Be Inc, the well known OS company that produced the multimedia-oriented OS BeOS, was acquired by Palm Inc. a year ago. The further development of BeOS was stopped, since Palm doesn’t seem to care about it. Will the BeOS spirit survive with the BeOS clones currently flourishing?
Since BeOS has stopped development, a lot of projects started to appear aiming to create a fully functional BeOS clone. Such projects are BlueEyedOS (formerly BlueOS) and OpenBeOS.
BlueEyedOS
BlueEyedOS aims to built a new BeOS on top of the Linux kernel and using XFree86 for the GUI. The project says that they already have the “equivalent” of the BeOS kernel and accelerated OpenGL that they need. Linux kernel also has a lot of hardware support and it is actively maintain by the open source community. Since Linux kernel has already a networking stack, the project will use it. As for programming tools they use Sun Java and GCC.
The project is developing the GUI (graphical user interface) to match the BeOS one, fast and multithreaded, some graphical applications to use the “basic components” like devices, user preferences, network, etc; a new API compatible with BeOS, but with some improvement; a new “structure”;
An architecture providing a flexible OS which can boot quickly and uses servers. Like BeOS, the BlueEyedOS API is composed of kits (BeOS source compatible kits).
OpenBeOS
OpenBeOS aims to recreate the a fully compatible BeOS-clone from scratch. And of course it would be open source. Individual servers and APIs (known as kits) are being re-written from scratch by an enthusiastic team of volunteers who want to continue the revolution started by Be Inc. The kernel is being based on NewOS, a microkernel written by a former Be engineer, and adapted by a dedicated team of hard core programmers. Under the guidenance of Michael Phipps, the project Head Leader, the project does well; it seems well organized and work is being done.
Already, they reached OpenBFS (BeOS File System) module to beta. It seems to work better and faster than the old once, and it keeps the compatibility. The development continue, we are curious to see what they have next.
A lot of programmers have joined the OBOS team, making the new clone very popular among the old users.
Opinion
I believe the effort made by both teams are much appreciated. My personal opinion is that the OBOS team approach is better, because they try to create a clone of BeOS as-is, not using a pre-build kernel, like Linux which doesn’t matches the BeOS kernel architecture.
Since Palm won’t develop BeOS any further it is a shame to let BeOS spirit die. BeOS R5 (the latest official BeOS release) won’t run on new hardware (patching BeOS to support AthlonXP or Pentium4 can be complex for newbies), so it would be impossible to keep running BeOS in a few years ahead. Lets hope these projects will give us the OS of the future, as OBOS team claims.
Of course they will make it. It just takes time. I didn’t see YellowTab mentioned in the article though. What about Zeta?
Thanks primarily to the great efforts of those working on OpenBeOS I have found the community to be more active and excited then it even did during the last year before Be Inc sold BeOS to palm.
The OpenBeOS team is very commited to the project, hard working, and keep the community well informed about progress (much as Be Inc was in the early days). There newsletters are very reminisant of the old Be Inc and the connection it had with the community.
So go OpenBeOS! This project and it’s members are exactly what the BeOS Spirit has been, it is surviving and will continue to survive for quite some time.
I love BEOS and check there site daily for updates.
IMO, OBOS embodies the true spirit of BeOS, but work by other projects on the Linux kernel provide a useful ground work for a fall-back plan. There can be cross-pollination between BeOS clone projects.
It’s a bit like BeOS was originally designed for custom BeBox hardware (purists may mention the Hobbit, other’s may think I’m referring to LOTR). When the hardware business did not seem worthwhile they ported to Mac hardware. When that went sour, they ported to x86. When that went bad a bunch of people started OBOS and other projects.
Brian Swetland has a quote page which has the following quote:
Change 966 by ceej@choline on 2000/07/17 16:34:25
to everything (merge, merge, merge)
there is a season (merge, merge, merge)
and a time for every checkin, under heaven
a time to add files, a time to delete
a time to branch, a time to sync
a time to edit, a time to revert
a time to laugh, a time to weep
Mostly the latter, though.
Is it the constant bickering, whining and in-fighting thats i going on? Is it the “introvertedness” of the community?
But, i’d like to give props to every team that tries to recreate the operating system many of us love and wish you continued success.
What exactly is that opinion/news item for? IMO it’s pretty redundant. The technical and organisational differences of the two projects were merely scratched at the surface. Please be a little more in depth. Even BlueEyedOS, who appear very closed mouth, have more info on their page…
Go OBOS!!
/me wonders what people will be able to do when those OS gets finished ? What “killer” apps will be available ? Gobe, ok, but what application would make me switch from w2k to one of the clones ?
One thing is sure if one of the clones supports PPC, il try in on my beloved bebox.
—
http://islande.hirlimann.net
From the title on the OSNews-frontpage I’d expect a little more depth about BeOS-clones. For example: what is YellowTab up to at http://www.yellowtab.de ? Do they have some new plans ?
And what about AtheOS and Syllable ? These are not real BeOS-clones, but take lots of ideas from the original BeOS. It would have been nice to see all BeOS-related initiatives listed here. But perhaps that’s something for a next article.
Just visited BeUnited.org again and they had even more projects to recreate BeOS on their list:
Cosmoe: http://www.cosmoe.com/
Leonardo: http://www.intuiware.com/leonardo/
And the correct URL for YellowTab was http://www.yellowtab.com
Just a quick note:
There are “other” communities developing PPC based hardware which may be of interest as possible new homes for OBOS (to keep a healthy distance from Apple who helped kill the original BEOS).
There is the (soon to be released) AmigaOne (check http://www.amiga.com for details).
There is also a “competing” Amiga Clone called Pegasos which may also be of interest (I think it’s by a company called bplan but I do not have the link handy)
Just some thoughts
Regards
Darren
More substance, please.
The Be-curious can find information at http://www.beunited.org/ …
I developer for OBOS, since it aims to recreate my Beloved OS, I aims to stay more compatible with the currently API. To me, this means a great deal. So screw you BlueEyedOS, you’re full of **** (fill in the blank
). You guys don’t know what you’re doing 
I hope support for the OpenBFS is included in kernels of open source OSs such as Linux or FreeBSD. Does anyone know if there’s a licensing problem with including the source of that file system in those kernels?
If other projects wanted to, I don’t think there would be aby licensing problems. The MIT license OBOS is under is generally less restrictive than say, the GPL.
Er… there isn’t any licensing problem. There is a programming problem. You can’t just include that source into the linux kernel, for example. There is a lot of work to do the porting.
Yes! Not only will they deliver a desktop OS, they will deliver a Server OS as well!
The HandHeld / IA OSes are not enough. There is still plenty of money to made in the desktop / server OS market after Microsoft remedies go into effect.
Long Live BeOS!
ciao
yc
Are you sacarstic? Cause you sound sacarstic….
I have nothing against the project. Hell! I have nothing against any project. But this is (IMO) the most ridiculous name for an OS. I’d like to know the reason. If it’s just to have a B and a E, Black Elephant would have been less ethnically connotated, and Beautiful Elegant more adequate, no ?
There newsletters are very reminisant of the old Be Inc and the
connection it had with the community.
A pale shadow of the former. There’s only one JLG.
I always thought MorpheOS would be a cool name for an OS
There is already one named MorphOS.
Black Elephant. I like that.
I’ve suggested it before, I’ll suggest it again. Two names I think would be great for operating systems, one in particular to the BeOS front:
WindOS
Phoenix <- rising from the ashes…
How many Be Os apps are there out there? Are new ones being written?
If it would improve perforance and stability, OBOS and blue-eye should ditch Be OS compatibility, and come out with a superior API.
Does either have the fragile base class problem?
In my mind, OBOS and blue eye should look at all the mistakes BEOS made, avoid compatiblity with rare Beos software, and start fresh, with a newer, more mordern and incompatible API.
Be inc itself broke compatibility in its updates i.e r4 to r5
Also, is C+ the best programming language for a new os? What about writing a new OS in Objective-C, or Java, or even C#? there are other object oriented OS’s out there as well.
if it was written in Java, it could run on any platform. objective-C is better OO than ordinary C+
anyone has any comments?
If I wanted Linux, I would use it. I want a cool desktop OS, not a BeOS-compatible OS.
The mixture of QT and GTK+ apps is bad enough. Who does really want yet another GUI toolkit?
If you want a BeOS-compatible OS, then help WinBe and you have made Windows BeOS-compatible.
The BlueEyedOS guys are not even able to release anything except for screenshots… they “fear” somebody could fork their work.
I wish all of the OS Beos teams the best. BeOS is still such a sweet OS to use, we all know of its linitations and bugs, and hopefully, these guys will produce something even better.
HarjTT
The mixture of QT and GTK+ apps is bad enough
How is having multiple widget toolkits under linux any worse than having multiple OSes under development? At least you can easily use Qt and Gtk concurrently.
If it would improve perforance and stability, OBOS and blue-eye should ditch Be OS compatibility, and come out with a superior API.
I’ve never used BeOS, but from the comments I heard, its API is, at least, as good as any other OS’ API. I don’t think it had performance problems, also.
In my mind, OBOS and blue eye should look at all the mistakes BEOS made, avoid compatiblity with rare Beos software, and start fresh, with a newer, more mordern and incompatible API.
Designing an API takes time, resources and leads to interminable teological discussions. The project wants to avoid that by just cloning something that already exists and currently works. After everything is done, they’ll start thinking about improvements in the design.
Also, is C+ the best programming language for a new os? What about writing a new OS in Objective-C, or Java, or even C#? there are other object oriented OS’s out there as well.
Same as above. As BeOS’ API is in C++, it is natural to reimplement it in the same language. Bindings to other languages will come with time.
Hi OSNews fans,
As President of beunited.org I just want to thank OSNews for all the great press coverage they continue to provide to our community and our efforts.
I am always surprised to see the amazing level of interest in any article that is posted here about BeOS. Opinions are always fiesty and fresh with new ideas about how the world should be. I congratulate everyone for taking the time to comment.
We here at http://www.beunited.org/ believe that the development of multiple versions of BeOS using open source approach is a good thing. It provides competition, and allows developers who have different approaches to flex their coding muscles. Our view is that we support all of them, but unlike the history of the Linux and UNIX splintered development we submit that at this early stage the cooperation of the BeOS projects is vital to maitaining compatibility though the development of a common set of standards. To that end beunited.org/ provides a forum for the members of these projects and the community at large to come together and define these standards. As a non-profit organization we have no part to say in the standards themselves, we only provide the forum and process for their creation and publication.
Visit http://www.beunited.org/ to get the latest news, code, and program announcements about our standards and development portal releases.
Thanks
Simon Gauvin,
President, beunited.org
Hmm, seems like the OSNews parser does not like hrefs embeded in comments… oh well.
Here is our address:
http://www.beunited.org
Hi all!
Hey guys, come on! Do you know that Linux kernel is the more advanced? It is already finished. Personally, I don’t like much of Linux. I LOVE BeOS of passion. But what’s the problem of use Linux as our kernel? OpenBeOS uses NewOS, correct? What’s a difference?
We not release not still, because we wants the things complete first.
Wait and see!
– Michael Vinícius de Oliveira
– BlueEyedOS Webmaster
I love what open Beos is doing but i wonder if there will be legal ramifications. I think recreating the os is awesome but could palm claim patent infringement and shut it down? Palm seems determined to shoot itself in the foot and anyone else that stands too close. Any feedback? Just for the record, i certainly hope that openBeos can exist happily sans interference from palm.
I have to wonder whether or not Palm cares at all. Personally, I don’t think they do. When Dano came out, what happened? Nada. When source code was released, the same. When was the last time palm had ANY contact with the BeOS Community?
Shooting itself in the foot? No. In the head? Yes.
I personally like OpenBeOS better, they are really continuing the community spirit as well as fufilling the “SHOW ME THE CODE!” mentality. Which is really impressive to me since a lot on the team don’t care too much for opensource, they’re simply committed enough to code anyway and help fufill the dream. I love reading the kit team project pages, participating in obfs testing, reading over the IRC interview logs, and following the mailing list. Community WAS AND STILL IS really important to BeOS, I think some projects have forgotten that. I don’t think it will ever overtake Microsoft, or even Linux.. but it will be one hell of a ride for us devoted BeOS fans, and shows that we’re going to go out with a bang instead of a fizzle, still united under our OS.
For me that would be both OS. OBOS as a closer reproduction of BeOS (even though not the same kernel) and BlueEyedOS as a more pragmatic version (a linux kernel, so what ?)
I stopped using BeOS when I changed computer two years ago, because it took months/year to have drivers just useable with modern hardware. So may be BlueEyedOS is a compromise and will not have a microkernel but it’s still for me the best solution so far. OBOS sounds great but I am not sure how they will address the same problem that BeOS had… drivers and apps ???
At least if they stay both source compatible (for the apps at least), they could benefit one from each other. But that means creating more a community of BeOS-clone users, not another war of OS.
re. Mark my words, PalmSource will deliver!
yc wrote:
> Yes! Not only will they deliver a desktop OS,
> they will deliver a Server OS as well!
The BeOS community has already been burned by a company owning the source code to their beloved OS going under and taking the source with them. I believe that even if Palm resurrected BeOS, it would still be a very hard sell.
re. Blue Eyed OS
Manik wrote:
> …But this is (IMO) the most ridiculous name
> for an OS.
Heh, Syllable’s a kinda’ funny name too…
I guess this is why it’s taking so long for the OBOS’ers to choose a new name.
re. is Be OS compatibility really necessary?
Anonymous wrote:
> [snip]
> …anyone has any comments?
The path for R1 has been chosen. Support them in their efforts rather than questioning their decisions about their own project.
re. The answer to all about BlueEyedOS
Michael VdO wrote:
> But what’s the problem of use Linux as our kernel?
> OpenBeOS uses NewOS, correct? What’s a difference?
I think that the problem people have with BlueEyedOS is threefold:
1. The BeOS community has an ingrained belief in doing things right from the beginning, even (especially) if it means starting from scratch. Folks have the impression that BlueEyedOS is compromising by including a crufty old kernel and display server like Linux and X.
You need to convince them (us) otherwise.
2. Your english. The few times I’ve looked at your site, I’ve seen many spelling and grammar mistakes. This gives the impression of incompetence (though I know you must be a very skilled and crafty lot to do what you do
.
3. The name. Sorry. I know it’s a subjective thing, but by definition, that means you needed to find a name that appeals the most to the majority. I bet that many folks think that the name was choosen simply so it would include the letters B-E-OS. [shrug] Not too creative. Though, I feel the same way about OBOS keeping the name OBOS. That is, I hope they don’t.
What about yellow tab and leonardo. Go to http://www.geocities.com/donan71 to see progress of all Open Source BeOS projects.
What both developers and users really need are videos, to see what the Be experience is like. I know that vid files are big and noncrossplatform, but I’ve seen companies use Flash movies to good effect, to show off their products.
Same goes with the Mac. Everyone likes screenshots, even when there’s no point. Vids are better for this.
Does anyone have a video demo of Be?
Anonymous: In my mind, OBOS and blue eye should look at all the mistakes BEOS made, avoid compatiblity with rare Beos software, and start fresh, with a newer, more mordern and incompatible API.
Blue Eyed OS plans not to have binary compatiblity. OBOS said they are faster implementing a design already there, and extend it for then on.
Anonymous: Also, is C+ the best programming language for a new os? What about writing a new OS in Objective-C, or Java, or even C#? there are other object oriented OS’s out there as well.
C++ is prefered by the developers because they feel they are more productive writing in C++. Java and C# aren’t suitable for OS development. Objective C may be good, but accroading to the guys at #openbeos is isn’t enough, probably would only be good enough for applications running on an OS.
Michael Vinicius de Oliveira: Personally, I don’t like much of Linux. I LOVE BeOS of passion. But what’s the problem of use Linux as our kernel? OpenBeOS uses NewOS, correct? What’s a difference?
OBOS wants to clone BeOS in and out. NewOS is a microkernel that is barely functional, BTW. I don’t think Linux kernel is suitable for a BeOS clone planing BC. For one, it doesn’t support features found in BFS…
Linux and XFree itself has its limitations, you would have to work hard to get around it. And with the amount of headway OBOS already made, frankly speaking, OBOS would be a better product in the end.
I think recreating the os is awesome but could palm claim patent infringement and shut it down?
IIRC, Palm doesn’t have any BeOS-related patents.
I think you’re wrong about patents.
Palm bought all BE’s IP including patents.
May be exBe folks have more information about patents but
IIRC one was about Be File System (or just compressed BeFS?).
Was there a patent for kernel ?
But I don’t believe there is any danger to OBOS team as long as they stay away from stolen BeOS code.
If there are published specs you can right software that matches these specs.
There is a nice demo video of BeOS 4.x floating around on BeShare… if you don’t have BeOS (which I suspect is the case) there is a BeShare client for Windows as well – http://www.ozone-o3.net/
Hi John!
> I think that the problem people have with BlueEyedOS is
> threefold:
1 – You forget that we have an app_server too. In future the app_server will be alone, without help of XFree.
2 – Yes, you are correct. My english is very poor. Here, in Brazil, my country, it is a problem, because the sintax of words is very different of the portuguese language. Don’t blame me
3 – Please note that “BlueEyedOS” is only a unnofficial name. It isn’t still the official project name.
BlueEyedOS FS will be == BeOS FS, ok?
– Michael Vinícius de Oliveira
– BlueEyedOS Webmaster
OK. So you’ve recapped what we already know, and you’ve put down a linux kernel that is already fully functional. The spirit of a dead OS has been exampled in the form of Amiga already. To finish, have you ever been through high school english?
>To finish, have you ever been through high school english?
Maybe he did not. If you are more grammatically correct than he is, why don’t you submit an article on osnews?
The content is what matters.
penguin wrote:
> To finish, have you ever been through high school english?
Ha! English is probably Michael’s *second* language. How many languages do you speak?
I’m not personally attacking anyone about the quality of the BlueEyedOS site — I’m offering constructive criticism: make a good impression, since the site is in english, have a native english speaker do some light editing to tidy up the language a bit.
By the way, “exampled” isn’t a word.
Also, Michael VdO: why is the BlueEyedOS team working on the XFree86 stuff at all if the plan is to have your own app_server? Why waste the resources on something you’re only going to throw away later? It sounds sort of reminiscent of the whole OBOS in-kernel networking vs. net_server debacle…
Why create BeOS again? It was once made, and we can learn a lot from it (just as we can learn a ton of stuff from AmigaOS, Windows, GEOS, NextSTEP, MacOS X, Linux, BSD, etc). Why make a copy when the same amount of work could give us something superior?
Ok, binary compatibility is one big thing, but other than that, there are never a reason to not make something good even better
Folks have the impression that BlueEyedOS is compromising by including a crufty old kernel
>>>>>>>
In most things, the Linux kernel kicks NewOS and even BeOS all over the place. I/O performance, VM system, even latency. In terms of the speed of primitive operations, Linux is *the* kernel out there at the moment. (Look up the lmbench marks sometime).
and display server like Linux and X.
You need to convince them (us) otherwise.
>>>>>>>
Great. While the OpenBeOS guys go and fix problems that Linux has already fixed, and live with software rendering for the next several years, us Linux/X folk will be enjoying a fast kernel with hardware-accelerated graphics.
I for one really like the BeOS way of doing things. That way was not necessarily to start over, but do whatever would lead to the fastest, cleanest system. Linux itself (not necessarily the UNIX userspace, but the kernel itself) is very fast and clean, and starting over is pointless. I hope the OpenBeOS project all the best, but I realize that their decision (in particular) to not use the Linux kernel relegates it to technical inferiority for a long time.
Depends for which latency you are talking about. As for i/O and VM, yes Linux is better than BeOS.
As for NewOS. You can’t compare it. NewOS is not even complete, not sure if it has a swap at all so far… It doesn’t even have a filesystem yet either (I mean the NewOS kernel, not the openbeos newos kernel).
> Ha! English is probably Michael’s *second* language. How many languages do you speak?
Yes, you’re right again. I speak my native language: portuguese. English I learn, here in Brazil, when I was a child. I’m not in the hight school still. Maybe the next year
I will be a lawyer or a computer scientist.
– Michael Vinícius de Oliveira
– BlueEyedOS Webmaster
To finish, have you ever been through high school english?
And have you been been through high school Portuguese? You have, ok then:
Ainda bem para ti, assim deves conseguir ler perfeitamente isto.
[Good for you, then you must be able to perfectly read this]
Um abraço para ti Michael, nao ligue ao que eles dizem.
>Why make a copy when the same amount of work could give us something superior?
Because some of us believe that the way towards a superior product is fastest reached by building OpenBeOS and then improving on it. It may be wrong, but noone gets hurt in the process. Only time will tell if we’re right.
Hopefully you can appretiate the hard work these people put in for what they believe in. And if you don’t you can always roll your own…
I think if people would go to the websites of the projects, they would find many of their questions answered, and wouldn’t ask them yet again here at OS News. The FAQ at OpenBeOS is especially comprehensive.
While I personally think that OBOS is the best way to go, I don’t understand the, er um, “dislike” of the Blue-Eyed OS project. They obviously won’t be able to create a 100% clone of BeOS using the Linux kernel, but that’s not their goal (read their website). People are overlooking the possibilities that B.E. OS could offer. And Cosmoe and Leonardo, for that matter. The more that can be learned about porting the BeOS API (which is perfectly legal with no copyright infringement problems, by the way, since the API was mostly published. Use of BeOS source code, the BeOS name, and possibly the icons are the only real legal problems), the more likely we will all benefit by having real (and good) alternative OS’s in the future, no matter how much the hardware changes.
And YellowTab is not making their own OS, they are making a *distribution* using the BeOS PE to start with (and which they apparently have a legal and binding contract for distributing), and probably OBOS when it is released (but you’ll have to ask Bernd to be sure about that).
If you want to discuss the merits or demerits of the technical choices made, try the forums. But let’s leave the FUD factors out, shall we?
“why is the BlueEyedOS team working on the XFree86 stuff at all if the plan is to have your own app_server? Why waste the resources on something you’re only going to throw away later?”
The reason is simple, it’s called ‘incremental process’.
Even if my english is just a “french touch’ed” english, I’m still a software engineer
. We use Xfree86 because we are developing an app_server(and the rest of the OS 
)!
The app_server really don’t care of the gfx card itself, it uses drivers. Because the ‘we will redo all the driver of the world while coding the app_server’ leimotiv is IMHO a technical nonsense. We delegate the low level rendering part to XFree86 in order to be able to develop something working well (drawing, app management, window managing, messaging…).
Later when all will be perfect, it will be relevant to use an other rendering method.
Regards,
Guillaume
“1 – You forget that we have an app_server too. In future the app_server will be alone, without help of XFree.”
I love you people! This more than anything makes me anxious to see where BlueEyedOS goes. My original worry was that the project was ‘just another X toolkit.’ I now wait to test drive the UI experience of BlueEyedOS.
Frankly, I see room for both OBOS and the BlueEyedOS projects. OBOS *could* be better, but it’s certainly going to take more work up front and take more maintenance in the long term than BlueEyedOS. The latter could be viewed as a linux distro, but with a truly radical change in the usability department.
Actually, XFS supports almost every feature of BFS (node monitoring, attributes) and more (ACLs, dynamic resize, real-time partitions) and comes with some really kick-ass tools (xfsrestore/xfsdump) to boot!
*Hug0*
Obrigado por tudo! Aqui estão meus sinceros agradecimentos!
Observação:- Quando sairá a nova versão (em binário e estável) do OpenTracker?
*Guillaume*
Thank you, master!
*mbishop*
Hey guy, contact Guillaume for more informations, ok? And thanks for believe in our project!
*others*
Você fala português?
Dou you speak portuguese?
Obrigado a todos!
Thank to all!
– Michael Vinícius de Oliveira
– BlueEyedOS Webmaster
Oi! meu nome é “Michael” mesmo. lê como se escreve. Nada de “Maicou”…
hehe!