Slashdot Mirror
Can BeOs Live On As Open Source?
OSBlue writes: "After Palm announced the buyout of Be, Inc.'s intellectual property & Technology and after some consequent indications from several key people that Palm has no interest at Be's products and especially in BeOS, a number of the BeOS believers tried to find a new home. Some found comfort in AtheOS, others joined BeUnited's effort to license the BeOS source code, while some developers formed efforts like BlueOS and OpenBeOS. OpenBeOS consists from a number of BeOS developers who are trying to recreate the BeOS Kits in a form of a new, complete and open source Operating System that has source and if possible binary compatibility with BeOS 5. One of the most important people in this effort, Michael Phipps, is interviewed by OSNews.
One is tempted to make the correlation:
:-) But ya gotta know what to pull the life support....
Amiga == OS2 == BeOS.
All ahead of their time technologically. All killed by stupid managment decisions. All still have freaks that refuse to acknowledge their death.
Oh, and I've used and loved all three
In my (granted limited) experience, and from what I've heard, I think BeOS is an excellent little OS, even if it does have a small niche. That said, I hope it can survive as Open Source Software. If you ask me the more OSS Operating Systems (OSSOSs?) we have out there the better off we are--giev more leverage perhaps against the bit monolithic software houses who produce their proprietary crap (MS WinXP anyone?)
Just IMHO...
Beware the Whyte Wolf.
With a gun barrel between your teeth, you speak only in vowels...
I've been an avid supporter of Be for years. I like the OS and have bought many versions. I think it would do well as open source for numerous reasons. The one that sticks out most in my mind is that it is truly an object oriented OS. I am an object oriented developer and I've looked through and attempted to write some code for linux, and I really didn't like it. BeOS on the other hand, is something I'd like to contribute to. I'm sure there are many other object oriented developers who feel the same way. Imagine if you could get the BeOS GUI to run on top of the linux kernel. What a product it could be!
mp3's are only for those with bad memories
we're trying to build a continent out of them. Perhaps you can extend Linux to provide what it is you like about BeOS.
"The Most Fun Possible on 4 wheels" is at SunBuggy in Las Vegas
Whil I do like BeOS, having used it a few times. I hardly see an Open source BeOS making any headway in the commerical market as linux has. As a hobby it seems like it could be fun and intresting. Other then that I don't see much in the future for BeOS
Snoozer.
I don't see why palm would want to opensource BeOS... They spent 11 million on it, and they are not likely to simply give it away. What reason would a pda maker have to release the source code to a desktop OS? Face it, BE is dead, and it ain't coming back.
I wonder what kind of stuff some of the old Be guys could contribute to GNOME/KDE.
Don't get me wrong, I like the BeOS, but with Linux starting to make headway and gaining momentum, it would be awesome for these guys to jump in.
For me, some of BeOS's goodness in KDE3 would make me drool, all technical issues asside.
Projects stay open and running even if no one is developing for it. More like a zombie state than anything else.
Since Be is a fully 0wned subsidiary of Palm now, I doubt we'll be seeing zombie BeOS anytime soon.
I think this shows are hard it is to have a niche OS market. Aside from wireless and imbedded technologies, it seems to me very unlikely that something as huge as an OS can exist on what basicly amounts to a niche. Where there are so many layers to an OS, adopting a whole new set of standards just to have a multi-media optimized OS just wasn't realistic.
That's what's so wonderful about distributions. You can make a niche distribution without the overhead of Yet Another EVERYTHING.
-- Ken Kinder ken@_nospam_kenkinder.com http://kenkinder.com/
Remember that Linux has only gotten really good and has gained the kind of acceptance it needs to be taken serious in the eyes of application developers in the last few years.
I would love to see an openBe implimentation, because it would be really nice to have an opensource OS geard toward multimedia instead of networking and programming. (Linux is many wonderful things, but it simply not geared for multimedia.)
It would take at least the same amount of time to reach 'critical mass'.
The next Slashdot story will be ready soon, but subscribers can beat the rush and slashdot the links early!
In a uniprocessor system the lack of reentrant kernel threads allows applications to control processor time. This cooperative multitasking is ineffecient, and systems intensive.
:)
Linux has never had cooperative multitasking. Never. Ever. It has always been preemptively multitasking. This one sentence alone shows that either you're wording things in a manner so tragically incorrect it's comedic, or else you simply don't know beans.
Another serious setback for Linux is the lack of a journalling file system. This makes data storage unreliable, and backup and recovery a dicey proposition. SGI said they would port the IRIX file system to Linux, but I haven't heard anything about this yet.
ReiserFS. Ext3. IBM's JFS. SGI's XFS, as found on Irix. Do I need to go on?
I would refute your post in depth and at length, but at this point I'm certain you're either totally uninformed or else trolling. Have a nice day.
If Palm isn't interested in Be's technology, and Be wasn't really a competitor to Palm, why exactly did Palm buy Be? Are there valuable patents or other IP?
Nope. As has been stated over and over again, BeOS cannot and will not be relicensed as Open Source software. There is simply too much proprietary, third party, technology embedded in it that it would take a lot of time, and probably a lot of cash, to strip away. It took SGI almost a year, if not longer, to get XFS released as GPL. Okay, the had to reengineer a good deal of the Linux kernel too. Besides, even if Be manages to strip out the proprietary bits you will most likely be left with a shell of code that will not compile, for a significant amount of time (*cough* Mozilla *cough*).
And IMHO, the "coolest" bits of BeOS have already made it into Linux -> 64-bit journalling FS with attributes, XFS! The other cool BeOS buzzword "pervasive-multithreading" didn't turn out to be that cool after all.
-adnans (ex-BeOS fool)
"In short: just say NO TO DRUGS, and maybe you won't end up like the Hurd people." --Linus Torvalds
I've used and love BeOS. I've also used tons of other OSs. I've loved things about all of them. with the exception of Windows almost all of those other systems have been considered niche. Some of those niche OSs have thrived. I'd love to see some air breathed into Be again.
-
If this can make Be live again that would be great.
If they could keep it living and get a good software base, I think i would have just found a replacement for Windows on my wife's machine, and probally mine.
I thought BeOS stood for "Be Open Source!"
Palaces, barricades, threats, meet promises
I managed to get a copy of the pre "freeware" version and I thought it was very well thought out, I would like to see more done with it...
oh and dont forget that it needs some Apps...
Thanks to file sharing, I purchase more CDs
Thanks to the RIAA, I buy them used...
The hole truth.
http://www.dibona.com/pictures/index.shtml
The eComStation folks have made the SMP kernel available in the client version, and they've been able to bundle an X server with it as well as get a third party (Connectix and Innotek) to create a version of Virtual PC for OS/2.
That, at least, gives OS/2 users the ability to run Windows and Linux in virtual machines.
Along with its legacy (DOS, Win 3.1) support and its ability to run ported Linux software via EMX, I think OS/2 is more interesting yet than most people realize...
Too bad it (and eComStation) are so expensive, but a demo CD may be in the works.
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
Now finally I've found out what the odd thing in the reddish squarish area in the Be logo is.
It's a skull, of course.
I could've found out earlier: it was just about the time they started to go down the drain when they changed the logo.
There is absolutely no reason to panic.
I guess Augustin and ESR will be posting to the front page soon.
Since C++ is a pathetically inappropriate language for OS development (IMNSHO of course;) I hope BeOS dies a graceful death.
There, now I feel better! :-)
Seriously if you want something that plays in mostly the same space (except it is written in C, is well designed, has hard realtime capabilities, and is a commercial success) check out QNX.
299,792,458 m/s...not just a good idea, its the law!
Galileo: "The Earth revolves around the Sun!"
Score: -1 100% Flamebait
Linux and X ?
Is it a good way to start a Desktop OS ?
- A consistent, easy-to-use and powerful GUI
- Strong multimedia support
- Runs fast on low-end hardware, and blazingly fast on high-end hardware
- A friendly, close community that welcomes new users
In addition, OS/2 has features that BeOS doesn't:- Great application support, namely DOS, Windows (including 32-bit Windows apps), OS/2, and even many Linux apps.
- Great hardware support (this was BeOS biggest problem).
- It's still being sold, marketed, and supported.
This last item brings me to eComStation, an OEM version of OS/2 that's effectively what OS/2 Warp 5 would be. I invite all BeOS users to check out eComStation. You'll be surprised what you find.And the men who hold high places must be the ones who start
To mold a new reality... closer to the heart
Debian for ex-BeOS users.
"There ought to be limits to freedom"
It's Madlibs!
--------------
Ok I am sick of hearing about THE GREEN PARTY and other wasted POLITICAL PARTIES talking about RUNNING FOR OFFICE. In my opinion there are only 2 POLITICAL PARTIES that are worth while thats DEMOCRATS and REPUBLICANS. I have a hard time understanding why these people just don't VOTE to ELECT DEMOCRATS or REPUBLICANS and even in that I don't understand why there is still a seperation between DEMOCRATS and REPUBLICANS . Both have there good points WHY DONT WE MERGE TO MAKE THE BEST WE CAN.
This would have been a pretty good troll if you had replaced "Linux" with "BeOS". But I am impressed that you at least got one biter. People on this site never seem to learn... -Signal 11
Of course, you wouldn't want to leave out NeXT, but then the freak that headed that up was bought out by apple, and see what we have today? Perhaps both Linux and FreeBSD/Darwin/OSX/Aqua could learn some things from the BeOS architecture.
- passion
Seriously if you want something that plays in mostly the same space (except it is
written in C,
is well designed
I don't know QNX, but my impression was that BeOS was actually pretty well designed, all in all (even if you hate C++)? Cleanly designed extensively multi-threaded system, neat filesystem, reasonably good scheduling (even if not hard realtime, much lower latency than any other desktop OS) etc. etc. So which flaws did I miss?
I like paying taxes. With them I buy civilization -- Oliver Wendell Holmes
1. They're trying to save an OS that never received its dues by extending its source availability.
2. Like the BSDs and Linux, this OS reworking has a relatively-limited goal (in comparison to other projects), in that they're trying to reimplement an already-extant system without extending its reach -- BlueOS sounds like it's bitten off more than can be chewed easily. It'll be difficult enough as is to do this well.
3. Phipps gives the impression he's not so interested in license wars at this stage. Given the stated community spectrum of opinion in the FAQ, this sounds prudent.
Here's hoping this article brings them others...
========================================
Death will come, and will have your eyes
-- Pavese
They bought them for *parts* of their technology that it can be re-used on a new PalmOS, but the main reason was because they needed the Be engineering team which is highly qualified for creating a new PalmOS. This is what David Nagel, the CEO of the new software subsidiary that Palm will be creating after the Be engineers join them, said at News.com a few weeks ago.
this is a bit of a lame answer but i'll post it anyway
besides the one(s) mentioned here:
http://osnews.com/comment.php?news_id=66
there are these:
http://bebugs.be.com/devbugs/hitlist.php3
From what I understand (having tried it a few times at various stages), beOS is like *nix with a kick ass graphics/windowing layer. Why not concentrate on bringing the beOS gfx layer to Linux ... I mean, the whole world seems to be holding its breath for the death of X11/freeX86. I'm not aware of the technical details, but is it feasible that the beOS graphics/windowing/desktop layer could be slapped onto Linux? If this could happen, beOS would be the desktop *nix box's killer app.
"Old man yells at systemd"
BeOS is faster and more stable OS than Windows2000, plus better multimedia systems then Linux, why not? I mean, aside from a great FileSystem and a kickass interface.
Non impediti ratione cogitationus.
I am an object oriented developer
No, get it straight. You are a number.
I don't mean to be frank, but what's the big deal? There are *plenty* of other OS's out there if one dies out, some much better. And most college students have shown that at its barest, anyone can custom-create an OS (or take an existing OS and modify it to their desires).
To me, it's like the people who still hang onto the Commodore 64 as "the ultimate gaming machine". I admit, I started my life on a TI-99/4A, but I haven't limited myself to it. This field is all about changing and adjusting and, quite frankly, the BeOS people are going to be left behind.
I disagree. The open source community has learned a LOT since Linux has taken off. A fully GPL'd or BSD'd BeOS would, if taken up by the community as the "officially sanctioned desktop OS of choice", become very popular very quickly. I have always considered Linux a server OS, but I've used it as a desktop because no other open source desktop OS is as advanced.
Maybe we need to get Apple thinking about asking Palm to open source BeOS. Would the Apple license be acceptable?
Scott
"Hokey religions and ancient weapons are no match for a good blaster at your side, kid."
1) You're a moron.
2) C++ is just as useful as C in many kernel applications. For example, Linux has many implementations of lists, trees, etc. Using templating, you can get 99% of the advantages of having specialized lists for each datatype without the hassle of actually maintaining specialized lists. Also, Linux has a lot of structures filled with pointers to functions. ABC's fill that role much more cleanly. Operator overloading (like overriding 'new') also come in handy. Plus, its not like anyone is suggesting writing the kernel using STL or somesuch nonsense!
3) The BeOS kernel is written in 'C'.
A deep unwavering belief is a sure sign you're missing something...
Why not take the things that BeOS did well and port them to a platform that has a chance of surviving? BeOS was beautiful to look at, the interface made sense, and the response to user input was blazing. All things that make for a nice desktop / multimedia OS, and all things that Linux could benefit from if it wants to compete on the desktop. Other aspects of BeOS such as networking, security, and hardware support were not so highly evolved and probably never would have been. Note that these are areas where Linux is somewhat evolved. I for one would love to see a marriage of the two.
Wanna get high?
We could call it BeSD :)
I think not, just opinionated. ;-)
I suppose my opinion could have been expressed less controversially, but in part this was a test to see how poorly Slashdot moderation actually works. That question has been answered at least. :-p
2) C++ is just as useful as C in many kernel applications. For example, Linux has many implementations of lists, trees, etc. Using templating, you can get 99% of the advantages of having specialized lists for each datatype without the hassle of actually maintaining specialized lists. Also, Linux has a lot of structures filled with pointers to functions. ABC's fill that role much more cleanly. Operator overloading (like overriding 'new') also come in handy. Plus, its not like anyone is suggesting writing the kernel using STL or somesuch nonsense!
However, BeOS does expose all system functionality through C++ APIs. Thus you get lovely artifacts like vtable slots reserved for future use. Also, it is often difficult or impossible to call C++ from other languages (with the notable exception of Java, which supports C++ directly with JNI).
3) The BeOS kernel is written in 'C'.
That's nice, but does nothing to address the issues with exposing C++ system APIs.
299,792,458 m/s...not just a good idea, its the law!
Galileo: "The Earth revolves around the Sun!"
Score: -1 100% Flamebait
I used BeOS x86 R3, R4, and R5 with 32M memory without a problem (on a PII 233).
I eventually upgraded to 96M memory. The only time I ever had to constantly swap was when using gcc (gcc? memory hog?), and it's easy to make gcc hit the swap with 96M too.
Each successive release did slow things down a tad and increase memory usage, but BeOS 5 runs fine with 32M.
An open source Be could really help the open source movement get to the desktop. Linux, though getting there, really wasn't designed with desktop use in mind. The mentality is just wrong. Though KDE and friends are very mature and wonderful desktop environments but the majority of the Linux software base is aimed at command line people. Try to get a user to go find and install random Linux software, try to get a user to find help for a problem related to Linux software. They'll get sucked into the "make all install" world quickly and they won't like it.
The Windows world is aimed directly at users with no interest in the lower level stuff, what is what the users like. When I go to Google and type in "linux red font" the first thing that comes up is "Using Truetype Fonts With Redhat 6.0", which starts off with some fstab stuff. That's well and good for you and me, but most users really don't want to have anything to do with that stuff. When I type in "windows red font" I get...ummm..."Quick Font Fixes for Red Hat Linux 6.2"...great. Well you know what I'm getting at.
I've always been a bit dubious of the probability of widespread use of Linux for desktops in business. Linux servers with Be clients, however, I think could really happen.
If the developers of TV and Radio thought this way I'd have to buy mulitple TVs to watch different shows and have different radios for different stations. Initial TV's did only "work" for their single stations but that changed quickly and now every TV in the world is practically the same and can receive the same programming as every other TV. Why can we just pick an OS and move forward rather then come up with a new one every so often because whats out there isn't good enough?
Any OS is just an OS. To geeks, one OS may be far superior for one reason or an other but to users they're all the same, they just look a bit different.
Any time some group decides to develop a new OS, all they are doing is deverting attention away from moving PAST the OS into something else, something NEW.
Don't get me wrong, I love Linux / UNIX. I'm typing this from a FreeBSD workstation right now. I personally love the windowmaker environment and the "UNIX way". I don't believe in wordprocessors (long live LaTeX---or LyX for you neophytes). I read my e-mail with mutt. I use lynx to browse the web most often. I use X to organize my terminals and set their geometry!!!
That all being said, I would not wish my computing lifestyle on anyone.
I'm also a closet BeOS user when I can be. Let me tell you what I like about BeOS.
Some of the other things that be had was a file system that you could do many database style things to. Ripping and organizing mp3s from the standard filesystem and OS features was cake!
Replicants and such were badass. I could imbed a webpage on my desktop with netpositive.
I could go on and on. I loved the system. I love Linux. I love them for vastly different reasons.
I love my Linux brethren because of their idealism, but sometimes they are too interested in ruling the world by exclusionary tactics. Don't assume that alternative OS users', their hacking ability and intellect belong to "the movement".
-Peter
. Penguins Surely Ca
They're not interested in BeOS,
and they're not interested in the name.
What else did BE own?
You were mistaken. Which is odd, since memory shouldn't be a problem for you
Why don't you merge code with the Atheos project, which is already a long way towards a "user" operating system. Some stuff Atheos has done would probably be better than the BeOS code, and some BeOS would be useful to add to Atheos.
David
beos doesnt have the hard real-time capability, but it actually IS written almost all in C... its the API that is in pure C++
"I would say that 99 per cent of what my father has written about his own life is false." - L. Ron Hubbard Jr.
Hang on BeOS users. For OpenSource coders who want to do something cool, but aren't sure what, this is a grand opportunity. These people have a plan, a working prototype, and the experience and skills with BeOS to make it happen.
Remember that BeOS is not Linux. The two will fail or succeed on their own merits.
As for reaching critical mass, well, that is a problem still with linux. I think OpenBeOS could do it though with careful planning and hard work. If not, then we get back our niche OS and best of all it will be Free.
-Peter
. Penguins Surely Ca
I first came across BeOS about six years ago when one of my more 'out-there' colleagues turned up one day with a weird box, lo a BeBox ! I was stunned by it, dual processor goodness, realtime video stream rendering, etc. It urinated from a V. Great height on the Intel boxen of the day, ~200Mhz.
:(
And they screwed it up !
1-2 years ago uploaded a copy of BeOS from some PC mag. Multi-video streams, multi MP3 streams, not a problem. Hell, it even had a free CodeWarrior IDE ! Hmmm, not QUITE enough apps to switch from Wintel, not QUITE *nix underneath
Now, BeOS is dead, long live BeOS ! Let this be a warning, the best & the beautiful do not always make it.
Ladies & Gentlemen, raise your glasses in recognition of our fallen comrades !
They want stability, simplicity, reliabilty & compatibilty. That's it. (unless of course they are into things pedaphilic, or they are a drug dealers, or they take work home with them, or they are paranoid schizos)
W2K & WXP are stable, reliable, compatible but definitly not simple (the average bloke doesn't give a fuck about multiple logons, they just want the computer to auto boot when they turn it on & quickly - gez W2K is a slow booter)
As far as Linux is concerned it has the same problems as W2K (you can go make a cup of coffee while waiting to boot all the way to the K desktop). Plus there's the esoteric Unix filetree
BeOS is simple, stable, reliable, but just lacks compatibility (drivers, apps). Its modified Unix file tree can actually be understood (its behind a 'BeOS' directory & only upgrades from BeInc go there, there's a complete mirror of it in the Home directory, so if some app or driver needs to add stuff to a system directory it goes there automatically during the pkg instalation. So you just install apps in a folder that you can actually call 'apps' in the home directory & you can install drivers in a folder called 'drivers' in the home directory, etc.)
W9X is simple & compatible, just no good as far as relibility & stability are concerned.
I agree, Glock27.
The kernel could be using C++, I understand the reasonings for that. However if you want to use GOOD C++ with Resource Aquisition Is Initialization (RAII) or have any non-trivial constructors, you NEED c++ exception handling, which can throw a big wrench into your special stack frames.
STL would actually be NICE to use within a kernel however!
The C++ API caused tons of stupid problems and workarounds in BeOS. Not only did they have to reserver vtable slots (in the right order!) they had to make sure that any methods that could be inlined were not if they might possibly be changed to non-inline in the future.
The BeOS api was started a long time ago - before the ANSI C++ standard came out. There are a number of design decisions that Be made back then that seemed reasonable then, but are not good now.
For instance, they should have designed their interface classes to contain pImpls - yes it is an extra redirection but it allows for dynamic library compatibility. Also, there should have been less focus on inheriting non-abstract system base classes. All user classes that the system needs to know about should be inherited from a pure abstract system base class. It would also potentially make their BLooper message dispatching system nicer - like libsignal++...
But one BIG problem with BeOS (besides bugs and slow networking) is their use of threads for EVERYTHING, in the language of C++ which officially has no concept of threads. Yes, you can do it but it is like fitting a square peg into a pentagram. Maybe Ada95 would have been a better choice? But that wouldn't be 1337 then I guess.
--jeff
ipv6 is my vpn
Ha,
:)
Weren't you a regular on comp.os.*.advocacy about 5 or 6 years ago? Only i think your tag was Warp 3
Maybe you can bring us up to date on the latest behind the OS/2 story (what's eComStation? Their site is having "index.phtml" failures)
Tips and Tricks for Mozilla
as long as the alleged source tree leak [begroovy.com] doesn't derail the project first.
As a precaution against legal liability, one of the more popular Beshare servers that hosted the tree, for however briefly, was voluntarily shut down.
Where does the school board find them and why do they keep sending them to ME?
and adding to the question...
:)
Is Timur Tabi still an active OS/2 user/advocate? I remember how proud I was the day he used my figshell program and complimented me on it
I think Timur's been playing with linux lately hasn't he?
BeOS could become the platform of choice for kick ass Open Source media software if it were to suddenly be seen as a great desktop OS that is easy to use for the common user, that is ALSO open source.
"You spoony bard!" -Tellah
And the slogan is "It's the OS that comes with it's own cardboard applicator."
Easy installation is the high point. I hear that it's a mess to remove though.
Could you imagine a beowulf cluster of these.
Once again people get in a fluff about what Big Business might be doing to their darlings. It's their own fault. People should stop wasting their time advocating or developing for systems that are not open source.
BeOS was neither revolutionary nor Open Source. You get no points for "slightly improved performance from implementing a few bits of modern research" in the OS game, especially if that implementation is at the mercy of a big fat joke of a company (Be) headed by a bigger fatter smellier joke of a CEO (Gassee).
Oh hooray! A highly multi-threaded OS! How those big 1-CPU and occasional 2-CPU desktop machines will scream!
And that's about what it's selling for.
I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas
What motivates people to use OS/2?????
making something opensource guarentees that it will never die but does NOT guarentee that it lives. for instance look at mozilla. look at openoffice. look at....
linux is the one opensource projects that i say is really 'living' - because people *know* how stable it is and how well it can do as a server.
as far as BE goes, it's good for multimedia or something (please confirm)... ultimately the larger part of the marketshare (i'm using wrong word here if they arent planning on making money as opensource..), they have to be installed at offices. obviously co-operates offices arent about multimedia and the type of 3d-games you can play in your cubicle (*ahem*)... ultimately people DO have to get work done. so as far as BE goes - dont be so sure that it would alive - alive as active alive, not as in just living.
my blog
Windows NT/2000/XP???
And bad management did not kill Be.
ZERO compatibility with EVERYTHING killed Be.
The big reason Be was so fast was because it couldn't do anything.
You 'Let's open source it and that will fix everything' freaks really need to get a handle on things.
This is a reminder of why software needs to be free. BeOS, QNX, MacOS X; these are attractive systems, but when you invest your time in learning them you surrender control to a corporation. You head down a road which can be pinched off into a dead end at any moment. That's why I steer clear of this stuff, no matter how nice it looks.
I'd rather have freedom than a nice GUI. As long as you have freedom, you might make a nice GUI. But without freedom, you can't even hold on to your nice GUI.
I wish these developers the best of luck in creating free software to replace BeOS.
BeOS was about latency, not throughput. When you move the mouse it moves NOW, not in a quarter of a second. Click on a button and it presses/releases IMMEDIATELY in response to what you did, and the new window snaps up NOW. It sucks as a router, but if having the GUI respond when you move it is what you think the system's first priority should be, then it's perfect.
Linux is about throughput and has sucky latency. Process switching is devoutly to be avoided. We won't put graphics in the kernel because that might slow down our packets per second as a router. We won't even apply the low-latency patches that have been floating around for a year and change. (Maybe in 2.5...)
Load a Linux box until it's thrashing and your rat pointer will jump inches at a time. You can't cut and paste text accurately when your mouse pointer jumps five letters a second and a half after you move it. That's how you get great throughput (batch up those transactions and run 'em through in CPU cache. Join together 50 mouse moves into one BIG mouse move!) But it sucks for GUI-ism, and that's what Windows users care about.
Moore's law will take care of half of this. As systems get faster, latency gets better because we deal with stuff as fast as it comes in. Nicing your X server down to -10 helps a bit, because it interrupts other tasks when it wants to do stuff, but that doesn't help applications you spawn and it doesn't do THAT much for the X server either because niceness is just a suggestion. Yet we don't even seem to do THAT as standard in Red Hat...
It's a question of what the system is optimized to DO. Getting good GUI performance from Linux has only been a goal for the past year and change. The KDE guys are trying it. The Gnome guys are trying to make sure the KDE guys aren't the ones to do it because they don't trust the KDE guys' judgement on licensing issues. The XFree86 guys are trying to undo the mistakes of the past 15 years. (That and get 3D acceleration, which is great, but 80 million triangles/second is reality (or the human perceptual threshold) and it won't be THAT long before Moore's law makes your low-end 3D card photorealistic images in realtime. And a hardware generation or three after that, software rendering will be able to do it. It becomes a much less interesting problem then...)
The kernel is fun, but the big block to Linux on the desktop isn't the kernel anymore. It's XFree86 and KDE. That's where effort can be used. But not until we understand the difference between latency and throughput. (Although getting the kernel to have seperate niceness settings for throughput and latency might be a good thing...)
Rob
Comments like these, is why I keep reading /. :)
I am simply amazed at the number of people out there who claim BeOS is so great. I bet most of those people never even tried it.
1. First of all, there is *nothing* revolutionary about the BeOS GUI. It has a few elements from Win, Mac, Nextstep, and various X toolkits. It has scrollbars, radio buttons, checkboxes, listboxes, "OK" buttons, minimize/maximize/close buttons, etc. just like you'd expect and nothing more!
2. Under the hood it had quite a few things going for it, since it could shed all the legacy stuff bogging other systems down. But other systems are now catching up in some areas, like 64-bit journaling filesystems.
3. The BeOS C++ API was relatively small and clean, but an OS kernel in C++? Overkill, way overkill. C++ added complexity where complexity need not be added. Also, the small API was too small - the drawing API was crippled in my humble opinion. Although X has some baggage, it does have a good, full-featured drawing API at least.
4. To all those saying "get rid of X!" you should go do some research. Yes, X isn't the best thing in the world, but most of you self-proclaimed programmers obviously have no clue about the internals of X (or BeOS for that matter). If you did, then you would understand that X isn't so bad. Yes, it would be awesome to get rid of all the X baggage, but for now, high-level libraries do a good job of concealing X's uglyness, while still enabling a reasonable degree of speed and usability.
5. Some day we will be able to drop X. Unfortunately all the projects I know of that are attempting to make this happen are trying to be *too* revolutionary instead of just losing all the legacy stuff. Berlin/Fresco has some awesome ideas but it is a pretty "out there" project as far as windowing systems go. For all you self-proclaimed "Object oriented developers": Fresco was worked on by John Vlissides, one of the authors of Design Patterns. If you don't know what design patterns are then you might want to get informed before proclaiming expertise in OO and window system design.
MP
I let anyone who visits my house (invited by anyone that lives in my house) use my PC (I can't stop them if I'm not there), but I don't want them to have the ability to send emails out in my name, read any file or change any file on the system. There's login passwords to ISP's in plain text in there, and technical reports that should remain confidention, plus lots of role playing stuff that could be taken out of context (not to mention a lot of fantasy art that may make people think I have a thing for tall women wearing little strips of leather and carrying 7 foot long swords).
In the office security is essential, if only to stop pranksters or the disgruntled from changing your files and settings. In the home at least you want to be able to stop people from assuming your identity (or your new girlfriend reading five years of your email outbox and saying tearfully "why didn't you write letters like that to me?").
Back to BeOS - how much effort would it really take to enable SSL and have a secure login screen? I suspect that it wouldn't take a lot to make everthing that comes in either authenticate itself or not be able to do anything outside of the program that asks for the data. As for file permissions, I don't know much about BeOS, but I know that it can support a few different filesystems, so it becomes a case of using one that supports file permissions.
The audio geeks love BeOS for it's low latency.
The multimedia geeks love it for the applications.
The proto-geeks love it for it's ease of use and stability.
It would be very bad to see Palm let it die. BeOS doesn't fit in the organiser market, but tiny PC's like the iPac are a growing market, and BeOS could fit squarely into the market of organiser sized PCs if Palm goes that way.
As much as I hate to say it, it doesn't belong on the desktop. BeOS is a totally rewritten Operating System that has been arround since 1995, Linux is a Unix and Unix has been in existance since 1969, and Windows has been in existance since 1983. BeOS has no real strong-hold here, I have used it and liked it, but countless times did I find myself hunting down drivers for it instead of using the OS. With Linux, 90% of the hardware I have come across had drivers allready and the hardware that I came across while using Windows had virtually 100% compatablity.
:(
BeOS may not die if released in Open-Source, but the problem is that there is allready enough Operating Systems out there and Be went in too quickly and too big.
It may live on in the Palm Market or other applications, but it isn't a server OS and it has no real chance of being a desktop OS. One good strong point is that a fair chunk of PC users detest Macintoshes for no real valid reason, my father is a good example, the BeOS GUI is Macish and therefore would be scounded upon.
Sorry Be.
The "look and feel" is a window manager thing, so there it's a case of picking a theme or W.M. that matches what you are used to. Failing that you can run X without a window manager and run StarOffice, for the whole win* start menu look and feel. Just don't run Enlightenment with maximum eyecandy settings and complain that X is slow - of course it is in that situation: it's a demo of all the possible features the developer could squeeze in. Without spare time and inclination it's not worth it, I know a few people that use win3.1 because it works for them and don't have the time or interest to learn how to use something else.
I use an NT4 box at work for other reasons, but find that I need to run XFree86 in a window to do things on remote machines every now and again.
The only one that can make you use linux on your desktop is your boss, and since you're using java, there's no reason for that. Since there's also *nix boxes on your network, and you can run most (openGL stuff need support on your display) X applications on your desktop, there's no reason forcing you to change your desktop even if you are running *nix only applications.
Yeah that's all we need..
another POS o.s. that can do absolutely nothing.
Worse still: a bunch of POS o.s.'s
that have no h/w support, crappy minimalistic
proggies, buggy api's all of which cobbled
together by a bunch of amatuer programmers.
BEOS IS DEAD, and I don't give a damn.
Maybe now palm inc. will make a REAL portable
that I might actually consider buying?
>>>>>>>>>>>I would love to see an openBe implimentation, because it would be really nice to have an opensource OS geard toward multimedia instead of networking and programming. (Linux is many wonderful things, but it simply not geared for multimedia.) >>>>>>>>>>>>&g t;>>>
ditto on that; BeOS should have been geared to multimedia exclusively. Apple has some hardcore fans in the recording world, mainly because of protools. a free OS that was geared towards visual art/music creation would be a welcome addition. Linux is nice, but i don't see any really high-quality sound apps (e.g., anything by SonicFoundry or Native Instruments) or anything comparable. I've leafed through 'linux music and sound' and have yet to find anything that really compares to the windows (or even apple) apps available. BeOS' death was drivers and supported hardware -- with an open version of BeOS, that wouldn't be as much of a problem, since there would be a more open (no pun intended,) environment for hobbyists to develop their own drivers and apps.
FreeBSD for the impatient.
1. One quick thing that comes to mind is the little sliding tabs on the windows, suprisingly useful... Also you have consistency, which would be revolutionary for the linux desktop.
2. Other systems are catching up, partially. Name a file manager that makes use of filesystem attributes like the BeOS Tracker, or any system that makes comprehensive use of these filesystem features. Additionally you have kernel level node monitors, querying on indexed fs attribute values, and a design that considers low-latency concurrent access of a disk. Beyond the filesystem, the overall design is meant for real-time applications. Most *nix varieties are not, and this becomes a serious limitation in real-time systems.
3. The BeOS kernel is written in C, absolutely no C++ in the kernel or device drivers. The engineers at Be understood what C++ would do to the kernel and hence there isn't even runtime support for C++ in the kernel. As far as the InterfaceKit goes, there have been numerous third-party OSS extensions that fill in many missing gaps, namely a layout manager.
4. X needs to die, bad. It gets the job done, and is very powerful, but has been developed in an incremental manner and provides capabilities that are overkill. There are aspects of X that high-level libraries can't hide, only making matters worse. There is considerable difference between "reasonable" and good or excellent. The BeOS GUI has its limitations, but it far surpasses X in providing a satisfying user experience. Additionally the BeOS programming models tend to be restrictive, and can artifically limit a program's design.
5. Hopefully an X replacement becomes viable soon. AtheOS seems to be a step in the right direction, but it doesn't seem that other systems will see much benefit from it. Trying to do too much is a good way to kill a project, things need to be done a bit more incrementally, for the sake of the implementors and the people who would need to write and use programs for a next-generation GUI.
All-in-all, the BeOS is rightly hyped on certain points ("buzzwords"), but fails to meet the hype of being TheOS. Hopefully some of the more advanced features of the BeOS, and the willingness to try something new will bring benefits to a wider community of computing.
Shanti,
Agni
You do know about the low latency patch for the Linux kernel don't you? It's been out for a long time and is an essential tool in the Linux music/audio community. We're all running patched kernels and we get lower input-output latencies for software synthesis than BeOS does!
Oh oh oh. Tee hee hee. Eric the half a be!
We have some of the best audio tools out there, Apache, X, PostgreSQL, Mozilla, etc etc and the BeOS API is soooo nice and clean to program in.
I wish the OpenBeOS guys the best of luck. They will get a lot of support as BeOS is just too good to let die. cheers
The boys at Redmond could learn a lot from looking at the BeOS application kits.
I've long felt that BeOS would be an a good OS for STB's for interactive Digital TV. It's multimedia friendly, simple to use and a light-weight s/w load, it's Ideal. Combined with Palm's direction toward the remote Internet Appliance market this surprises me. However their loss looks to be be Open Sources gain.
I didn't know that existed, I thought the fastest was a 200MHz. Sure you don't mean a Pentium MMX, those came as 233MHz. :-)
A quick google search gave me this : 150MHz, 180MHz and 200MHz....no 233MHz, don't tell me you dared to overclock that baby!
i am using beos for 3 years as a tiple boot along with bsd and mandrake and i love it,whether palm supports it or not
sam
i am using beos for 3 years as a triple boot along with bsd and mandrake and i love it,whether palm supports it or not.
sam
Hi,
I have two amiga's a 500 and a (1200? 1000?, I forget) I'll ship one out to you if you pay the shipping only. I have other little goodies for 'em to...
britt@mac.com
Regards.
Britt...
To whomever mod'ed this down so as a troll.
FFFFFFFFFFFFFFFFFFFFFFFFUCK you.
I love Linux but... I have my doubts about it on the desktop. Be OS on the other hand seems to be a great desktop OS.
Open Source office of the future Be on the Desktop and Linux servers?
(loosely ripped)
Yes, I'm still active in the OS/2 community. I'm a Linux developer by day, but that doesn't mean that I like Linux. It has some neat things about it, but I still think OS/2 (especially eCS) is far superior as a desktop OS.
Yes, this bigotted profiling has *got* to stop. It's wrong. It's evil.
Every day, thousands of people are stopped, just for using win98. Computer hardware is biased, assuming that win98 is more likely to crash than *nix. "Application" error, it says, as it stops the user for "investigation." "General Protection Fault", as it pulls the application away. Then, as civil libertarians try to investigates, it blue screens, all based on the notion that win98 users are less reliable.
This user profiling must be stopped *now*!
:)
hawk
the average bloke doesn't give a fuck about multiple logons
I have feeling that when this "average bloke" logs on to find his work has been deleted by his 5 year old son, he might start to consider multiple logons.
Also, he might get a bit concerned when a virus completely mangles his system, along with his work, and everyone else's work to boot.
Yeah, I've been active on the OS/2 newsgroups for several years. Not c.o.o.a as much (way too much petty bickering for my tastes - even more than here).
:-)
eComStation is a fledgling attempt to produce an alternative client OS by Serenity Systems.
Currently it's based on OS/2 Warp 4, and they've cut a deal with IBM to get the latest fixes and such for the system. They've also managed to bundle a bunch of software with the OS, including StarOffice 5.1a, Lotus SmartSuite, the HobLink X11 server, and some other stuff.
Part of the technology included is something called WiseManager which supports some sort of remote client bootstrapping from a centralized boot server. I don't understand it.
Another part includes LVM (the Logical Volume manager) which was part of IBM's Warp Server, and that lets you create logical disk partitions that span multiple disk partitions and multiple disks, and it uses IBM's JFS journaling filesystem.
I think it's an interesting attempt, and I have a copy of the initial eCS 1.0 release that I plan on playing with in the next month or so. As I learn more, I'll probably make more comments about it here and there...
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
Yeah, Timur's still around. :-) He and I both use Linux as well as Warp, but I think I like it (Linux) a lot more than he does. :-)
I will note, though, that Linux hasn't replaced my OS/2 setup yet. Probably won't for a couple of years yet. But I use it as a secondary desktop and on a couple of servers on my LAN.
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
You said C++ wasn't appropriate for a kernel, I pointed out that the kernel was in C. Second, there are hacks in other (C) OSs too. For example, paremeter passing to the kernel is always kinda ugly. Hacks are just a part of life in the kernel. Also, the C++ issues exist in Linux too. KDE/Qt gets around the FBC problem by breaking binary compatibility every major release. Given the smaller BeOS software base, the reserved function "hack" really isn't that bad.
A deep unwavering belief is a sure sign you're missing something...
Geez, stop complaining about threads! There is nothing wrong with pervasive multi-threading. Given the crappy responsiveness of the Linux GUIs, Linux people are in no position to talk. A lossy messaging system (as in BeOS) is an implementation problem, not something wrong with threading. Just learn how to do proper locking and everything will be rosey. The kernel people have to deal with it all the time (SMP locks), its not that HARD.
A deep unwavering belief is a sure sign you're missing something...
Why do O.S. designers have to be such bigots?
All they had to do was make their device driver model accept Windows device drivers. But nooooooo. They were better than that. So, now, they are dead.
Lack of device drivers killed OS/2 and that had IBM backing it. What made BeOS think they could ever compete with Windows for the attention of H/W vendors?
Ok, so Windows D.D. suck. Who cares? As long as they work, use them. You will get instant market share.
The next problem is to get an office package as good as MS Office...
I use threads all the time! I just don't like them to be forced. However I'd like them better in a language that officially SUPPORTS them. C++ is not that language, unfortunately - think thread safe std::string, STL and exceptions, race conditions in singletons without a double guard (therefore the inability to have non-trivial constructors in static members of a class without wrapping them in a thread-safe dynamic singleton allocator object), debuggers, core dumps, profiling tools, memory leak checker tools, lack of warnings when using shared data without a semaphore, etc..
I've been using BeOS for ages and actually had a PPC BeBox too! Still have BeOS installed (which kernel panics when my real Roland 100% standard MPU-401 MIDI interface receives a MIDI byte).
Worked with www.lcsaudio.com where they are STILL SHIPPING commercial software for BeOS. At one point LCS had more BeBoxes than any other company besides Metrowerks. I used to be an enthusiast.
Not anymore.
I can't count how many BeOS applications are written with threading problems! Is it because most programmers actually suck?
Yes, most linux GUI's responsiveness sucks. No, having a seperate thread for each window would NOT always help - Most X apps only have one window anyways. However, having seperate non-gui worker threads WOULD help and should be done when appropriate.
--jeff
ipv6 is my vpn
I'd love to use Linux again. But after using BeOS for a few years, I've grown used to a few too many luxuries.
/boot partition to a CD-R (great for system backups, and sharing the experience)
I'd love a Linux Distrlibution which, out of the Box does the following:
1. Has Metadata Attributes in the Filesystem, which can be easily accessable by the File Manager.
2. Has a pretty, consistant and responsive GUI.
3. can mount Different Sessions from Different CD's graphically
4. mounts Audio CD's just like any other filesystem
5. works with my Video Capture Card straight out of the box.
6. can easily be replicated by burning an img of the bootdisk and img of the
If someone made a Linux Distribution which did these things, I'd be ecstatic. It doesn't need fancy things like Blistering Audio Performance, or the ability to turn a processor off on an SMP system, or even the ability to play 5 DivX Videos at once on a K6-3/550 without skipping. Just a nice, clean, logical GUI.
The thing with BeOS threads is that they force you to seperate display logic from the actual program. This allows windows to redraw quickly without waiting for the application to catch up to it. This is a problem in many Linux GUI apps. Also, by seperating the display logic out, it takes away many of the pains of multithreading. The windowing thread shouldn't be doing anything super-complex, just displaying precomputed data. As such, multithreading shouldn't really make designing programs that much harder.
A deep unwavering belief is a sure sign you're missing something...
Well, I agree with you.
--jeff
ipv6 is my vpn
Plus there's the esoteric Unix filetree
Hmm?
The Unix file tree makes a lot of sense to me.. a lot more than the silly meaningless Windows file hierarchy.
Anyhow, what do you need to read from the tree directly for anyhow?
All your stuff use your own organization under $HOME in *nix, whereas its Windows where you have to browse and navigate throught its large, meaningless file hierarchy.
OS/2 was a 32-bit smoothly multitasking operating system in 1992 when Microsoft was offering little more than a cooperatively multitasking shell on DOS. Compared to Windows 3.1, even OS/2 2.1 rocked, and Warp 3 and 4 have added polish over the years.
:-)
:-)
Not only that, but OS/2 came with a GUI that still hasn't been equaled in many respects (the WPS), a scripting language that programs could tap into (REXX), and support for DOS and Windows programs that was both robust and flexible.
*That's* why I moved to it. Alternatives like BeOS didn't exist yet, remember, and Linux was little more than a kernel and a collection of text-based utilities, not the useful OS it is now. Xfree86 worked on VERY few cards, and there wasn't a lot of X-based software worth using, IMO.
Most of the folks who still use OS/2 are older hobbyists like myself who moved to it back in its heyday. It's a much harder sell now.
Why do I still use it?
I use it because I have OS/2-native tools like Colorworks, Embellish, and Web Organizer that I really like to use and that I've already paid for, because it has modern-enough internet clients like NFTP, links, Netscape Navigator, and slrn for me to get by, and because there is still enough in the way of software development going on to keep my interest going in the platform.
Also, I've already paid for my two OS/2 copies, so the relatively high price tag isn't a factor any longer.
Does that help?
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
What if a whole lot of people pooled some cash and bought the BeOS off of Palm and gave it to the Open BeOS and Blue OS Projects.
:)
Palm only paid $11million and the part we want is the part they don't want.
Just a thought
BeOS FOREVER!!!
You're right. ;)
it was in a computer lab. Half the machines were 200 P-pros, the other half were 233 MMX... I had to go open the case on that one to see what it's CPU was
Not to be picky, but it's based on Warp Server for e-Business (Warp 4.5) which is why it includes the LVM and JFS. SMP is also added as it is part of the server. SMP doesn't work in Warp 4 unless you add parts from the server, so the easiest route was to do server with or without SMP, as you can select it in WSeB's install.
Adam McNutt
avid young OS/2 user
BeOS users always had ability to write own drivers and do these things very modular os, problem was that everyone was playing waiting game, waiting for Be to update these things like microsoft does, same with just about every kind of app you can think of. Well guess what, now there are no excuses, Be no longer holding userbase back I expect lots of good things, and already have seen many...