AtheOS Interview 147
JigSaw writes: "BeNews has a very interesting interview with Kurt Skauen, the AtheOS creator and almost its sole developer. In the interview, Kurt is discussing the design of his OS which features a (nearly) micro-kernel, memory protection, 'true' multitasking, real C++ OOP design from the ground-up and all the rest of these buzzwords. AtheOS uses its own GUI, it does not rely on X or KDE libs, so porting Konqueror to his OS was a bit of a challenge."
i wish (Score:1)
Re:Memory requirements (Score:1)
Re:Responce to thee article (Score:1)
Re:AtheOS (Score:1)
Actually, building things from scratch can be easier. Often enough, the really difficult part is to learn all the idiosyncrasies in the code base others wrote while with your own pet system, you learn it as you write it.
Writing an OS from scratch takes some talent, but mostly a *lot* of patience. Otherwise, it is not that difficult if you have a good grasp of basic system programming concepts (which are easy to pick up from textbooks, I dare say most is common sense) and if you know your hardware (so you can write the necessary drivers to get started). Learning all the ins and outs of a foreign system can be much more cumbersome than desiging your own, and yet it is often required in order to make even rather minuscule changes. Or, to put it in other words, it is easier to deal with your own common sense than with the collective mindset of a bunch of completely different people.
Not that I want to put down the achievements of this programmer - I'm very impressed by his effort. However, I am not so much thrilled that he could do this but that he had the patience and the stamina to pull it through.
Re:oh please (Score:1)
Re:Fuck Moderators (Score:1)
linux is far better than bsd (Score:1)
Linux is the most inferior operating system (Score:1)
Re:linux is far better than bsd (Score:1)
Re:Dear god... (Score:1)
Re:Lack of direction (Score:1)
Re:Question for kurt... (Score:3)
Re:What I don't understand... (Score:4)
Re:vmware? (Score:5)
you can update to 0.3.4 once installed.
http://www.ethernalquest.org/VmAtheOS.zip
I haven't tried this, don't have the bandwidth.
No more complaints (Score:5)
Re:Spelling? Grammar? (Score:1)
all you base are belong to us . JeffK lvies aigan get a new cup for you cumputar
Re:Memory requirements (Score:2)
Accually the i386 could access several terabytes of memory. However when you go over 4Gb you have to get into segmented memory. Memory segments were very difficult to get right in DOS (not sure about other systems, but few used segmented memory) so nobody bothered implimenting them. (Accually this isn't true, microsoft implimented segmented memory in win32, but that seems more to break OS/2 then because there was a gain from it)
4 gigabytes comes from having 32 bits. (2 to the 32nd power is 4 billion something)
Re:Dear god... (Score:1)
Re:Lack of direction (Score:1)
AmigaOS compatibility is much more interesting. And given that there's alreagy an open-source effort at recreating AmigaOS out there, AROS (at www.aros.org), I think there'd be more future in the AROS and AtheOS teams co-operating. AmigaOS has a massive amount of support and software, but not much future: Amiga Corp is going over to a version of Tao's Elate (www.tao.co.uk.) A lot of Amiga fans would be keen to have a free AmigaOS for the PC, I'd think...
Re:AtheOS impression... (Score:1)
Networking is a particularly poor example to cite here because afaict atheos's network stack is in-kernel, and Your Favorite OS's is also on the way there.
Re:vmware? (Score:1)
-- Thrakkerzog
vmware? (Score:2)
Re:Kurt on slashdot (Score:1)
The two operating systems are both fundamentaly different. Read the Atheos webpage (when it's not so congested) and/or the article. Atheos is not intending to be the next GPL BeOS clone. All Atheos has in common with Be is that Atheos gleaned some ideas from the BeOS. Nothing more, just like GNOME which borrowed lots of ideas from anywhere else.
Atheos is not and never will be BeOS, deal with it.
--
Slashdot didn't accept your submission? hackerheaven.org [hackerheaven.org] will!
Re:More curious microkernel stuff (Score:1)
Linux also has this capability through modules, though AtheOS is likely to have stretched this concept a little further.
This has nothing to do with a microkernel.
Question for kurt... (Score:2)
Anyway, it would be interesting to hear about what steps Kurt had to go through to bootstrap to OS to a usable state. Was initial development done under Linux, DOS/Windows, or some other OS? Was there a lot of rebooting going? I've never written my own OS, or even just code that is loaded by the BIOS as an OS would be, even if all the code does is write "Hello World!" to the screen.
Kurt, if you're reading this and you don't mind, a bit of background info would be nice. Thanks in advance.
Re:More curious microkernel stuff (Score:1)
The kernel itself doesn't contain any drivers nor any filesystems. They are all loaded at runtime.
Yeah, but they still run in kernel space rather than user space, so it's not a microkernel, but not quite monolithic either...
john
Re:AtheOS is what I've been griping for... (Score:1)
It is basically a single user unix-clone with
a non standard gui.
Something like Eros is far more innovative.
See www.eros-os.org
Unfortunately, its a lot harder to make something that might actually change the state of the art.
Re:the one question I have... (Score:1)
Security model
Component model ( used from the ground up)
Global filesystem
Anyway.....
Re:responses. (Score:1)
(just like win32):
Every bit of new kernel functionality needs a new system call. (ioctls are skanky system calls)
The filesystem is hierarchical
Security is acl based, or non existant.
By acl I mean the kernel has a list of who can
access each object. Yes, this is how unix works.
Re:AtheOS is what I've been griping for... (Score:1)
It has almost all the same warts.
Ie it is another expression of the same core design. Bunches of system calls, joke user based security, global filesystem for extra security holes.
They are very similar compared to what can be done.
Re:what you want is plan 9! (Score:1)
Its been ages since I used it at Uni (York, where Vita Nuova sprung from). It was on the undergrad dev machines during my first year before it was savagely replaced by Linux and NT dual booters to wide applause (and tears).
Much of what I remember about it was 'fun' - it does lots differently and the whole the gui apps (its not X) had very different behaviour to what we expect but its is surprisingly smooth to learn and become accustomed to.
But, outside of white coat scientist walks we all run drooling toward eye candy and Linux and its plethory of color won the day among the students.
Re:Dear god... (Score:2)
Re:what you want is plan 9! (Score:1)
The restricted government usage clause.
~Tim
--
Re:what you want is plan 9! (Score:2)
~Tim
--
Re:More curious microkernel stuff (Score:1)
Re:Dear god... (Score:1)
How about MySQL? Significant enough for you?
There are dozens more, but you didn't even bother looking did you?
Slashdot Bug (Score:1)
Re:Big F'n' Deal (Score:1)
You know.. it shouldn't bother me, but it does. *sob*
Troll (Score:2)
He says that he's wanting to perhaps switch the core system to the LGPL so that he doesn't discourage people using other licenses. He's doing this to allow 3rd-party developers more freedom in designing their software. He never mentioned taking the whole OS to a closed source license for profit reasons. You're either trolling or grossly misinformed.
You know what? Even if he felt like doing that, then your choices are (1) shut up and (2) like it. It's his personal project that he's poured blood, sweat, and tears into. You're obviously not a user of AtheOS, so it's not like it effects you anyway. You claim that he's "no Linus" and that his "poject is no Linux," but let's see you write your own OS from scratch, especially one as advanced and fully featured as his. His OS is in many ways on par with Linux and BeOS for functionality, and he's been doing it all himself. Let's see you do better.
Futhermore, exactly where does he bash on other OSes or brag about how great his project is? Nowhere. You're just making up crap to troll, and it's not even a very good one at that. This man has done something that 99.9999% of the population simply isn't capable of and you dare to slam him for it because of a simple potential license change?
Whatever. You people are the reason that the Free Software movement feeds on its young. I can't believe you'd slander this man over such a petty thing as going from one Free Software license (GPL) to another (LGPL).
Re:Troll (Score:2)
In particular, some of the features that his system has are particularly thorny issues in OS design. He has a pluggable VFS system. He has a kernel which is preemptable. Do you have any idea how hard that makes it to do pratically anything in the kernel when every single component of it has to be reentrant? He has full SMP support. This adds another level of complexity to design.
Of course, it's easy to bash another man's achievements when you don't understand their magnitude.
Re:Um (Score:3)
Well, that depends. You see, Amiga is more than a piece of hardware. It's more than an OS. It's more than an internet appliance. It's more than a video editing workstation. Amiga is a metaphysical state of computing perfection -- an undefinable nirvana between hazy dreams, shattered promises, and perfection long since lost. Amiga is a way of being. It is enlightenment. It is a freedom gained from oneness with vapor than only many, many years of guru meditations can achieve.
So, no. AtheOS is not available for Amiga, but it strives to be, much like us all.
Big F'n' Deal (Score:4)
It's not like most of the American-born readers (and editors) of this site have a better grasp of the language. That doesn't make them less intelligent, just less articulate. Plus, I'm halfway sure the guy isn't a native English speaker anyway. Let's see how well you do in your second language, if you even have one that is.
I respect his C++ skills better anyway. I have to have a lot of respect for a man who will singlehandedly write his own preemptively multithreaded kernel, taggable journaling filesystem, GUI & event-driven programming layer, and system drivers, and who ported GLIBC and a web browser to his home-brew system. Sure, the grammar threw me for a bit of a loop initially, but just how high did your average native American Slashdot reader score on their verbal SATs/ACTs anyway?
This guy has a crazy mix of genius and dedication that could've changed the world if applied to things like military strategies or politics or if they had appeared on the scene at the right time and place. How different do you think things would've been had he started his OS 5 years ago? He'd be Linus right now and have his own cult following. (Of course, his OS might've gone in a completely different direction without his BeOS-loving friends to influence him.)
Of course, this is yet another pointless response to a blatant troll. Disregard as you see fit.
Troll (Score:5)
He says that he's wanting to perhaps switch the core system to the LGPL so that he doesn't discourage people using other licenses. He's doing this to allow 3rd-party developers more freedom in designing their software. He never mentioned taking the whole OS to a closed source license for profit reasons. You're either trolling or grossly misinformed.
You know what? Even if he felt like doing that, then your choices are (1) shut up and (2) like it. It's his personal project that he's poured blood, sweat, and tears into. You're obviously not a user of AtheOS, so it's not like it effects you anyway. You claim that he's "no Linus" and that his "poject is no Linux," but let's see you write your own OS from scratch, especially one as advanced and fully featured as his. His OS is in many ways on par with Linux and BeOS for functionality, and he's been doing it all himself. Let's see you do better.
Futhermore, exactly where does he bash on other OSes or brag about how great his project is? Nowhere. You're just making up crap to troll, and it's not even a very good one at that. This man has done something that 99.9999% of the population simply isn't capable of and you dare to slam him for it because of a simple potential license change?
Whatever. You people are the reason that the Free Software movement feeds on its young. I can't believe you'd slander this man over such a petty thing.
X Windows (Score:1)
Wow (Score:2)
Re:vmware? (Score:1)
Re:AtheOS (Score:1)
I once did that in a few years ago - in assembler. The debugging was the real pain : compile on computer A, load on a disk, move to computer B, boot, see how it behaves, then go back to computer A and iterate over and over. The most painfull was before I had a display driver written as the only way to get an info from the PC was thru the beeper (ie beeps : situation A happened, no beep : situation A didn't happen... why ?)
On the other hand I wouldn't dare writing for the Linux/*BSD kernel - way too much code written by others to understand before writing just one line.
What I don't understand... (Score:5)
Um (Score:2)
stuff (Score:1)
Also I love this color scheme.
Re:Memory requirements (Score:3)
Not the same as memory space.
Spyky
Re:AtheOS impression... (Score:2)
Re:AtheOS is what I've been griping for... (Score:2)
Re:More curious microkernel stuff (Score:2)
Re:Memory requirements (Score:2)
Re:Memory requirements (Score:2)
Re:what you want is plan 9! (Score:2)
http://plan9.bell-labs.com/plan9dist/screenshot.h
I wish the GUI was "beatiful" instead of the source.
--
the homepage, if it's slashdotted (Score:5)
I have zero tolerance for zero-tolerance policies.
Re:Kurt on slashdot (Score:2)
Quite frankly, I'm a BeOS nut. I haven't had the chance to try AtheOS yet (I've been waiting for a more featureful build), though. I noticed some discussions a while back on the BeNews site (I think) about making AtheOS source- and binary-compatible with BeOS. Seeing that Be Inc.'s future is cloudy, I believe that I can speak for much of the BeOS community when I say that this would be much appreciated and would lead to wider acceptance and usage of AtheOS. I'm wondering, however, about the feasability of this. Since AtheOS is still in its early stages, changing the application structure wouldn't cause as many problems as with a more entrenched operating system, but I'm not sure how difficult this would be to do with the current AtheOS codebase (I'm not much of a programmer). If it is feasable, are there any speculations on your part as to whether something like this might be undertaken, or would it be something you would consider when there are more people working on AtheOS? Thank you for your time.
This is Cool (Score:2)
"One World, one Web, one Program" - Microsoft promotional ad
Re:My take on AtheOS... I want my shell, dammit! (Score:2)
As for all those features "In the kernel", how do you think Linux does it?
Under AtheOS, the appserver (The bit that does all the GUI elements) is in userspace. Drivers are dynamically loaded, as are filsystems. There is no need to recompile the kernel to add in a driver. In fact there is no option to compile a driver into the kernel. Linux has far more "stuffed into the kernel" than AtheOS.
AtheOS isn't going to appeal to the hardcore Linux and BSD users. However, i'm inclined to say, "So what?"
If you like Unix & X, use it. If you want something a little easier to manage as a desktop OS, keep your eye on AtheOS.
Re:responses. (Score:3)
Well, AtheOS:
- Has a POSIX.1 API
- Uses bash as it's shell
- Uses GNU CC & Glibc
It's as close to Unix as it really needs to be, for me to feel comfortable using it. If you're used to Linux, you feel close enough to home using AtheOS for basic tasksRe:AtheOS is what I've been griping for... (Score:2)
Hopefully some of the BeOS developers will take up this project as a hobby.
A Lone OS Developer? (Score:3)
Kurt is probably very thankful for "32-bit OS Design and Development With Pre-emptive Threading in C++ for Dummies".
Re:More curious microkernel stuff (Score:2)
Re:AtheOS is what I've been griping for... (Score:2)
More curious microkernel stuff (Score:3)
I'm not even sure why people still care how microkernelish every new OS is...
Adress space != memory (Score:3)
goatAtheos? (Score:3)
Am I the only one around here who is a little leery anymore of clicking any links in the .cx domain?
eh, i guess that's what i get for browsing at -1...
Re:what you want is plan 9! (Score:2)
---
Re:what you want is plan 9! (Score:2)
---
Re:dont want simulated, I want the real thing! (Score:2)
Re:The AtheOS homepage. (Score:4)
The server link is only 2Mb/s so it tend to be a bit hard to access when slashdotted.
Re:Kurt on slashdot (Score:5)
Yes. I have been developing AtheOS solely under AtheOS for about 3 years now.
>Is anyone working on a GUI RAD tool for AtheOS (something in the spirit of KDevelop or Glade)?
No.
>Are you trying to gather the momentum among application developers now or is it too early?
It's a bit early. I have never ever anounced AtheOS anywhere myself since many important "desktop features" and other things that I whould like to have firmly defined from the beginning are still missing.
>Do you have a Component Model yet? If not are you planning to add one soon? A CM would definitely help keep the look and feel consistent and help avoid code duplication. Two things that marred Linux's acceptance as a desktop OS.
Nothing like COM or CORBA but AtheOS have an object oriented design and it is component based in the sence that you can build abstract components that communicate over the flexible "builtin" event system.
Re:the one question I have... (Score:4)
I think AtheOS is actually a very valuable project for the Open Source world because not everyone is a Unix fan. A modern GUI but non-Unix operating system under GPL is a good thing for the OS scholar, as it gives a different perspective on how to do things.
/brian
Re:what you want is plan 9! (Score:2)
PS: BTW, according to the license you have to agree to before you can download Plan9, you cannot export Plan9 out of the US without a license! So much for putting it on your FTP site....
Re:the one question I have... (Score:2)
Because he has a desire to create and the ability to focus that desire towards his own OS.
If you don't understand then you are, indeed, missing something.
Re:X Windows (Score:2)
Re:what you want is plan 9! (Score:2)
"Do one thing and do it well", a maxim that works excellently, but that most developers seem unaware of.
Re:what you want is plan 9! (Score:2)
actually, that's not true. what the license does say is that if you distribute the system with modifications, you must make the source code available (on request) to lucent (the intention is that they can make such modifications available in the general distribution).
this is no worse than the GPL, which stipulates that if you distribute the system, you must make the source available too.
Re:what you want is plan 9! (Score:2)
not true. you have to comply with the US export regulations (as does the export of every other piece of software), and as it contains some crypto software, you aren't allowed to export it to one of the "7 terrorist nations" or somesuch crap.
this is US government's doing, not lucent's (although it has to be said the lucent lawyers do seem to err on the "safe not sorry" side at times...)
Re:what you want is plan 9! (Score:2)
beauty is in the eye of the beholder!
i agree it doesn't look like conventional GUIs that you see everywhere else, but then again the point was to rethink things without prejudice.
you'll notice from that screenshot that there's very little lost screen real-estate. no arrays of cryptic icons, no titlebars, no buttons, no text entry boxes, etc. that's one design goal: screen space is precious, so use it for information not crap.
also, it can't run any other WM, as it doesn't use the abortion that is X windows. this allows the window manager to start up in a fraction of a second, and take about 1/10th of the memory of window managers under other systems. most applications start up faster than you can notice.
no, it's not an OS for weenies that want to point-and-click on a few predetermined actions all day, a la Microsoft Word. it's small, incredibly powerful, and a joy to develop for and use on a day-to-day basis.
what you want is plan 9! (Score:4)
It's open source, written by some of the best people in the business, (the source code is beautiful) and its whole raison d'etre is to investigate how things should be done, instead of how they are usually done.
you might find some of your fundamental assumptions challenged and your paradigms shifted, but isn't that the point?! it keeps fundamentals of UNIX while discarding all the crud that's built up since the 7th Edition...
Re:what you want is plan 9! (Score:2)
its whole raison d'etre is to investigate how things should be done, instead of how they are usually done
diametrically opposed to this statement:
it keeps fundamentals of UNIX
Re:Big F'n' Deal (Score:2)
C'mon dude... you left out a comma!
Re:AtheOS is what I've been griping for... (Score:2)
Re:QNX (Score:2)
If I had ruled the world back then, I would have made QNX the top dog, and we'd have a great OS paradigm for the next 20 years. No slight to Linux, which has its own strengths; but I'd say that Linux is more important for social reasons, i.e. open source, than for intrinsic architectural reasons. But that's not how things worked out. In the dark days, I kept telling people "It wouldn't be bad for humanity if the QNX guys just gave up and LGPL'd the whole system." The technology was superior enough that it could have eventually provided a viable alternative for a whole class of high-performance systems. This is sort of the path that Linux acceptance has in fact wound up taking, but QNX would have had a bit more technical depth at each end of the spectrum, I believe, ranging from gnarly big databases to tiny embedded systems.
But anyway, that's not how things went, and *my* path would have required that a good OS company admit defeat -- which wouldn't be a nice ending, even if the rest of us profited with a pretty open-source OS.
JMHO -- Trevor
The Accomplishment of One Person (Score:3)
Great work Kurt.
--CTH
--
Atheos vs. BeOS (Score:3)
Over on the BeNews discussion board for the article, a lot of people are saying that AtheOS has (and presumably will continue to have) poor driver support.
They forget two things, however:
- AtheOS is GPL (though Kurt is considering moving it to LGPL).
- Linux has the best hardware support of any non-Redmond OS out there.
Combine these two, and it's almost a certainty that AtheOS will get drivers, simply because it's drivers can be based on Linux kernel drivers.It might be an interesting exercise to port most of the Linux kernel drivers to the AtheOS architecture.
Re:AtheOS is what I've been griping for... (Score:2)
AtheOS is what I've been griping for... (Score:3)
Simply put, it's an open source project that isn't just a clone of an existing product. Open Source works great when everybody has a pretty set model of how it should work. Linux was easy because they just had to make it work like the existing assortment of Unicies.
But AtheOS is good because they are investigating how things should be done, instead of how they are usually done. Granted, BeOS already did a lot of that thinking, but they still are going off in their own directions.
It'll be interesting to see how this works out. Is it possible for an open source project to create something new and truly innovative?
Re:the look (Score:2)
Re:QNX (Score:2)
QNX does have a gui subsystem (Photon) but I think its main purpose is to provide QNX developers with a native QNX environment to avoid cross compiling. I don't think we'll see Photon as a end user front end. Having said that QNX is hiring multimedia plugin developers so I might be wrong here. QSSL seem to be quite enigmatic about the direction of QNX-Rtp. Maybe someone here will enlighten us on this matter.
While AtheOS and QNX have very little in common both are worth checking out as both are gems. QNX is beautiful underneath. It just laughs in the faces of Linux zealots who think microkernels are passe.
Atheos is promising and because it's developed as an open source project there is a chance that the community might take it in a certain direction. But then again it might drift without focus. Time will tell I guess. It's looking good for now though.
Hope this helps somewhat.
Re:Big F'n' Deal (Score:2)
I don't have to Imagine. I just come to slashdot.
Kurt on slashdot (Score:4)
My questions to Kurt:
Can Atheos compile itself yet?
Is anyone working on a GUI RAD tool for AtheOS (something in the spirit of KDevelop or Glade)?
Are you trying to gather the momentum among application developers now or is it too early?
Do you have a Component Model yet? If not are you planning to add one soon? A CM would definitely help keep the look and feel consistent and help avoid code duplication. Two things that marred Linux's acceptance as a desktop OS.
A note from the Free Software Foundation: (Score:3)
Give him a good slashdotting (Score:2)
Re:What I don't understand... (Score:2)
Because X11 works and because it works well. X11 even allows applications from different desktops and toolkits to work together reasonably well.
Lack of responsiveness of some GUIs on Linux has nothing to do with X11, it's a consequence of toolkit design. Talking to developers of several toolkits for X11, I found that many of them have disdain for the X11 APIs and just don't expect it to do well. When their toolkits perform poorly, they don't bother to look to their code, they just assume it's X11's fault. That way, performance problems in toolkits never get fixed.
The fact that "that other OS" can be sluggish also suggests that it isn't X11's design. X11's design may not be all that pretty after nearly two decades of evolution, but it gives a lot of functionality that other window systems lack, and the mechanisms X11 uses for local communications (unix domain sockets, shared memory) are reasonably efficient.
the look (Score:2)
Max
Compared to Linux (Score:3)