPC. You want async I/O, which Linux doesn't support but Windows does.
> You're welcome to implement inotify for procfs right now.
I more meant that if filesystem drivers had always been external to the kernel, then it would have been harder for so many syscalls to develop filesystem-specific behavior (e.g. things that work on block FSes but not nfs/tmpfs/9pfs/procfs and so on--inotify, atimes, inode stability, sparsity, stuff like that). Being monolithic made it easier to get lots of FS-specific exceptions, which results in some things requiring you to carefully parse manpages or debug EINVALs based on where files live. The benefit of that approach is that it enabled fast development of new features, but looking back I'm not sure if it was worth it.
> Nonblocking I/O only makes sense for IPC.
I want O_NONBLOCK to not fail for disk files. It could lie and block, or only report ready once things were in the buffer cache, or only report ready once part of the VFS response came out of the hardware interrupt/command queue, or any one of several options. I don't know which if any of those count as "async I/O", but the core problem here is ubiquitous syscalls (read/open/inotify_add_watch) whose failure-or-not is contingent on the type of object you give them (blockdev file vs. non-blockdev file vs. semaphore vs. socket vs. pipe vs. timerfd and so on).
> You want async I/O, which Linux doesn't support but Windows does.
Linux kind of has async block device I/O with io_uring now, but they repeated what is basically the Linux original sin mentioned above and made it contingent on whether a device/driver/filesystem support asynchronous operation or not.
Windows overlapped I/O definitely did it first and better though.
Have you tried asking an LLM to fix that?
Hurd specifically includes a userspace. Its kernel is Mach.
Hurd is not just a kernel, but yes, it does provide approximately the same functionality of Linux.
They're both kernels, one of them just happens to actually be useful
No, what they're referring to as a GNU/Linux is in fact GNU Hurd.
Of course it’s a large and complex project in terms of code, but in terms of industrial impact it’s unlikely to ever amount to much.
Linux just has too much momentum and there are too few real benefits to Hurd.
This is not GNU/Linux. GNU Hurd is a totally different kernel.
The comment was calling Hurd a Linux killer. One hopes that was intended as a joke.
Hurd is a 'Linux killer' in the same way TempleOS is.
I know this is in jest, but because of the "hurr durr" connotations, only the absolute most bearded of free software zealot neckbeards are going to ever run GNU Hurd beyond simplytest driving it.
Open source continues its unbroken streak of showing us why naming things is important.
Right, which is why something with an offensive name like git could never take off.
Anyways, HURD's biggest problems are performance and perennially missing important features (2026 and no USB support)
Hardly any Americans would recognize “git” as an offensive word. That would not have been a barrier to adoption in the largest single software market in the world.
I grew up in one of the other ex-British colonies and we didn’t use “git” either. It’s a pretty localized word. Apparently it only came to prominence after the empire was already collapsing, so didn’t spread much.
Monty Python's Flying Circus spread it far and wide in the English-speaking world.
Was there any insult that Monty Python didn't use? Some of their gags literally involved opening up the thesaurus and reciting an entry. Most of their sketches involved hurling insults at one another. It was a Petri Dish of British slang!
Those of us who have been using gimp for 20+ years have learned to just accept it. Just take the name they give without complaint and like it.
Maybe the name was more fitting than I realized.
> but because of the "hurr durr" connotations
I have no idea why you think this. You might be terminally online. Hurd is just like "herd" because it's a microkernel, and hurr durr is some social media/gamer shit a tiny amount of people started saying decades after Hurd started being developed.
[dead]
[flagged]
I'll be honest by saying I for one have never heard hurr durr before. It might be a bit more localized than you think.
Frankly though it's unnecessary to be so personal regarding the people involved, or even the name.
Yes, at this point Hurd is no more than a curiosity. That has nothing to do with the name and everything to do with that we're not really looking for another OS. There are already a bunch of OS's to choose from.
For production machines you can already choose between Windows, Mac, Linux (plus ChromeBook et al), and even BSD. Within each of those there are several mainstream flavors, and other minor ones. No one is realistically punting for Hurd in any kind of "real" environment.
That said very minor OS's do provide interest to academics and hobbiests. New ideas and concepts can be tried before they get folded into main-stream OS's.
That seems like an awesome reason to use it in an era of most software is an AI slopfest.
Go for it. I'll probably stick to Linux, though.
it's GPL, not plus-N-word license.
GNU Hurd Durd?
I wish that there was one day where I could log in and just feel like I was on the HN that existed when I first got my account. A day with a moratorium on all posting about LLMs, LLM startups, generative AI, just actual Hacker News, i.e. news which is interesting to hackers, and not a compendium of every terrible antisocial Bay Area echo chamber idea being sold as humanity's savior by narcissists with God complexes.
You might not be able to do that, but you can visit the HN front page as it existed on pretty much any arbitrary date in the past.
https://news.ycombinator.com/front?day=2015-08-05
Nothing changes in 11 years.
* Apple sucks
* The internet is too centralised
* The financial system is unstable
* Browsers got a new API
* Some communist country regulated tech
[flagged]
It seems Samuel Thibault is still there, sweet.
ps: some slides from earlier this year https://fosdem.org/2026/events/attachments/7FZXHF-updates_on...
full vid here https://fosdem.org/2026/schedule/event/7FZXHF-updates_on_gnu...
Does anyone more familiar with Hurd know how they addressed performance issues encountered by other Mach-based kernels? I remember reading that one of Mach's bottlenecks was validating that each IPC message had proper permissions/access when sending them between processes due to the large number and frequency of messages.
Mach's problem wasn't that it had slow IPC, but rather that in terms of microkernels, Mach was pretty damn big.
There are some very fast microkernels out there, like the L4 family, which negate the IPC overhead of microkernels by being small enough to fit entirely in the L2 cache of most processors. Linux may only have the single IPC call per round trip, but it's a fucking huge kernel and there is typically a ton of cache thrashing going on.
>small enough to fit entirely in the L2 cache
Shouldn't the "hot" path fit in the L1 of a "modern" processor (100 kB+)?
Are you saying there is cache thrashing because callers often sercice rheid own requests themselves? If you don't want to service requests in the same thread, doesnt it mean you have to spend entire core(s) for running the kernel?
Is that a bottleneck anymore? Maybe something that was appreciable in 1994 isn't a big performance hit in 2026?
Isn't it even worse now? The faster computers get, the more that software-induced bottlenecks hurt.
I could see just powering through because modern hardware is fast enough to do so being an option.
Adding 20% to a 10 second operation is a lot longer on a wall clock than adding 20% to a 1 second operation.
What even is a 20% delta in performance? Like a five year old difference in hardware design.
If I could get a micro kernel with a better security profile than Linux, that doesn’t sound so bad. I could easily be using a ten year old chip for my day to day browsing and probably wouldn’t notice.
> If I could get a micro kernel with a better security profile than Linux, that doesn’t sound so bad.
I suggest that you look at genode/sculpt:)
If anything, Mach style IPC has gotten worse because of the greater overhead if kernel/user transitions. Spectre really did a number on context switching overhead.
Hurd is like 36 years old at this point. More like adding 20% to a 10 second operation, compared to adding 20% to a 10ms operation.
People lost their shit when Spectre mitigations added 20% overhead to kernel calls though. Why wouldn't they do the same for microkernel overhead?
The same people are perfectly fine with Electron crap all over the place, they don't really care about overhead.
Not true in a data center - 20% less work for same CPU is like 25% more cores needed for any applications where the scaling is driven by CPU utilization. That is $ - and people have made things so far from efficient many medium size companies have 100k cores to pay for.
I'm having a hard time believing that data centers are chomping at the bit to use the Hurd.
If people really cared about performance, we'd be doing something other than the bloated way we have been doing things.
https://winworldpc.com/product/qnx/144mb-demo
I'm not a Hurd contributor ( or any OS kernel / library ).
I'm not sure this holds today. One example I can see is related to crypto. We used to have specific hardware for computing cryptography functions but it's now handled directly in standard hardware and the software has not evolved ( but it's been faster and faster to compute checksum functions )
I mean, every major arch has crypto specific acceleration instructions.
Yes, it's worse because processors clocks are much faster than memory clocks.
Yes, you want to do more IPC, its basically linear with the increase in performance. So the amount of overhead stays the same basically. At least in IPC benchmarks and those tend to be important.
Making that fast is fundamental to any OS.
So for an I/O heavy program like a web browser, what percentage of CPU time is spent in syscalls? Someone must have that data for Firefox. 1%? 10%? And then that moves to 1.1% (or 11%)? And then every other program is even less affected? Curious if anyone is in the know.
Firefox isn’t io bound it is spending a ton of CPU parsing network buffers and rendering UI.
Nginx is IO bound in general - for my headless go stateful server, i could get syscall6 up to 20% of the profile time. It was a fight between GC in the gRPC code allocating for headers and the syscall6 for top single profile footprint.
Looking forward to reading the results of your benchmarking Hurd vs. Linux.
It depends on the OS, some OS have some way to do IO that gets around that. iouring or things like that. I don't know how Hurd does IO.
Most microkernals have methods to do IO in a different way. If they don't then likely they would pretty slow when doing I/O heavy operations.
Is there reason to think they've addressed them?
Most RTOS for high integrity computing deployments are microkernel based, and the large majority of cloud deployments are running on top of microkernels, which is basically what a type 1 hypervisor ends up being, even more so, if it they are serverless workloads running directly on top of the hypervisor.
Lets get to a usable feature set before optimization.
This is what I like about Haiku OS, they have focused a lot functionality and have that fairly solid, now optimization and stability is being worked on more closely.
> 9pfs for the Hurd
From the wiki:
> 9P is perhaps not as robust or as fault tolerant as NFS
How is it not robust? What metric did they use? Fault tolerance is not explained here so I don't know what they mean by this. You can use aan(8) on 9front which runs on each side of the connection and holds it open if the network breaks, then resume on reconnect.
> I plan to build bootable images and boot the kernel on Apple M1
GuixSD with GNU Hurd would be a really fun way to repurpose an old MBP.
My M1 is currently running Guix on Fedora, I would love to see this!
I hope Hurd has its day. Maybe in 100 years or something.
Instead of rewriting Bun, maybe Anthropic should have thrown some tokens at finishing Hurd.
I hold out hope for (human) work to coalesce on creating a stable/secure microkernel (SeL4 anybody???) and getting it mature enough to build an OS on.
Once that is done, then you build out the absolutely massive collection of userspace drivers that would be needed to replace linux by leveraging LLMs and other automated techniques. The good part is that by keeping the drivers in userspace, the security risk of generative techniques for driver development is pretty low.
I thought work was finished to allow Hurd to use Linux device driver code as is?
We'd have to name it Turd then.
That'd be GNU/Turd to you!
You can't polish a Hurd.
Fun fact: Duke Nukem Forever shipped before GNU Hurd.
Maybe they are still in the same level of forever. Oh well.
"I come here to code GNU Hurd and chew bubblegum, and I got a whole lotta bubblegum..."
Come GIT some!
Sorry, only pun I could think of.
Relevant xkcd: https://xkcd.com/1508/
I mean, we're already ahead of schedule for Elon Musk stuff and BloodDrone, so who knows...
Next year will be the year of Hurd on the desktop, I am sure of it.
Hurd will have its day once Linux makes it on the desktop :-)
While I upvoted this, as being true for the general population, for me Linux has made it on the desktop (and laptop) much more than 2 decades ago, around 2003, when the remaining problems that had forced me to use dual-booting with Windows had been solved (i.e. when it became possible in Linux to play movies, read DVDs and use TV tuners). When that happened, I wiped out Windows from my computers.
If you have a pro to help you, already for a long time Linux has been fine on the desktop for anyone. My parents, over 80 years old, have used Linux on their PCs for many years, and they did not even know what "Linux" is. That was possible because I installed and configured for them the operating system and all the applications they needed for reading and writing documents, Internet browsing, e-mail, music listening and movie watching etc.
Windows is easy only because it comes preinstalled. I have installed Windows at work on many kinds of computers and during installation I have encountered much more problems impossible to solve for a non-expert than when installing Linux.
I've managed to run Linux just about exclusively since probably just before I was a teen in the early 2000's. I think I bought Red Hat 8 or 9 or something like that at CompUSA on a whim and proceeded to call customer support to get help doing basic tasks on the desktop which they very kindly helped me with. Man I was feeling like a seriously 1337 h4x0r.
I didn’t have a pro to help me. But I used Linux a lot on the cloud/cluster. I didn’t want to futz. So I bought a computer with Linux preinstalled. Then I got one for work. Then a desktop from a different vendor.
There is an extra cost associated with it (I’m paying more for the same hardware), but for me totally worth it to have a functioning Linux install on delivery.
Pretty much. I have been using Linux since 1998 but it became my only system back in about 2009 I think. It just isn't that geeky blued together unix-y thing any more. So long as your hardware is supported, thats it.
Yes, the games support was rough but that is almost a non issue nowadays. Thanks to Proton and Steamdeck support, I don't even check compatibility any more on games.
Linux has made it to the desktop, packaged inside VMs served by Apple, Microsoft and Google.
That's not really what everyone means.
However the recent success of gaming-related distros such as SteamOS and Bazzite seem to be succeeding where Usability-focused distros such as Mint, Pop!, and Zorin couldn't get deep traction
This should be a lesson about providing a product versus providing a technology.
Ubuntu is "like Windows but not" which is mostly relevant to people boycotting Windows. Steam Deck is "a gamebox" which people who already like games might want to buy. Bazzite is "a thing that turns your computer into a gamebox." Of course, people don't actually want gameboxes either - they want to play games - but it's a few steps closer.
[flagged]
Because it is way cheaper?
It's around $1 / M on claude code plan. Most users report around 200M tokens on the $200 plan.
I don't think anyone is paying full API pricing.
No, but that is what they will ever get from OEMs selling PCs at computer stores used by normies.
Also lets not forget those distros play Windows games, because hardly any studio bothers with native Linux games.
They play both, which is much better for adoption anyway? And it's contributed substantial improvements back to wine and other projects.
It is hardly any different that saying gaming on Linux is great because MAME exists.
Game studios will keep buying Windows, using Visual Studio, and let Valve do the needful.
Additionally people keep acting as if Valve and Gabe will be around forever, they won't, none of us will.
MAME and Proton are not the same at all. You can reuse Wine/Proton libraries even on Windows.
It just...has made it. Plenty of power users just use it as their primary OS. I think official figures under-estimate it because many people daily-driving will spoof user agents of other OSes.
And there's Android.
Ah, Android, with a userspace based on Java, Kotlin, C and C++ standard libraries, Android native APIs, without anything related to GNU/Linux, unless it is a rooted device, or one gets to package a full Linux userspace as an APK, which isn't fully compatible due to security enforcement mechanisms.
I would look into those statistics regarding desktop world wide adoption, distribution per OS.
Anyone who wants it can have Termux which gives you a full Linux experience under Android with GNU tools if you want it, though.
No it does not.
It gives a Linux experience within the security sandbox from Android, which is why there are limitations when using it on regular devices, via PlayStore.
Also it requires working around the NDK limitations on available POSIX APIs.
Not with Fdroid's release.
Oh come on. :-)
Linux is not userspace. Linux is a kernel, and android is a parallel fork of the kernel. So android is a version of linux but is not GNU/Linux.
Android/Linux, if you will. We dont have to be shy this.
Yeah, exactly because of that, lets not do victory laps for a kernel.
GNU/Linux, the Linux desktop, has zero relevance on Android.
No one, not a single person, buying Android tablets and phones at places like MediaMarkt, cares about GNU/Linux Desktop.
They care about what apps via PlayStore, Samsung, Huawei and Xiomi stores, they can use, just like all normies.
People using Linux in 1992 would say something similar to people using any of the current flavours of Linux.
No they wouldn't, given that the only reason I got into Linux was the lackluster POSIX support on Windows NT.
Linux hit 10% desktop usage recently
Not really, that was a fluke, still below 10%, although getting closer with 7.53%.
https://gs.statcounter.com/os-market-share/desktop/worldwide
Both numbers are big, no? Especially given the complete dominance of linux in many other settings.
I think it is pretty impressive if you consider ChromeOS, with all of Google's resources behind it, is at 1.47%.
Yeah, if we ignore Windows 70%, and Mac's 20% of desktop share.
As a sign in the window of a tiny shop in my quiet suburb says "Laptop and phone repairs - Windows, Linux and MacOS". If that doesn't count, I don't know what else could.
Eric S. Raymond opines that now that Microsoft makes its profits mainly from Azure, some product owner in Microsoft will be allowed to enshittify Windows for collection short-term profit, thus paving way for Linux on the desktop.
https://x.com/esrtweet/status/2084605702203424896
Hasn't this been happening for a while now?
Ah so X/Twitter is where ESR is nowadays. Kind of not surprised.
Nah, Claude will finish it over night
That's what they tell us, but for some reason software still looks just as bad or worse than it ever has.
First there will be the Desktop Linux of the year breakthrough. Evil folks say it will coincide with the end of the world via a total global catastrophy but these people don't understand beauty - such as TempleOS, the most perfect OS of all times.
I can't help but feel disappointed that after 35 years the first thing they're talking about is an SVG logo. I mean, there's a certain novelty that it was made on a laptop running Hurd itself, but still I was left with a sense that progress is meandering and whimsical.
I remember being excited by Hurd's ambitions after studying Tanenbaum's OS book in university and drinking the microkernel kool-aid; now I'm too old and jaded to believe it'll be a viable alternative.
There was a brief period about 10 years back where there was a sudden burst of activity on Hurd, but that started to dwindle shortly afterwards. Alas, it is like Fusion power, always moving forwards but under funded to get it across the line into something more.
For every year development Hurd moves forward, so does everyone else's, only the they have much more resources.
It’s like farmer’s market craft fair tech. Same vibes.
Last neckbeards standing?
Someone should do the blasphemous thing and have Opus 5 finish HURD.
Actually in all seriousness reviving dead interesting FOSS projects is one use for AI coding I've thought about.
Are AIs good enough at doing novel work that they have yet to encounter?
My guess is that the GNU/HURD people are doing a lot of novel work that doesn't yet exist.
How novel is it really?
You think what they're doing isn't covered in the intersection of OS books, OS papers, other OSs, osdev, random GitHub projects?
AI to some extent learns APIs. If they are doing new things they are probably introducing new kinds of interfaces.
My opinion is Sol or Fable could absolutely "finish" Hurd (whatever the definition of "finished" is), but it would be difficult to review and understand everything it did, nobody would trust it, people would raise copyright concerns, doing so would violate all sorts of GNU cultural norms, and it would sit in a repo unused.
Thus my "blasphemous" comment.
The D compiler port to Hurd is very interesting
Tales from the crypt.
I would love GNU Hurd to succeed.
That said, if there's a resurgence in microkernel, it will probably come from the Rust community.
I agree. Operating systems have become too fast.
Next year will be the year of GNU Hurd, as well as The Desktop Linux Year and people really using HaikuOS on real hardware for years. You read it here first!
Ha I just booted Haiku on my desktop yesterday and it ran surprisingly well. Kind of took me back to 2001. If Firefox is stable on it I actually probably could switch from Windows (my needs are pretty simple at home).
Wow.