The OpenBeOS project leader, Michael Phipps, sent us a status report and he even makes a guess as to when a packaged OpenBeOS alpha release might be ready: “The status is very good. The networking and FS groups are coming along really well. To the point where they are testing functional (albeit occasionally going to kernel debugger land) pieces. The other kits are mostly moving along quite nicely. Printing, Input and ScreenSaver are nearly done. Storage and Midi are moving very rapidly. You know all about prototype 5 of the app_server, and work is in progress on #6. Media is moving under the heroic efforts of Marcus, and there should be some exciting Kernel news shortly. As far as an alpha, it would all depend on what people are looking for. Anyone can download and build today. We encourage (and warn, since nothing is super well tested yet) that. As far as a completed, packaged alpha, I would not expect one for **AT LEAST** a few months. Probably more.“
Michael Phipps On the OpenBeOS Progress
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
I’ve downloaded the source, and built. Though, I did not have much time to test. Building should be A LOT simpler with the re-work of the cvs tree.
All I can say is, kudos to the obos team(s).
It’s great to know that my future OS is in good hands.
Just don’t forget about app development people! We need more than a handful of good apps to run when we roll out the red carpet for OBOS R1.
Hopefully the latest news will encourage people enough to start developing for BeOS again.
after reading up on Beos, then about obos, i am very excited about this operating system and the neat buzzwords that beos users get to brag about. So, as i wait the months for obos to complete(which seems to be a VERY short time for such a project) i wanted to download the Beod personal edition(i thought i read you can just dl the os and install without partitioning?) well, the link i found at bebits.com was for devtools, not the os. Can anyone help me out here? i really want to give this a try.
i withdraw my question on where to download, the devtools were the first in a list of downloads, at the bottom of the page, there was the correct link….grrr, so thats why they have scrollbars
The OpenBeOS project is making great strides and it’s great to see such important and difficult things such as BFS and networking getting close (I’ve tested the net_kit myself). However to say that an alpha release is a few months away, is simply promising way to much. This is a HUGE project and although parts of it are going much faster then any of us expected, there is still a long way to go. The app_server, media_kit, drivers, and let’s not forget the kernel are far from releasable points.
Don’t take this as a knock to the OpenBeOS developers, I can’t stress enough how great they are doing (and suprising everyone in the process), let’s just not get everyone’s hopes up by promising a release in a “couple months” that is more likely a year off. What can we expect from OpenBeOS then? Well I would suspect that within two months there will be usable replacements for the BeOS net_server and BFS. From that point other replacements will emerge (print_server, prefrences, input_server, kernel, etc) one at a time. Thus those of us who still proudly run BeOS will be able to slowly move over to OpenBeOS one step at a time, at first running a hybrid of BeOS R5 and OpenBeOS R1, and by about a year this hybrid will become almost exclusively OpenBeOS. At this point it will be possible to make a OpenBeOS release.
What is the deal? Suddenly there exists so many OS options, it is not just Windows and Mac, there is Linux, the new Amiga OS, and BeOS Open Source, ect, ect.
I still have BeOS Personal edition on my new PC, even though there is not a video driver for the cheap “card” I have. (SIS 630/730).
I keep BeOS because it is such a nice OS. Still, it seems to me everyone is going in too many directions. Its a pity everyone can’t agres on a “third OS”. And it would be nice it they went with BeOS, but I am sure if everyone could agree to refine one flavor of Linux or the Amiga OS, we could end up with a decent OS.
OBOS is really making progress, and fast! im excited and will, as soon as there is a some-what-stable release defeinetly, try it out and if i like I will make sure to contribute with some obos apps
Thank you for bringing a polite but realistic point of view to this conversation. A top-to-bottom, feature complete replacement for BeOS is no small undertaking, and a few months seems like an (understandable) case of overexcitement. I’d also like to point out that the beta cycle for this package could be a massive undertaking. Not only to eliminate bugs, but to get the new code to behave the same in all cases as BeOS, such as error codes. That’s what it will take to get everything from drivers to Tracker to Gobe Productive running on a BeOS clone. Not that it can’t be done, but it’s still quite early in the game.
well, the link i found at bebits.com was for devtools, not the os
You should try scrolling down the page a bit! =] I presume you want the Windows Installer;
http://www.bebits.com/bob/11317/BeOS5-PersonalEdition.exe
For completeness the Linux Installer is;
http://www.bebits.com/bob/11657/BeOS4Linux.tar.gz
And the update (for either installer) is;
http://www.bebits.com/bob/11319/BeOS-5.0.3-PersonalEd-x86-Update.zi…
i withdraw my question on where to download
Hey – two can make that mistake…. =]
Still someone might find the links useful…..?
A new “status report”, but what can you really find in the CVS tree:
– no kernel
– no drivers
– a incomplete (userland) network stack
– jokes like the app_server which is just like a wrapper to the BeOS app_server
– a big surprise: a working BFS, nicely recoded with the help of the BeOS leaked code. Just remember the guys who claimed to have the leaked code few months ago (BGA,…)
– small ‘preference apps’, working great, but always using just BeOS things (no OpenBeOS code used)
With more than 200 coders, it’s a pity.
Binary compatibility is already out if you look at the InterfaceKit classes (said to be given by B.E.OS by the Inferno guy, is it true?) and some SupportKit classes. The end of the dream.
BlueEyedOS: better? Not really, no binary compatibility and still NO binary releases except screenshots
Noone knows more about it, and noone seems to want to know more (see OSNews, BeGroovy,…).
I one day they release something, Michael will have ask himself what is trying to do…
The best thing could be a realistic ‘merge’. Where is BeUnited?
RD.
“However to say that an alpha release is a few months away, is simply promising way to much.”
They never said it was a few months away.. Notice the “at least” that Michael used in all caps? Jeez…
Adam
Another link that could be of interrest might be:
http://beos.cosmopolit.net/index.php?seite=Download
Where you can download (the unofficial) BeOS 5.0.4 DE (ISO). It was put together by some german enthusiasts with updated drivers, DivX support, etc. Installed it on my box alongside win2k and it runs really well. It could be worth pointing out that this version won’t be installed “inside” Windows, but as a regular OS.
Dude.. The reason why there is no kernel is because the OBOS kernel coders and network coders are working on the NewOS kernel. No OBOS hasn’t forked the OBOS kernel yet so what can you really expect?
Anonymouse coward, Why dont you stop complaning and make your own damn OS. You are rude to even accuse somone of stealing code. its open source, it would be reported and kill right away. people like you need to get a grip on life, they have only been at the project for a few months. They are not just taking on the task of creating an OS from scratch, they are trying to make it binary compatible. for the time that is spent on the project so far, they are at a very nice start.
If you are going to be an a$s, do it somewhere else, or show people you are better at it. until then shut your hole.
Hi,
this is great news!!! I am really hoping that OpenBeOS will become a full replacement for BeOS in the not so far future. Till then you can download 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
I might be dreaming, but wasn’t there a BeOS 5 Personal Edition that could be installed on Macs? I thought I saw people talking about it in forums, but never paid much attention to it. Now that I have a Mac to play with, I figured I would play around with different OSes until I can buy Lightwave and do actual work in OSX.
There was no code stolen to make the write capability of OpenBFS. Axel and BGA have done an excellent job implementing that on their own. Still alpha-quality … but implemented =)
>A new “status report”, but what can you really find in the >CVS tree:
>- no kernel
Ummm,
http://newos.sourceforge.net/
This is the kernel OBOS is based off of. The kernel kit is working primarily with making it ‘BeOS compatable’.
>- no drivers
Theoretically, any existing BeOS driver should work …
>- a incomplete (userland) network stack
The BeOS network stack (NetServer) IS userland.
>- jokes like the app_server which is just like a wrapper to >the BeOS app_server
Umm, again, no. They are rewriting the functions from scratch. They just happen to emulate and wrap it in a BeOS program for easy testing. You will notice too that there is a version that addresses the display hardware ether through a driver or directly. Although not as mature, it still works without the BeOS app_server.
>- a big surprise: a working BFS, nicely recoded with the >help of the BeOS leaked code. Just remember the guys who >claimed to have the leaked code few months ago (BGA,…)
Actually it is implemented from scratch using the BeBook as a guid, just like all the other parts of OBOS. The BFS system just has a slight leg up because there was actually a book published about the internal workings of the BFS.
http://www.amazon.com/exec/obidos/ASIN/1558604979/qid=1020272623/sr…
>- small ‘preference apps’, working great, but always using >just BeOS things (no OpenBeOS code used)
Most of the code is available. Some of it just hasn’t been checked into the main CVS yet. Each team has it’s own site and its own CVS and it’s own way of doing updates and testing.
>With more than 200 coders, it’s a pity.
I seem to remember Michael mentioning that most of the work for each team is being done by a few ‘hard core’ members, while the rest mainly just test, submit bug reports/patches, etc. Also, these people work in there SPARE TIME. Most have jobs and other commitments that come first.
A project of this scale takes year of work to get finished. Most people have no idea the complexity and challenges involved in writing an operating system, let alone one as mature as BeOS. The technical details alone put most other programming projects to shame. I, personally, am suprised with how much they have acomplished is such a short time frame. Good work guys.
Mr.Anonymous the Coward,
Your claims are just pathetic, including “no kernel”, “incomplete (userland) network stack”, “stolen BFS” and so on.
P.S. Sit down, peek any file from “non existing” kernel and think over if you can write yourself something similar. I’m absolutely confident, if you could, you would not have made posts like yours.
>- no kernel
Ummm, http://newos.sourceforge.net/
This is the kernel OBOS is based off of. The kernel kit is working primarily with making it ‘BeOS compatable’.
OpenBeOS will be based on NewOS, what’s more NewOS is not yet fully BeOS kernel compliant. Maybe didn’t you read the mailing list or studied the code.
>- no drivers
Theoretically, any existing BeOS driver should work …
Yes ,”Theoretically”
>- a incomplete (userland) network stack
The BeOS network stack (NetServer) IS userland.
Did I wrote: “an userland (incomplete) net stack” ? incomplete is the current stack.
>- jokes like the app_server which is just like a wrapper to the BeOS app_server
Umm, again, no. They are rewriting the functions from scratch. They just happen to emulate and wrap it in a BeOS program for easy testing. You will notice too that there is a version that addresses the display hardware ether through a driver or directly. Although not as mature, it still works without the BeOS app_server.
There are 2 things in the ‘proto5’:
– a ‘display driver’ which only wrap drawing functions to BeOS one…
– the code from YNOP which is really incomplete which implements really the drawing functions
But these 2 parts are simply not an app_server.
>- a big surprise: a working BFS, nicely recoded with the help of the BeOS leaked code. Just remember the guys who claimed to have the leaked code few months ago (BGA,…)
Actually it is implemented from scratch using the BeBook as a guid, just like all the other parts of OBOS. The BFS system just has a slight leg up because there was actually a book published about the internal workings of the BFS.
Humm,… call me stupid or not, but my full day job is project manager in a software company, I made a lot of things for BeOS years ago (before 4.5), and I’m really surprised to see a ‘superman’ who can with the BFS book rewrite it. The author just describes the mechanisms, not how are exactly used the millions of bits composing a BFS partition. How can I say that BGA use the leaked code? Simply because he said it on BeShare… and because I have this precious code.
>- small ‘preference apps’, working great, but always using just BeOS things (no OpenBeOS code used)
Most of the code is available. Some of it just hasn’t been checked into the main CVS yet. Each team has it’s own site and its own CVS and it’s own way of doing updates and testing.
Preferences apps are useless without an OS (!=BeOS).
Even if tomorow the NewOS is ready, you will not be able to run this apps: no libbe.so, no app_server, no drivers… and no binary compatibility (see the classes signature to see why)
Sorry for this objective point of view that noone wants.
RD.
-1 : Troll
Considering whoever posted this decided to be anonymous, it is reason enough to have serious doubts about his/her point of view… Anyway:
> – no kernel
http://www.newos.org. Already boots, supports SMP, has a better VM then BeOS ever had and is progressing quite well (tough it is far from being complete).
> – no drivers
The OpenBeOS kernel team is basically working on those. USB 2.0 support, for instance, is being written by Michael himself.
> – a incomplete (userland) network stack
Incomplete, yes (tough it is already possible to get some BeOS apps working with it). Saying it is userland just shows your ignorance. It is a kernel land implementation that can also be compiled as auserland app for testing purposes.
> – jokes like the app_server which is just like a wrapper to the
> BeOS app_server
You know even less than I thought.
> – a big surprise: a working BFS, nicely recoded with the help of
> the BeOS leaked code. Just remember the guys who claimed to have
> the leaked code few months ago (BGA,…)
Now this is a *SERIOUS* thing to claim. OpenBFS don’t have *ANY* code from the original BFS nor it is based on any code from it (unless, of course, what has been publically released by Dominic in his book). And no, I do not have access to any code from BeOS, though I am sure there are people who do (as it leaked at some point).
> – small ‘preference apps’, working great, but always using just
> BeOS things (no OpenBeOS code used)
Are you crazy or what? All the code in the preferences app has been created by the guys from the OpenBeOS project. It is obviously OpenBeOS code.
> With more than 200 coders, it’s a pity.
What? You say there are 200 coders working on it? That’s new to me.
> Binary compatibility is already out if you look at the
> InterfaceKit classes (said to be given by B.E.OS by the Inferno
> guy, is it true?) and some SupportKit classes. The end of the
> dream.
Again, you have *NO* idea about what you’re talking.
-Bruno
Bruno (BGA) is also co-author of MDR (Mail Daemon Replacement) and his solution turn around a BeOS Bug.
Bruno and others have made good works and good apps.
http://www.bug-br.org.br/zoidberg/
> Humm,… call me stupid or not, but my full day job is
> project manager in a software company, I made a lot of
> things for BeOS years ago (before 4.5), and I’m really
> surprised to see a ‘superman’ who can with the BFS book
> rewrite it.
Maybe that~s just because you’re not as good at programming as you want to believe. The “superman” you refer to is Axel Dorfler who did an *AMAZING* job creating the BFS recovery tools and an even better job recreating BFS. Also several Be engineers happen be my friends and they gave invaluable help (Travis, for instance, pointed us some hidden IOCTLs present in BFS). Also even people outside the BFS team helped us (someone I don’t remember right now pointed that BFS always mounts volumes called “__ro__” as read-only).
> The author just describes the mechanisms, not how are
> exactly used the millions of bits composing a BFS
> partition.
He describes most relenat stuff. And give pointers to others. What could not be figured out using the book has been figured by experimentation. In case you didn~t notice, one major part of the code that is still missing is exactly the journal support. You probably know, as you’re such a great programmer, that this part can not be made fully compatible to the way BFS works exactly because it is something live. The end result is that we are implementing the Journal in the way we think it should work and it *WILL NOT* be even compatible to the BFS journal.
> How can I say that BGA use the leaked code? Simply
> because he said it on BeShare…
I also said I want to conquer the world. *IF* I ever said that on BeShare, you may be sure that I was joking.
> and because I have this precious code.
Good. Let’s get this straight… Would you please say where exactly we are using BFS code? I am really curious about it.
-Bruno
###I still have BeOS Personal edition on
###my new PC, even though there is not a
###video driver for the cheap “card” I
###have. (SIS 630/730).
You too huh? Let’s hope that part of the
OBOS project will be updated drivers….