The OpenBeOS kernel has finally forked from NewOS. Michael Phipps has checked in the source code for the OpenBeOS kernel to the CVS repository. Initially, this code differs little from NewOS. Over time, as the code is tweaked and amended to fit the requirements for creating an R5 clone, the kernel will diverge from its NewOS roots.
OpenBeOS-NewOS Fork Complete
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
does this mean that OBOS finaly has some code put together that can work on its own? I mean not running ontop of the current r5?
No, not by a long shot. But it means that we are one step closer! Keep up the good work teams, the progress so far is looking very promising.
No, not yet. The NewOS kernel will of course boot by itself, but that does not mean that the rest of the OpenBeOS KITS were built on top of this kernel or that they will work with it. What the fork means is that the OpenBeOS guys are happy with the current development of NewOS to allow them working with a forked version that will have the right kind of features to continue development the way the BeOS dictates.
We are a long way still from a fully working BeOS clone.
Aside from being an homage to BeOS (which is kinda creepy if you ask me), what is the point of all this? If everybody works hard and doesn’t lose interest, you’ll eventually have a clone of a product that died years ago. Why not do something current?
if you check the http://www.obos.com (i think thats it) and read around you will find out that version 1.0 of obos is to be a complete clone of the last beos version, after version 1.0 of obos tho, they are already taking ideas and concepts for what version 1.x/2.0 will be. i believe that after 1.0 is out, they are going to stop with the “lets clone beos” to the “now lets make an even better system”.
IMO, because they want to re-create BeOS, it makes sense to start up where R5 died off. It is good to have some consistency in the evolution of the operating system. If they started on something completely new, it wouldn’t be the “same thing” as BeOS and the BeOS community wouldn’t take much interest. Once they have a working replica of R5, then the OBOS team can focus on where they want to take the OS. Of course, I could be totally wrong.
sorry, the url for openbeos is http://www.openbeos.org (which ofcourse it says in news release.
rude: you posted like 1 minute and 9 seconds after me, quite impressive, i totally agree with you by the way
may i ask whats your problem? Seems that every time i read a rude remark i look up to see your name as its author.
What constitutes the “death” of an operating system?
Hi,
try this – the BeOS Developer Edition:
http://bezip.de/app/1194/
This is an enhanced distribution of BeOS Personal Edition with additional and updated software put together by some BeOS fans.
Ciao,
Sebastian
Yeah, it’s really nice. The tracker crashes on my system the instant it boots up.
mike burns and Rude Turnip–
Yes I get the face value statements, but when I consider equally obvious factors like the passage of time and different circumstances, the stated reasons seem…a bit goofy.
I can understand a misguided parent who has lost a child thinking (under the delusions of grief) that if they could clone said child, that everything would be just like before. But even if it was possible, other factors like the inevitable passage of time to plain old entropy would ensure that the resultant creature is not the same child.
By the time the clone would reach the dead child’s age, the child would have been twice as old, and would have grown accordingly. Same with BeOS. If there is no working system now, by the time you get to the starting point the world will be a very different place. And I suspect that the BeOS frozen in time will be passe by then. Is that a good starting point — behind the times?
Richard Fillion– My problem? Right now it is you. You’re the one who is choosing to pick fights and not discuss operating systems.
DavidGentle– What constitutes the “death” of an operating system? Well, when you can no longer buy a copy, it’s a pretty good sign! Personally I’d make the time of death the minute development stops. There used to be an exception — OS/2 nutcases. They were so obsessed with OS/2’s demise that one just about had to throw them a bone, and let them rant on and on about how OS/2 wasn’t dead. It was entertaining at least. I don’t see too many of them any more…
I’m not a member of the OBOS project, but this has been stated quite a few times…
What reason to re-create an OS originally released in 2000? Simple: It’s a consensus on which the team members can agree upon. No time wasted on endless discussions about APIs, interfaces etc. etc. Instead they can progress relatively quickly, as can be witnessed at the moment.
And if you want to go down the “why not use Linux” road: I’ve tried Linux, several times. I don’t want Linux, it doesn’t appeal to me at all. I want a cleaner, more user friendly alternative OS. OBOS might be the one.
The NewOS website says it runs on the Sega DreamCast as well. Does this mean there will be a OBOS for DreamCast as well?
The NewOS website says it runs on the Sega DreamCast as well. Does this mean there will be a OBOS for DreamCast as well?
I dont’ think so. The OBOS-Team will first focus on the x86 Platform (there won’t even be a Mac Version).
Sebastian: The BeOS DE is, if I am correctly informed, just a copy of the PE edition, that has been enabled to “stand alone”, combined with some goodies (drivers, apps, etc.). Here in my box it works all right, but I have heard of other ppl having all sorts of problems. My install crashed completely when I tried to install BONE/OGL/MEDIA. So, if you would like to really play around with Beos 5, I still recommend people to try getting a real copy of BeOS R5.
Speaking of that, would there be a problem using a pirated copy of BeOS 5 now? Legally: No! … But morally speaking? Is there anyone still selling BeOS that would be hurt by people using pirated BeOS? (No, not rethorical question.)
Speed: By the time the clone would reach the dead child’s age, the child would have been twice as old, and would have grown accordingly. Same with BeOS. If there is no working system now, by the time you get to the starting point the world will be a very different place. And I suspect that the BeOS frozen in time will be passe by then. Is that a good starting point — behind the times?
Well, time is not all that matters. Imho, Windows have not grown (much) better the latest five years and neither have Linux. And I can’t see any of them changing direction anytime soon. So, in my opinion a ressurected BeOS R5 with updated drivers, that works on new computers will still be able to compete when it gets released in a year or two.
Another good thing: You will not have to use OBOS. If you are happy with Windows or Linux, stick to them. I and many with me eagerly looks forward to the first OBOS release.
…there be a problem using a pirated copy of BeOS 5 now? Legally: No! …
replace with:
…there be a problem using a pirated copy of BeOS 5 now? Legally: Yes! …
I bought my copy april, friend of mine bought it 2 weeks ago, at the same official reseller.
Wow! What happened to my english? Sorry …
I think you are forgeting something, I don’t think many BeOS users are 100% BeOS users, I bet most use at least 1 or 2 of the other platforms daily since it is hard to use BeOS 100%. This gives us a very good comparison about what is good & bad about each OS. I like to view the Windows, Linux, Mac as a triangle of OSs, each very distinct from the other as are their users but gradually merging to a common std. And BeOS is already at a pt somewhere in the middle as are its users.
It shares culture, clean style (& ppc?) with Mac.
It shares same cheap x86 as Windows & Linux.
It shares the nix cmd line with Linux & MacOSX.
It shares with Linux a more developer fringe feel.
I could go on
There is very little in BeOS that isn’t available or couldn’t be duplicated in one of the other platforms except perhaps speed, responsivness etc.
It already has many of the best attributes of the other 3 plus it has the speed I can’t get in the others. But it doesn’t have much software or drivers true, I can barely live with what it has, but I have all 3 other OSs to fall back on.
In a more polarised world most users don’t care for some of nix cmd line, or Apple or MS, or cheap HW, expensive HW, or over glitzed styling or no style, or fragmentation or OSS but to each his/her own.
If BeOS does finally go away, most of the apps that I use exist under Mac, Windows, Linux are actually in better shape (Gobe or sim, Opera, MW-IDE,….) & a zillion others so I would really only miss the Tracker exp. But I would still end up using 2+ OSs.
But if OBOS can resume where R5 left off, I don’t think time will have been wasted. Look at what is happening in x86 HW, the world is moving to more universal std HW like FW, USB2, SATA, generic chipsets, few graphics chips to worry about, ISA is long gone. Of course drivers still have to be written, as they do for Linux.
So why do people climb mountains, beats me, but nothing you say is going to stop me using any of the above OSs.
sorry for long rant
NewOS’s license is (from the website) “essentially the BSD license”. OpenBeOS (or whatever it will soon be called) is using the MIT-license.
What I don’t get is, why would anyone contribute *OS* code to a project licensed in this way? I mean, suppose you come up with some cool new file system, or algorithm for quickly handling digital video internally, or a codec, or whatever. Under the MIT-license (and its ilk), MS can come along and pick it up to use for themselves. Nothing inherently wrong with that, except that they might also change it (ever so slightly, in some undocumented way) so as to make *your* version incompatible with the MS world. Not sure how clear I wrote that, but I think you know what I mean.
For OS code, the GNU GPL is much preferable (imho). With the GPL, your work can’t get hijacked and possibly turned against you in the way mentioned above.
Anyhow, because of this, I would think that it’d be much more rewarding to contribute OS coding work to something like Cosmoe.
Someone on this list (I believe) once said that folks like MS look at devs who contribute BSD/MIT licensed OS code as suckers doing MS’s work for them. That sounds correct to me.
—j
I mean, suppose you come up with some cool new file system, or algorithm for quickly handling digital video internally, or a codec, or whatever
There is nothing stopping anyone from developing such things under the GPL! That is not really the “OS”, but more an add-on to it. Hell, you don’t have to release the code period to any such things.
However, if OBOS was under the GPL, you would HAVE TO release the code, if you linked to anything OBOS. By OBOS being more “free” than the GPL allows, developers have more freedom as well, which, IMO, is a very very good thing. Especially if you actually want to make any money.
Under the MIT-license (and its ilk), MS can come along and pick it up to use for themselves. Nothing inherently wrong with that, except that they might also change it (ever so slightly, in some undocumented way) so as to make *your* version incompatible with the MS world.
To clarify that clumsy remark,
1. You contribute something cool to your chosen OSS OS.
2. It gets adopted by your OS — neat!
3. It starts getting adopted by other OS’s as a standard.
4. MS notices it.
5. MS adopts it (with minor changes, to make it incompatible with yours) for use in their next version of Windows.
6. 90% (or whatever) of computer users begin using your (ahem, MS’s) cool new feature.
7. All of a sudden, you notice that:
A. You’ve gotten no credit for your work.
B. You’re cool new feature is incompatible with the
MS-replacement that everyone is now trying to
emulate.
C. New commercial s/w that may have been written using
your new feature (for your OS even) will now use
MS’s instead. Sorry. :/
A. You’ve gotten no credit for your work.
if you read most MIT type licenses, if you use the code, you don’t have to publish the source, but you do have to redistribute the license, which, has your name it it. so, you get credit. though, the user of the code, cannot use you name for marketting, w/o your permission.
There is nothing stopping anyone from developing such things under the GPL!
The distinction here is whether or not the OS links to and runs your code. In that case, if your code is GPL’d, it can’t go in OBOS.
However, if OBOS was under the GPL, you would HAVE TO release the code, if you linked to anything OBOS.
Not if the OBOS system libs were LGPL.
…Especially if you actually want to make any money.
Funny, I just bought Visual SlickEdit for Linux for $300. I guess you *can* make money selling s/w for a GPL’d OS (as long as the OS’s libs use the LGPL).
—j
The distinction here is whether or not the OS links to and runs your code. In that case, if your code is GPL’d, it can’t go in OBOS.
no, the distinction is if the code is part of OBOS or, distributed separately. the examples you gave were add-ons, and therefor can be GPLed.
Funny, I just bought Visual SlickEdit for Linux for $300. I guess you *can* make money selling s/w for a GPL’d OS (as long as the OS’s libs use the LGPL).
Funny, I never heard of Visual SlickEdit. I wonder if any of that $300 went to marketing?
no, the distinction is if the code is part of OBOS or, distributed separately. the examples you gave were add-ons, and therefor can be GPLed.
Hmm… thanks. I’ll go back and read the GPL more carefully.
I never heard of Visual SlickEdit. I wonder if any of that $300 went to marketing?
http://www.slickedit.com
They run ads in DDJ. Quite the spiffy editor/ide.
Sebastian, have you ever posted anything usefull? Every time a BeOS topic appears, you post the same text.
Everybody who is interested in BeOS already knows about the Developer Edition.
Now stop posting the same crap all the time!
By the time the clone would reach the dead child’s age, the child would have been twice as old, and would have grown accordingly. Same with BeOS. If there is no working system now, by the time you get to the starting point the world will be a very different place. And I suspect that the BeOS frozen in time will be passe by then. Is that a good starting point — behind the times?
I’m guessing you are totally unfamiliar with the Glass Elevator project being run parallel to the OBOS project?
KAMiKAZOW: Ok, I’ll agree, I’ve done enough advertising for the guys putting together the BeOS Developer Edition – I’ll stop it! Please don’t mind it, they have done a great job in my opinion, so I thought they deserved it.
Ciao,
Sebastian
Well, time is not all that matters. Imho, Windows have not grown (much) better the latest five years and neither have Linux
>>>>>>>>
Think about this sentence for a minute. Are you honestly going to tell me that WinXP and Linux 2.5.x are anywhere near the same OS as Win95 and Linux 1.x? Hell, the the past year Linux has gotten stuff like a journaling FS, preemptibility, block-layer rewrite, new driver API, new scheduler, etc. It’s now to the point where it beats BeOS in every way except in GUI speed/elegance.
Well, time is not all that matters. Imho, Windows have not grown (much) better the latest five years and neither have Linux
>>>>>>>>
Think about this sentence for a minute. Are you honestly going to tell me that WinXP and Linux 2.5.x are anywhere near the same OS as Win95 and Linux 1.x? Hell, the the past year Linux has gotten stuff like a journaling FS, preemptibility, block-layer rewrite, new driver API, new scheduler, etc. It’s now to the point where it beats BeOS in every way except in GUI speed/elegance.
“By the time the clone would reach the dead child’s age, the child would have been twice as old, and would have grown accordingly. Same with BeOS. If there is no working system now, by the time you get to the starting point the world will be a very different place. And I suspect that the BeOS frozen in time will be passe by then. Is that a good starting point — behind the times?”
IMO cloning BeOS R5 has it’s advantages:
-Cloning BeOS takes less time than designing and creating something from scratch. Ordinarily, work on most of the OS components wouldn’t have started yet without a usable kernel. But because there is already a working system to test them on, every part os the OS is being developed right now.
-For me as an R5 user, I can start using some of the work of the OBOS team before there is an R1 release. For example, there is already an improved mail-daemon available. Hopefully I’ll be able to replace that crappy net_server with something faster and more reliable soon too.
-While a new design might be more up to date, it would have to start out with no apps at all. Granted, BeOS doesn’t exactly have an overwhelming number of apps, but still some is better than none. For example, there will be a great browser in the form of BeZilla, or perhaps even something less bloated that uses the Gecko rendering engine. There’s a great file manager and desktop in OpenTracker. There’s a great mp3 palyer called SoundPlay which is still being developed. There’ll be a DVD player (VideoLAN). If we’re lucky, Gobe might even decide to release their latest Office suite, since sources at Gobe suggest it would take little more than a recompile.
It’s now to the point where it beats BeOS in every way except in GUI speed/elegance.
What about consistency?
How many sound servers are there for Linux? (oss, alsa, arts …) Some are compatable, but, on my linux box at work, many apps can’t use the sound card, while other apps are.
The different GUI toolkits give different apps different look and feels.
Mime database? Many different fs browsers maintain their own list of apps/filetype association
etc, etc, etc.
Not to mention the best user oriented filesystem ever. BFS rocks!
-Bruno
>>Not to mention the best user oriented filesystem ever. BFS rocks!
-Bruno<<
That’s because you’re behind the BFS helm now… keep up the great work Bruno!
Here’s a question of any OpenBeOS developers reading this: are you going to copy BeOS to the extent of copying the crappy virtual memory manager? From what I’ve seen and been told, the BeOS VMM was inefficient, didn’t support mmap(), and didn’t do cacheing at the file level (only at the block device level).
Do you intend to clean that up at all?
It’s now to the point where it beats BeOS in every way except in GUI speed/elegance.
What about consistency?
How many sound servers are there for Linux? (oss, alsa, arts …) Some are compatable, but, on my linux box at work, many apps can’t use the sound card, while other apps are.
The different GUI toolkits give different apps different look and feels.
Mime database? Many different fs browsers maintain their own list of apps/filetype association
etc, etc, etc.
>>>>
None of that is technological. Linux provides a lot of tools to the user. You can compare Linux to BeOS. Linux is many things to many people, BeOS is one OS for one type of people. Nothing prevents people from taking the tools that exist on Linux and targetting a coherent distribution at one group of people. KDE-3, for example, is a damn good try at this. At’s all integrated, it has features BeOS could only dream about, it looks good, and its UI is quite consistant. As long as you use only KDE apps (of which there are probably more REAL ones than BeOS ever had) you don’t have to deal with all the sound servers and toolkits and whatnot. Everything is integrated and centralized, including MIME types. Just because all of those other tools exist doesn’t harm the platform in any way, you don’t have to use them.
you have to IGNORE most of the stuff out there…
re: speeds orginal comment.
To drum up developer intrest.
The NewOS website says it runs on the Sega DreamCast as well. Does this mean there will be a OBOS for DreamCast as well?
I dont’ think so. The OBOS-Team will first focus on the x86 Platform (there won’t even be a Mac Version).
Although most of the team is working on x86 is important to mention that the whole system is being designed with portability in mind. With that said, the current kernel (as it is almost identical in every way at this point) to NewOS does run on the Sega Dreamcast, and as the code matures towards the OpenBeOS goals keeping this port supported would be rather easy, as long as there was a developer interested. As far as the “mac” version goes, there are team members who are currently working on PPC support for OpenBeOS, including the new macs G3-G4s that Be never supported. However one difference on this platform (due to licensing issues with the metroworks compiler used on the PPC version of BeOS) the PowerPC version of OpenBeOS will not be binary compatable with BeOS R5 (and will instead opt for GCC 3).
To me, I agree with OBOS philosophy. They are building a clone of BeOS r5, but they are making it in such a way they could extend easily later on. Why? They want ISVs (ie, Free Software makers, like Bezilla) to continue developing, and when OBOS R1 is released, it could run Bezilla and other apps, and ISVs can migrate easily without much pain.
With the looks of the development (unless a flame war gets on the mailing list and developers leave), by 2004, it should be self hosting. I can’t think of one OS (except AtheOS and Cosmoe… but aren’t they cloning parts of BeOS?) that at least plan to have most of BeOS’s API features. For example, Tune Tracker would have taken more money for cost of development if it was developed for Windows.
So in other words, if you don’t agree with what OBOS is doing, just respect their decissions. If they fail (though i doubt it), poor them. You can say “See? I told you so.” But something tells me that wouldn’t happen.
I’m guessing you are totally unfamiliar with the Glass Elevator project being run parallel to the OBOS project?
Ah, why are you asking me? Don’t you know if you’re guessing? What is Glass Elevator? Why should I care?
Not to mention the best user oriented filesystem ever.
LOL, not hard when there’s only one user on it!
Do you intend to clean (VMM) up at all?
Right…as I have been asking. Wouldn’t it make sense to improve, rather than just clone?
What about consistency?
How many sound servers are there for Linux?
No offense, but that’s a pretty boneheaded assumption. Just because you can doesn’t mean that you must! Having choices is hardly a liability.
To drum up developer intrest.
In what — reinventing the wheel? Seriously?
They want ISVs (ie, Free Software makers, like Bezilla) to continue developing, and when OBOS R1 is released, it could run Bezilla and other apps, and ISVs can migrate easily without much pain.
Forgive me, but it seems an awful lot like you’re saying that all this work is just so that other people can have “busy work”. Surely there’s more to it than that! Is there grant money at stake? Something like that?
What about consistency?
How many sound servers are there for Linux? (oss, alsa, arts …) Some are compatable, but, on my linux box at work, many apps can’t use the sound card, while other apps are.
Linux? Consistency? If you believe that Linux would ever achieve consistency, go someplace else. I’m speaking this as a Linux user (and a zealot).
As for sound servers, there is one basic one, Alsa. This is part of the Linux kernel. Arts, for example, is a sound mixer. It doesn’t in any way replace Alsa.
As for some apps not getting sound, you should get the dependencies of the apps. If you use something like Debian (or Mandrake), when you download or install a app, it would download the dependencies for you.
Nontheless, Be’s sound server is really nothing to shout about comparing with today’s systems – after all, it’s old and outdated (for example, it doesn’t support Dobly Digital Surround 5.1, which OBOS can’t because of patent issues.)
However, I still can’t see any other OS now (except maybe Longhorn) that contains the features of Be, like BFS (hehe, Bruno). Below the hood, BeOS is better than most of today’s OS. On the surface, it is an outdate piece of junk (I mean, for example, does it burn CDs from the file browser, does OBOS have any plan support for Mount Rainer, most likely to replace floppies and so on? Naahhhh….)
I’m not insulting Be or BeOS is any way, OS features wise, when R5 was released, it was the best. Sadly, poor company management cause Be to die off and BeOS on the surface turn rusty.
—-
LOL, not hard when there’s only one user on it!
There is certainly more than one user using BFS.
Right…as I have been asking. Wouldn’t it make sense to improve, rather than just clone?
Which begs the question, how to improve without first cloning? They decided to outright clone R5 and after R1, they would extend it. So, instead of going into API debates, or design debates, they are actually doing something. They are building an OS faster than Linux or Windows or Mac OS. So what happens after R1? They start extending the APIs, adding new features and so on.
Besides, wasn’t the Linux kernel a Minix clone once?
In what — reinventing the wheel? Seriously?
If we didn’t want to reinvent the wheel, wouldn’t we all be using XEROX’s PCs to use GUIs?
Also, wasn’t Linux a Minix clone once? And where is it now? Third largest OS in terms of market share in the desktop field, second largest in the server market, fourth largest in the PDA business, etc.
Forgive me, but it seems an awful lot like you’re saying that all this work is just so that other people can have “busy work”. Surely there’s more to it than that! Is there grant money at stake? Something like that?
No, they are building a clone of R5. While doing that, interested OBOS developers can continue developing for R5 until R1 comes out. Go ask the people behind BeZilla why they want to continue development for R5? They would answer OBOS. Seriously, its a good stragety. If they made a clone of Windows (which isn’t the best design choice anyway), they would face a up hill task of reverse engineering an undocumented implementation of Win32, while stepping on patent stuff. Also, to throw in features found in BeOS and expecting applications to use it… so in other words, ISVs would have to port. So how about Mac OS – most of the features in it are patented, no point. Linux? There is an open source version. Why reverse engineer something that is already there?
So, just say they decided NOT to clone OBOS. What do they do? Debate about APIs and design choices? Because they are cloning BeOS in the first place, they don’t debate it. And if they decided not to clone BeOS, they wouldn’t be this far. Nobody is asking you to support the OS, but trust me, I’m very sure OBOS would be big one day.
(Besides, OBOS is a not profit community)
> > Not to mention the best user oriented filesystem ever.
>
> LOL, not hard when there’s only one user on it!
Let me try to understand this. Even if there was only one person using it, what effect would it have in it being good or not? No other FS implements features like like queries and node monitoring with the same integration level wih the OS that BFS does. Heck, in fact no other FS implements all features that BFS does.
If all you want to do is act like a troll, you’re doing fine. If you want to convince people of your beliefs, then you should at least try to talk about stuff you actually know about.
-Bruno
Tough I am part of the OpenBeOS project and the team lead for the OpenBFS part of it, I am speaking based on my point of view. Don’t take anything I say for granted.
The primary objective of the project is to clone BeOS (up to a point, I will explain bellow) for various reasons. Most important of all, in my point of view, is to set an specific objective and to reach it. We are doing this because we really believe BeOS is as good as we want to convince everyone else it is.
As I mentioned, we will be cloning it only up to a point. Some parts of it were essentially broken (net_server comes to mind) and the current OpenBeOS Net Kit does not intend to reproduce the same problems it had, but we are aiming at binary compatibility. Some apps may break and have to be recompiled (because they implemented workarounds or exploited know bugs), but we intend to allow most apps to work out of the box.
After we have completed R1, we will have a working OS that may not be up to date as we would like it to be, but we will have a solid foundation where we can improve upon and we *WILL* be doing that. From this point on we will do whatever we have to to make it the best possible OS. We may never achieve that, but we will try.
-Bruno
The OpenBeOS effort is progressing along faster than I or probably anyone else ever imagined would happen. Ok, the minor exception to this is the MIDI Kit and Game Kit. What’s with those teams, anyways? Can we expect to see a burst of ANY activity/noted progress from them soon (this month or the next)?
But I am far beyond “mildly curious” about what’s going on these days… I’m sweating and panting and salivating like a rabid animal, waiting for the next note of progress!
GO, OBOS Teams, GO! Y E A H !!!
As is an appropriate statement: “BeOS is dead… long live OpenBeOS!”
Jared
“OpenBeOS: The New Future We’re All Waiting For!”
To drum up developer intrest.
In what — reinventing the wheel? Seriously?
Linux… UNIX (Minux for the X86)
and believe it or not SOME PEOPLE want a BeOS-like system.
Yeap, thats right, not Linux (or UNIX of any flavour), Win32 or anything else. BeOS. But, as you have said it’s dead, so we are starting again.