This document describes the feature of the BeOS kernel to monitor nodes. First, there is an explanation of what kind of functionality OpenBeOS have to reproduce (along with the higher level API), then the author (Axel Dörfler) presents the implementation in OpenBeOS.
Node Monitoring in BeOS/OpenBeOS
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
Axel Dörfler is one of the most impressive programmers I have heard about in recent history. He seems to be singlehandedly implementing and documenting the majority of OpenBeOS.
Yup, I agree. It seems that OBOS is mostly the work of two-three people who know what they are doing and the rest are just larking or at very best, don’t know how to proceed. In the second team of people belongs that guy who does the app_server (that guy needs to be replaced by someone who have a clue about graphics server programming).
Alex is indeed one of the very few people who do work on OpenBeOS, and who do know what they are doing.
I agree Axel Dorfler seems like a very talented man. But when can we have an iso to download?
>But when can we have an iso to download?
When hell freezes over. Come on who really believes that OBOS will ever get anywhere. Axel might be a talented programmer but for creating a real useable desktop OS you need SOMETHING more than a few talented programmers.
Following to their status page all tricky parts of the OS (really almost all parts of the OS) are still in early pre-alpha (reads: they have almost nothing) stage.
BTW I just decided to start the development of an MacOS X clone for x86. I have decided on a name to use “H20 OS” so I am in pre-alpha stage already *ROFL*
Unfortunately, I will have to agree with Max. OBOS is nowhere near to deliver anything usable today.
I believe that Axel and the rest talented programmers should join YellowTAB and work on existing, usable code, instead of trying to re-invent the wheel. If these programers are doing OBOS for fun and for the experience of it, then sure, OBOS *is* a great way of spending time in your afternoons, but if you are seriously interested in providing a future to the BeOS platform, then YellowTAB has a product that works NOW and they will need additional help TODAY.
BeOS is already behind the times, as Be hasn’t worked on it for years, it doesn’t support latest stuff (firewire, USB2, bluetooth, ata133, latest ati/nvidia, full 3D on modern hardware etc etc). If these programmers really want to make a difference, they should join YellowTAB and work on adding these features and bringing Zeta (essentially BeOS 6) up to speed, instead of re-inventing the wheel and start from scratch. It took Be 10 years to bring BeOS to the level it was in March 2000. It will take another 10 years to these 5 OBOS devs to bring OBOS into the same level (they are less devs working on it, and they don’t work full time on it). So, put up your work at YTab, IMHO.
BTW, this discussion doesn’t really have much to do with BeOS’ node monitoring.
I disagree Eugina, I think that OpenBeOS is helpful for yellowtab because it can liberate them from their licencing restrictions from the Be source.
When? In 10 years? It is useful only if they had this solution ready, already, or in the next 1-2 years. Unfortunately, OpenBeOS is way far still from such “liberation” or completion.
And what licencing restrictions from the Be source are you speaking about? Even if in 5 years YTab ship OBOS instead of BeOS, they would have already ship BeOS. And if PalmSource would want, they would sue them by then anyway. In other words, OpenBeOS is only helpful here if they ADD new features to Zeta, NOT by replacing parts that are already there and YTab already uses (which also create potential bugs in their reimplementations).
I disagree Eugina, I think that OpenBeOS is helpful for yellowtab because it can liberate them from their licencing restrictions from the Be source.
Uh, how can it helps YellowTab when it’s nothing finish? If they join to YellowTab and they will get the things done much more quickly, because they are going to get the pay and more hours of programming. Also, include the more promise than the non-promise project of OpenBeOS.
What I mean with the above is that the “replacement” of the “fixing” of the low level stuff, should be handled by YellowTAB themselves, as they have the source code. If OpenBeOS wants to really help BeOS, they should transform their project into a development group in ASSISTANCE of YellowTAB (*or* whoever CAN use/sell/distribute the original BeOS code (which is here and works)), instead of trying to re-invent the wheel and redo what YTAB already has here today and already works.
In effect, they should be what GNU is for the Linux kernel. Without GNU, Linux would go nowhere. GNU provided the NEEDED applications and some low level software (ld, gcc etc), and OpenBeOS should do the same for whoever wants to offer a BeOS that works *today*. When YTAB finally ships Zeta, they will have the hardware support problems I mentioned above, *plus* the application base problem.
If OpenBeOS is here to HELP YTAB, as Wing said above, then OpenBeOS should change its role in regards what it has to offer. It should provide the needed apps for Zeta to survive (plus low level services when possible, e.g. the new BFS filesystem Axel wrote). Pretty much do what the *original* BeUnited (that myself and Deej founded in March 2000), which had as a goal to port/develop big apps that would push the BeOS forward as they were apps that were needed in the platform. The original BeUnited goal failed miserably, because as in the OpenBeOS case, from the 120-130 developers who actually signed up, only 5 or 6 actually sat their a$$ down to do what they promised to do when they signed up in their respective development teams: develop.
People critisized me for… critising OpenBeOS. But I’ve been there with BeUnited, I know what is like. It is very hard to get people excited enough to actually do develop stuff. It is already hard in the Linux platform, which is currently flourishing, but on a half-dead platforms like BeOS, is just immensely difficult to have these individuals work for the BeOS. This is why my expectations are too low for any such project (and this also includes Syllable, not just OBOS – nothing personal, just observations based on personal experience).
Didn’t somebody from YellowTab in some interview (yeah I know it’s vague) say that they were planing on using pieces of OBOS as they became avalible. Maybe YellowTab should hire these developers, but not have them directly working on Zeta. Instead of creating an exact clone from scratch, they could make and improved version of each part, and that could be folded back into Zeta. They could eventually replace all the old BeOS a little quicker than it will take for OBOS to be finished all by itself. ot only that but if they are replacing the parts with enhanced parts Zeta is moving forward, so it’s not so far behind when it’s done.
> Instead of creating an exact clone from scratch, they could make and improved version of each part, and that could be folded back into Zeta.
There is a problem in this.
Creating something from scratch (with questionable “improvements” as you should not forget that binary compatibility should be kept), is 10 times more time consuming than YellowTAB working on directly improving the existing parts. There is no reason why you would need to re-implement the whole kernel AND keep binary compatibility! This is a huge task! YellowTAB themselves should be working on improving the current one, and OpenBeOS should be working on third party ports and applications or parts that don’t require you to have the BeOS source code to move on. This way, the BeOS lives on, while having two teams each one fighting for the same cause, is waste of time in today’s extremely demanding OS market. People are _expecting_ their drupa-coupla DVD-R/CD-RW/EUGENIA-RW device to work. Five years ago, they didn’t. Heck, 10 years ago, noone cared about networking stacks either! But today, writing an OS is extremely demanding. There are way too many modern stuff to be supported, and if they are not supported, you only get people being “unhappy” that their favorite hardware is not supported, or that this or that application is not existant. And having one such person in every forum saying “Zeta or OpenBeOS sucks. It didn’t boot. Or it didn’t support my PSD-2005 image format or didn’t support web cameras on my Yahoo! IM or MSN IM.” it is enough to drive away others.
We are living in very demanding times and OS manufacturers are strugling to support all that cr@p. Even MS had trouble with WinXP. And Zeta/OBOS is destined to work in the jungle of x86 (Apple is indeed in a more favorable position in this big issue, as it controls most of its hardware).
This is why I am saying that OBOS should become what GNU is for Linux. Sure, GNU tries to write its own OS, HURD, but after 10 years they are still NOWHERE near even a beta. But GNU showed real strength in other areas, and helped the Linux kernel! OpenBeOS should do the same for whoever has the rights on distributing BeOS (in this case, YTAB). There is no other way for mutual survival. Help each other and [barely] live, or stay apart and see the BeOS fade away with time.
I wouldn’t invest a single minute into YellowTABs afforts.
They have no solid business strategy at all and displayed their utter unprofessionalism all to clearly in the past.
If they ever manage to release that “Zeta” thingy (I doubt it) they will be out-of-business a few months later.
It’s no suprise that out-of-all Linux companies only RedHat has half-way solid finances. They are professional, they have a solid business strategy, they are sane. Trying to compete against Microsoft on the desktop market is like declaring war on the US. You can only lose.
RedHat knows that Linux sucks on the desktop. They make their business by selling support contracts for a unixish server/cooperate OS. That’s a solid business model. It worked for decades and for dozens of cooperations. And it still works today.
I think it is sad that MS has a de facto monopoly on the desktop but the market isn’t fair. The only chance for an alternative OS to survive is the non-profit way of Linux.
>They have no solid business strategy
>It’s no suprise that out-of-all Linux companies only RedHat has half-way solid finances
Believe it or not, I agree with you. Of course and all these small OS companies are going nowhere in getting any significant market share, let alone surviving in the long run.
But this is irrelevant in the frame of this discussion.
You see, we are talking about people who do want to work on BeOS stuff in their free time. And YellowTAB is the ONLY company today that does have the right to distribute BeOS and it does have the original source code. So, if the OBOS people do want to do something productive with this desire of theirs, they should work for YTAB, as YTAB is the only BeOS provider today that has something that WORKS.
It is irrelevant if YTAB is going to survive in the long run or not. XFCE or Fluxbox will never overturn Gnome or KDE either. But people do want to work on it, so in this frame of mind, OBOS should invest in the YTAB business OR *whoever* can distribute the original BeOS. It is much better than wasting years and years on trying to re-invent the wheel and at the end ending up worse than they would ever thought. At least, supporting YTAB, it would give this community 1 or 2 or 5 years of life. If it works, and YTAB makes it surviving, great! If it doesn’t, that’s ok as well. We can at least live with the memories of 2 or more years of a supported BeOS. And that worths it.
I half agree about pooling OBOS/YT/B.E.O.S etc.
One issue for OBOS+YT is the source code, if any OBOS developer sees it, they can’t work on OBOS can they? So any help they give will have to be at a distance, which is the way it was going to be anyway as new parts became available.
If the OBOS devs were to work on such things as FW, Usb2 etc that would sap any energy to do OBOS. Tough call.
at least they would hav a beos like kernel
an something to work with instead right now all open beos has is a kernel that is not even in alpha….
i personally think beos should join up atheos an syllable
at least they would have something
Whoa. I wasn’t aware that YellowTab had distribution rights to the BeOS source code. I’ve been out of the BeOS loop lately, so maybe I missed something. If they do have the source code, that’s really cool. I might have to take a closer look at YellowTab then.
Yes, YellowTAB has the BeOS source code.
>i personally think beos should join up atheos an syllable
Not really. AtheOS is not BeOS, these are two different OSes and cultures, joining them won’t bring anything new.
I think that Eugenia does have a good point in pooling their efforts together for the common good, when starting in the possision they are in now its not only sound buisness advice, its almost a necesety if you want to survive to release a second piece of software.
On hardware compatability… TBH unless you have M$ cash and resources you will NEVER support everything. So picking out the most suitable and easy to support hardware to me again makes good sence. So I won’t get 3d Hw-OGL with my Geforce… big deal, I can get it with a radion and on the bang per buck scale they are better value. I don’t care if I’m limited to 15-20 gfx cards past an present, so long as they work as they are supposed to. Quite a few motherboards out there work without too much hassle, and that includes budget boards for those of us who are cash starved… A very strong point in their favour (although AC97 sound STILL isn’t as good with BeOS as it is with Windows… never mind..)
Will it be the next big thing? Maybe, maybe not. No one can say. 20 years or so ago, could anyone here say that when Bill Gates set up his company MSN that it would dominate the market in two decades and become a power unto itself? I doubt it. If nothing else I would hope that Zeta inspires people that there are other ways. Just like linux is doing, only with a much more friendly interface and easy to understand way of doing things.
If it collapses, so what? I don’t know about you guys and gals, but I sure as hell am enjoying this ride, no matter how short it may turn out.
Just think, how big did Be inc get when it went X86 and software only? If they had not pulled the crap they did re the whole stb debacle, BeOS would have been a serious threat (which M$ actually acknolledged in several leaked emails and so forth). Personally I hope that Zeta becomes the next IT success story, I have reservations, but I do hope. Marketting is the key to all of this, when did someone last try selling an OS specifically to the home user, as a genuine home computer solution? Windows is ubiquitous, of that theres no doubt, but its massive and with XP, restrictive. Zeta could make a bundle if they target this right.
you can’t have it all. it will not happen anytime soon, for a community-spnsored project like openbeos. however, i think it’s not even that important: beos/openbeos never really had many users, and the one who still like to work in beos are hardcore enough to use supported hardware. If someone offered me the alternative between openbeos with little hardware support, or no openbeos at all, I would certainly take openbeos with what little hardware it supports. Let’s be honest, BeOS with old ATA transfers still gives better user experience than Linux with ATA100.
BTW, I am installing BeOS on a brand new computer this morning. My new MIDI station!!
I know very well that YT has Be source code … but my question is … how much? Did they license portions or the entire deal? I’m a bit confused as to why Zeta won’t be fixing a number of fairly large bugs (the >1GB RAM problem, for example).
BeOS does have a functional ATA100 driver… I’m using it as I type this hehe. One of the other people complained that BeOS didn’t support ATA133, well, all I can say is, beyond one brand of hard disk manufaturer (the people who came up with ATA133) no one else actually supports it. Show me any motherboard out there that is made not by the creater of the standard that has this? (and some promise ATA pci cards) I doubt you’ll find one, its ATA100, or in the latest boards, this new Serial ATA standard. In real terms even ATA100 isn’t fully utilised as IDE hard drives poop out at around 20Mb’s (on average) per seccond, ATA 100 spec offers four times that amount of potential speed. So what would ATA133 support in real terms give us? Extra bandwidth that still isn’t used, big deal…
>I know very well that YT has Be source code … but my question is … how much?
AFAIK, as much as it is needed.
>Did they license portions or the entire deal?
I won’t comment on this.
> I’m a bit confused as to why Zeta won’t be fixing a number of fairly large bugs (the >1GB RAM problem, for example).
Because this is not an easy thing to fix. And it is not a bug. It was a design decision, many years ago. I talked to my husband about this, who knows about the actual problem, and he says that this is one of the biggest hard problems to fix on BeOS today, it is just very difficult. “Fixing” it, it would mean that many apps that make assumptions about the layout of memory won’t work anymore. And other limitations, like the 32 MB of addons per app, it will be divided by two by making the limit up to 2 GB RAM. The point is, that when these decisions were made, it was 1995 and people didn’t have more than 8 MB of RAM, it was making sense back then to have these limitations, in order to optimize stuff better. OS/2 had similar problems too btw. All older OSes that didn’t continued to get updated by its creators do.
That’s horrible idea.
YT in my eyes now looks like a bunch of small hackers with totally unproffessional leader who can’t stand critics, lies to everyone around
and has ego bigger than Mount Everest.
If they ever release Zeta it will be just a little “debuged” official Dano. Nothing more. Additional apps are nothing special. Drivers are already on BeBits. Generally all really worth things were written by others, not YT.
And OBOS idea is to be opensourced which is totally not the YT way.
It’s their (OBOS devs) decision, not ours anyway. Their life, they can do whatever they want with it
But if they want to also make all of us happy, yeah they could change app_server guy. Maybe join Syllable, or (if we want something really new and really close or even further to BeOS modularity) Unununium.
I really am not sure what way would be best, but i’m sure of one: joining YT is not one of the good ways (for us, and in the end probably for devs too – they would have hands tied later if they wanted to make some opensource thingie similar to os).
Back to topic:
Axel’s idea for changes in mode monitoring is really good. It would make much easier life for developers and much cleaner apps code
Shard, do you know what you embody in my mind in that comment of yours?
The Amiga community and the way it is forked into the AmigaOS and the MorphOS community. Each community throws fireballs against the other. And you know what they have accomplished with it? Nothing. They are the laugh of all the people who are not part of these two communities and they are watching them from “outside”.
I highly suggest you be careful what fireballs you throw, trust me, you don’t want to see the BeOS community in another all time low. Unite and see a possible future. Fork and BeOS will die a slow death, as it already does. Think about it.
Having spoken to Bernd at length I can honestly say that he is a good guy, passionate about BeOS too. So he doesn’t like critisim… not many do, me included.
Bunch of Hackers? Fair enough thats your point of view, I’ve not looked into this in great depth. “just a bugfixed Dano” Have you ever used Dano? I have, its a damned big improvement on R5.
YT might not be the best people to do this, but they are there and willing to give it a shot. Not many companies would do the same (palm suddenly comes to mind). Give Zeta a chance, you never know, it might actually work!!!
Agreed RE the whole MorphOS AmigaOS thang. It is a complete sham. And it reduces the Amiga community numbers by the bucket load with each new statement from the parent companies involved. The whole community is a laughing stock compared to what it once was, which was the Amiga’s best asset…
Well that was my opinion. And to say the truth BeOS comunity is already splitting.
I agree it would be better to keep joined effort, but not under YT’s leadership. Maybe it sounds funny for someone or maybe not, but that’s my opinion.
You’re comparing BeOS community to Amiga’s and i think it’s very similar. I wasn’t in Amiga community than, i just read few things about what happened, but look how familiar it seems. Here comes company, fans have their hopes high, give money to “santa” and than… they see they were cheated. Amiga fans were cheated many times, by few companies. Do we want the same, or maybe we can learn from their mistakes?
I know Aros is developed for a long time now, and still not complete (although it looks quite close now), but it will be there after another few companies will harvest puppets ;] And will be progressing. And fans can see what’s happening, and even help.
If i’ll ever give money for some software again it will be only donation for opensource project (and only if i see some progress first).
well.. maybe i’ll buy halflife 2 and thief 3 ;]
Betas are being produced, products that appear close to completion are doing the rounds at trade shows. They have got quite a bit done. OBOS looks good but beyond a few people who actually get off their arses (like axel) nothing is actually done. As Eugenia said, its about 4 years before we’ll be seeing anything like a complete OS (I suspect a little longer having seen how long its taken AROS to get where it is, but there you go.)
YT may not be your cup of tea, but give Zeta a chance to prove itself, if nothing else it puts BeOS back into the public sight which can only be a good thing in the long term. OBOS may be the way to go in future, but thats no good if you have to wait around for X amount of years for anything to get done. Zeta will be out this year (with any luck). Lets see how it actually performs before people dismiss it.
If you don’t want to pay for software then thats your choice, but to dismiss an OS purely on that basis is a complete waste of everyone’s time on this board as it doesn’t prove anything, except your being a skinflint (read thrifty in the bad sence) that is….
>I agree it would be better to keep joined effort, but not under YT’s leadership.
Sorry, but you have no choice. YT is the only people who have the original source code, who have something that works right now. There is no other choice, if you like or dislike Bernd, is completely irrelevant at this point. OBOS is in no position to deliver anything, and Palmsource doesn’t care not even 0.1%.
I don’t think Zeta and OBOS (will) fork the community. They have very simmilar goals (design, binary compatibility,…).
IMHO the “problem” with those BeOS-replacement OSes is BlueEyedOS. It seems to me that the BlueEyedOS people do no longer have the same philosophie about operatin systems as the BeOS (OBOS, Zeta) people.
It’s not about my money, it’s about who will get it and for what.
Sure, maybe i am wrong and YT will proove to be a good company releasing few next BeOS/Zeta versions and moving BeOS forward.
I just doubt it.
BeOS works for me ok. THX to great guys writing drivers i can upgrade my soundcard and gfx card if i need/want to. I can wait. I don’t have to buy something just because it looks new.
Sure, BeOS lacks here and there, and i would like to have updated net_server now, but it’s not critical. And not for the price of freedom (i’m not talking about money now). I can resign from part of freedom only for someone i trust – i don’t trust YT.
I don’t see BlueEyedOS as a problem. B.E.O.S is NOT a direct BeOS clone, it is what GNUstep was for OpenSTEP: a re-implementation of the toolkit under Unix. Nothing more, nothing less.
I can see your point, if you don’t trust them that that is end of argument as far as I can see. I can’t force you to trust them any more than you can force me to distrust them…
If BeOS works for you, then thats great, I know it works for me also.
I think that only time will tell who is right on this one. Quite frankly, I, like a great deal of others in the community don’t want you to be. And I suspect if you think about it, neither do you.
(argh.. i had quite long post and than hit CTRL+R instead of SHIFT+R… ;(
I have a choice. I don’t have to jump to YT just because they may have source.
We don’t need strict leader. We need only some standard API to have apps running on all BeOS-like OSes. Something like POSIX but for BeOS and GUI
There is BeUnited, but i can see it’s too weak to actually keep standards (at least for now). If people will want to keep BeOS standard and can talk to each other without acting like fanatics, we can keep basic API with some platfrom specifics similar way to *BSD world.
I was not talking about the technical side of BlueEyedOS, I was refering to the communities.
Some people will say “Oh, it looks like BeOS and it is even source compatible. The hardware support is also higher”, while other people say “BlueEyedOS sucks. It’s betraying the philosophy of BeOS. It has a monolythic kernel and is soooo unlike BeOS.”
This is what I meant.
OBOS and Zeta communities are IMHO not forked. Many people say “I’ll use Zeta until OBOS is ready. Once OBOS is ready I’ll use what’s better supported.”
First of all Axel rules pretty much everything in the world. When the Linux kernel hackers pray to their DevGod, they in fact pray to Axel….
OBOS isn’t dead, OBOS isn’t Stupid. OBOS is what it’s all about. There is a reason to why they want binary compatibility. They also sort out a lot of bugs that existed previously in BeOS. They add more Posix stuff. They do a lot of things which look promising for the future.
The rant in this discussion about Linux blaha and only hope etc etc… that’s just depressed unhopefull Zeals talking. Sure there’s hope. Plenty of it.
For those who don’t remember or haven’t read why people have left the BeOS platform, I can give you a hint. It’s NOT because BeOS isn’t the best desktop OS out there, because most people agree it is. The fact they left is because BeOS didn’t have a promising future.
I promise you, the very day that OBOS release something (eta 6months?) then there will be a huge bandwidth problem, because I tell ya, at least 100 000 people will get that ISO in less than 2 days.
Keep up the good work Axel and co!
thanks for keeping it real
Why to poke OpenBEOS team for not working on YT?
Poke Bernd. If YT needs qualified worforce, it should hire it. Point.
Or do you propose real OpenBEOS developers work for YT for free? Just because they love BeOS and must support any BeOS revival effort? Strange idea, just contradicts OpenBeOS personal ideology. They started OBOS because it is quite clear in today’s desktop monopoly, that only open-source OS may be kinda immortal.
Eugenia it’s something to post is opinion but you act plainly evil right now with no reason at all and especially not as “informative” purpose.
I can be really evil myself http://www.bebits.com/app/2136 don’t ever tell me bogus thing ever again about what should be done for BeOS and what not. That simple port represent you more than text.
Some like Shard say YT will do nothing good, perhaps, it’s is opinion and i generaly respect him a lot and he say is opinion without having to centrate is life around that hate like you do.
As for now YT may be a bunch of hacker (i’m not their QA department so i don’t know) but at least they did not released so far rushed stuff like M$ and even apple do most of the time. Ho yeah the guy that bring me my journal each saturday is just a journal guy after all, glad to read osnews i was forgetting to bash him about is meaningless life for all those years! Beware journal guy it’s going to be your worst weekend ever!
Including myself, i really know loads of hardware-geeks having all the latest and greatest and we walk into the shop every other day to buy stuff we don’t really need. However, I have never got to know ANYBODY with more than 1 GB RAM. So who are we talking about here..? Fat servers need more than 1 GB, the averge Net+ surfing BeOS-guy probably doesn’t need more than 32 MB (hehe.. 😉 ) — So the 1 GB limit is rather a non-issue.
The 1 GB limit is certainly an issue. I used to have 640 Megs of RAM in my machine. When I threw a 128 MB GeForce3 into my machine, BeOS would stop booting, forcing me to downgrade to 512 Megs of RAM.
Now, if Zeta doesn’t fix this bug, I’m extremely unlikely to purchase it. I have another 512 megs of RAM that I can throw into this machine (from another computer that died), and it *really* boosts performance.
Adam
hoping for a self-fulfilling prophecy? Just because you were convinced OpenBeOS won’t make it, it’s no reason that you can’t change your mind later. You remind me of that middle-east analyst that used to be guest at BBC World news, who was convinced the US troops will be cought up in street fighting and will have to pull out of Iraq because of huge losses. He kept that line up until they took the Baghdad airport. Then he just disappeared from the TV.
Which is good.
Of course, operating systems are difficult to assess. For many people (me included) BeOS as it is now, is a very useful and interesting OS, which is a pleasure to use. For others, it’s outdated and unsupported. For those, even when OpenBeOS was to get 1.0 release, it won’t be good and all the work the OpenBeOS developers invested in it, will be in vain.
I think YT will definitly help bring interest back to the BeOS front. It already has somewhat, even before anything has been released. geez.. there are so many pesimists that nothing would EVER progress if anyone actually listened to them. Lot of progress has been made in OBOS. What the hell is this article about ?? Something that has been learned, implemented, and shared. If we all listen to the farking pesimists, we would all just go out and buy a new copy of XP, and even small steps would not be made . <sarcasim> Hell… why even take a computer operating systems class in college anymore??… theres NO way that it would ever be useful since MS just about owns the computer now. What good would it do? </sarcasim>
OBOS will go slow; it will take time. It will never keep up with a multi-billion dollar corporation with full time programers. These are people that enjoy computers, and enjoy exploring the path not taken.
I’m not going to judge YT’s work until AFTER I see it and try it; but people that have say good things about it.
…. FAM or dnotify. I’m not sure what’s so great about this.
Shard, you have a big ego too, so what? It’s dorks like you who drive away the geniuses like David Reid. Were you actually a programmer at the level of guys like he and Axel, you would not make such pathetic comments.
This thing, OpenBeos should help Yellowtab is silly since yellowtab plans to use openbeos code.
I’ve read in their forums that they are not planning to update the mediakit but wait for the openbeos-media kit to get finished, and use it instead! And I guess the same goes for other parts of Openbeos.
Yellowtab isn’t really a stable company, well no one can tell it is, or it is not because it haven’t got any incomst yet.
If the Openbeos guys would help yellowtab, the code would belong to yellowtab. Then if yellowtab would go bankrupcy, that would be the end again.
Yellowtab isn’t Be. Face it, they are hobbiest programmers, I bet skilled ones. But since they have no public market plan or no investigations, and no other staff than programmers (?) giving it a bet now would not be very smart in my opinion.
1) OpenBeOS won’t do third party applications for BeOS, it also won’t join YT. We have started doing a whole OS and that’s what we will deliver – certainly in less than 10 years 🙂
2) I am fine with Zeta – I wish Bernd and YT great success, and I am sure we’ll be working together closely in the future – for the good of both. We won’t work on the closed BeOS sources though, we don’t want to be dependent on any commercial company ever again. I hope Zeta will “reanimate” the BeOS vibe again, and we will profit a lot from it when we’ll start to deliver the best BeOS ever created.
3) Thanks for all the fish, but I am certainly not the best programmer ever :-). Furthermore, there are lots of people working on the project (and also, for example, NewOS) that I have the highest respect for – more than just two or three.
4) Shard may not like Zeta or the level of professionalism Bernd has shown in the past (and may show in the future as well :-)), but he’s still a very nice guy, even if he expresses he thoughts in such a direct way. He also had nothing to do with David leaving OpenBeOS 🙂
5) Why shouldn’t Eugenia think we’ll take 10 years to complete OpenBeOS – have we proven her wrong yet? I mean, we certainly will, but is that her fault? I have enjoyed reading tons of her news articles in the past, and I hope, she’ll continue to deliver.
OpenBeOS is a huge task, and does take a long time to accomplish it. But I am sure the results will justify the efforts needed to create them.
Yeah i have a big ego too – still i don’t lie to people, and i can stand criticism usually (that depends on how someone criticize also, and even if i’m in rage first ;] i can think about case later and take into account other’s opinions).
I’m surely far away from being as good as Axel, but i think i can make my comments (even if they’re pathetic in Your opinion).
Even if David Reid is so good coder, he’s not so good in working with others (i’m saying that after reading mailing list archives). That’s why he left probably. Not because of some dork like me ;P
BTW what can one person do: MenuetOS, AtheOS, TraingleOS.
Probably that’s why parts on which Axel (not only, other parts like media too) are progressing faster than others – one person working on something mainly alone, without interruptions and laziness, can do much more than a group of people.
Sooner or later Axel and those few others will finish OBOS
Eugenia is 100% correct in this matter of OpenBeOS and Zeta.
If the effort for BeOS is a unified effort, a strong platform can be established.
As the community has contributed things like drivers and apps, I feel OpenBeOS could put effort into providing drivers for Zeta. They could help grow YellowTab, and YellowTab could also employ these people.
I think the YellowTab people are trying to give the community a product they lost, and for themselves create a niche market like Apple has going for itself.
Delete my post. Because I like and agree with Axeld post.
I have personally worked with David, he is an absolute joy. One of the finest programmers and people I’ve ever had the pleasure to work with. The fact that OBOS could not hold on to him is a sign of *their* weakness, not his. His only weakness may in fact be that he got involved with a project like OBOS in the first place.
“one person working on something mainly alone, without interruptions and laziness, can do much more than a group of people”
Take your own statement and apply it to David’s situation. It’s clear to see why he was so upset. All the mailing list archives are good for is too see the after effects of a mismanaged project. Had the others stayed out of David’s way, there would have been no problem.
I won’t mention my opinion of you, my observation as someone who has been involved with BeOS longer than most, is that you’re completely clueless.
“Had the others stayed out of David’s way, there would have been no problem.”
That’s not what i said. I ment that even one man alone can do very much.
So even if only few people really do something on OBOS they can make it reality. Yeah, OBOS has lacks in management, but still those few can make it.
If You join some project with other people, and You can’t agree with rest, than it’s not their fault. You can be wrong also. No one has right to think he/she is ideal and makes no mistakes. Of course they can be wrong too, but it’s their project not Yours.
David left OBOS and AFAIK joined another project, so what’s the problem? He does what he want (probably), and OBOS people do what they want.
Axel THX
If Axel is as good as many people say he is, there isn’t any reason that OBOS can’t be completed in a reasonable time frame. Look at AtheOS. While it’s now a comercial quality OS, it is pretty damn good for one person who coded it in his spare time. OBOS has more support behind it along with more hackers working on it. On top of that they don’t have to design it, that part’s already been done. So they may have it done in a few years. On the other hand you can look at ReactOS and look at how far they are. It took them a while to get where they are, and it could take a couple more years before they have a windows clone that’s any thing near usable(for non console usage). Who knows, OBOS may be achiving usefulness by this time next year, or ten years from now.
oops, AtheOS is not a commercial quality OS. Sorry, i didn’t mean to say that it is now a commercial quality OS, just a typo.
David was one guy doing a big amount of work for OBOS. His real life got in the way of his OBOS development and certain OBOS people would not accept the fact that he would be gone for periods with no way to get a hold of him. Due to their child-like behaviour and total lack of patience, David quit. Now who is filling David’s shoes?
AFAIK David turned completely away from the BeOS community except to watch from afar as one would watch a TV series. That’s what I got from his recent editorial on BeOSJournal. Please correct me if I’m wrong. Either way, I’m sure if he’s reading this stuff, he’s quite entertained.
Shard, keep in mind that you can seriously hurt OBOS by slighting YT. If you think this is untrue, than you are certainly not the “nice guy” Axel says you are. I would also say that you did contribute to David’s leaving because you sour the BeOS community with your “direct” speaking.
Your slander against Bernd/YT/Zeta is as stupid as Bush’s preemptive strike war on terror crap. You view the closed source system as some kind of terrorist and you’re lashing out at it in unreasonable ways.
Just be quiet and if the whole thing flops *then* you can gloat as much as your greedy heart desires. Until that time comes, please be truly nice and help *promote* BeOS stuff (that includes Zeta damnit!).
Eugenia brought up the Amiga, and let’s stay on that topic for a while longer.
MorphOS, which is the new contender, has brought in a lot of code from AROS, which is the open-source Amiga clone. The same could probably be done by Zeta, licences and human factors allowing.
OTOH, I must have been following AROS for seven years now, and they’re still a few years from target, I think. Most of the structure is there, but certain programs need to be written, and there will probably be a GUI overhaul sometime before they can reach a 1.0 state. This is for the OS, never mind hardware support or actual applications.
Therefore, I’d say that Eugenia’s ten year estimate for OBOS isn’t that far off. Reimplementing BeOS can’t be simpler than reimplementing AmigaOS, and the staffing situation at OBOS can’t be much better than AROS, either.
Still, open source development always takes astronomical amounts of time, but there is always the chance that they will eventually get there. One day, even GNU’s HURD may work, though it may not be while we are still alive.
Nevertheless, in spite of the not quite ready state of AROS, there has been enough source in their CVS tree to boost the development rate of MorphOS, while the main corpus has been done by paid or well-determined programmers, whose efficiency usually leaves open-source developers in the dust.
What I mean to point out, is that cooperation between companies and open-sourcers is possible, and in fact favourable (even for both parties, since MorphOS has fed the AROS CVS a bit, too).
As for Eugenia’s perceived split in the Amiga community, I think that it’s overstated. Certainly, there is an inevitable split of a kind when people eventually choose their platform, but all the bollocks is mainly caused by a few trolls who probably couldn’t care less about the fate of any Amiga-like system.
Still, this kind of split can’t really happen in the BeOS community for a long time. Either you buy YT, or you don’t. OBOS or B.E.OS are still non-contenders, just as AROS has never been able to split the Amiga community in any way.
Creating something from scratch (with questionable “improvements” as you should not forget that binary compatibility should be kept), is 10 times more time consuming than YellowTAB working on directly improving the existing parts.
This is a myth, working from scratch can be faster. It allows you to fix things at the design stage rather than the development or maintance pahses which is commonly considered to be 10x and 100x slower respectively. Read a book on Software Architecture if you don’t believe me…
There is no reason why you would need to re-implement the whole kernel AND keep binary compatibility! This is a huge task!
OBOS is not so straight forward as I might suggest above because it’s bound by an existing design and binary compatibility but I think there’s some room for improvement. Companies need to ship products and this invariably leads to developers taking shortcuts – as the Be developers apparantly had to do sometimes. That said OBOS will not be 100% binary compatible.
Sure OBOS is going to take time, but it’s going to be good.
If it was to be abandoned we wouldn’t have projects like the one recently mentioned to implement the BeOS API on Windows – it’s based on OBOS code. I somehow doubt YT would want to do this, be allowed to do it or have the time.
However OBOS gets a bed press because it’s taking time, at BeGeistert they explain why things take time and I think you’ll find that progress is much better than appears.
OBOS is made up of many parts but many depend on libraries and the kernel so these have to be done first and this so far has taken a great deal of time.
As for YT, they too are doing a good job and recognise that they’ve made PR mistakes, Zeta will be more than a bug fixed Dano.
As for YT cooperation with OBOS I think you’ll find people developing and contributing to both projects.
As for developing an OS taking a long time, if you went back in time with this knowledge would you tell Linus not to bother? (Note: BSD users should refrain from answering this ;-D )
can someone explain what’s wrong with darkwyrm’s code???
what’s the reason for this darkwyrm bashing? prototype #7 seems pretty good to me! or are you people just talking out of your arse?
Eugenia wrote:
I believe that Axel and the rest talented programmers should join
YellowTAB and work on existing, usable code, instead of trying to re-
invent the wheel.
Hmm.. I think that they should be doing something similar to what B.E.OS
wants to do, and simply create a window manager and a BeOS API on top of
X and POSIX. This would unite many disparate projects (Cosmoe, B.E.OS,
BeFree (just mentioned today on osnews), others?) while at the same time
make much less work for themselves so something usable could be delivered
way sooner.
The GNU/Linux community wants X. A lot of C++ devs want a really nice
OO API. BeOS’ers want to program in the BeOS API again. Everyone wins.
Of course, the OpenBeOS’ers motivations are their own, and maybe they
aren’t particularly concerned with providing me a BeOS API that I can
use on my current system.
Even if Yellow-Tab gets this Zeta thing out the door, I don’t think I’ll
install it — it just seems too “Weekend at Bernie’s 3” to me.
Nicholas wrote:
As for developing an OS taking a long time, if you went back in time
with this knowledge would you tell Linus not to bother?
OS’s were a lot simpler (and smaller) back then, no?
Further, Linus wrote a kernel. The OBOS team’s greatest work is everything
besides the NewOS kernel that they’re using.
Yeah Sagres, I could do without the Darkwyrm bashing too.
Well said.
First: I’m not against Zeta because of closed source (I bought BeOS r5 Pro). I am against OBOS joining Zeta, because of closed source (You should know why – it could be end of OBOS).
In fact i wanted to buy Zeta, end even be reseller in my country.
If David is entertained than good for him
It was his decision to turn away, so why You keep crying? Also maybe it’s more healthy for him to not to be involved? As i see too many people are very nervous about too many things and it’s not healthy.
That’s why i’m not in any OBOS team – i would not have time nor mood (i’m lazy person) to work on it all the time, and i don’t agree with everything they do.
OBOS use some code from me and i’m happy with that. If they didn’t than it wouldn’t be end of the world (most probably someone would write the same thing and with better code
You, Eugenia and others can share Your opinions but not me? So You say only one side has right to share their opinions, and other side can’t?
I don’t agree with that.
“Just be quiet and if the whole thing flops *then* you can gloat as much as your greedy heart desires.”
That wouldn’t be wise, don’t You think? It would be like giving a motorcycle to someone not knowing almost anything about it and not saying that it can be dangerous. And after accident, saying “oh, i thought You shouldn’t do that”. I’m comparing like that to say that it’s good to talk FIRST, not to say Zeta is certain death.
If someone wants/needs Zeta than why not to buy it?
Just don’t blind people with false promises, or hide some facts to make people think about it as a holy grail (as some people already do).
“Until that time comes, please be truly nice and help *promote* BeOS stuff (that includes Zeta damnit!).”
I will not promote what i don’t trust (YT). That’s why i’m not marketer
BeOS is best OS for me (even if i have to reboot to windoze to work & play CS
, but doesn’t have to be for others. When it’s reasonable i talk about it to people.
there are some concerns that beos, OBOS will be “dated” by the time it is done. I don’t think so. It takes a lot of time for technology to be commercialized. Some examples.
CDMA: invented in the 1940’s, gained broad appeal in the 90’s
OFDM, invented in the 70’s (not sure but at least 25 years ago). Just gaining implementation today and still not “mainstream)
Unix (jesus how old is unix 30 years)It is just starting to gain acceptance on the desktop.
Punk rock: started in the 70’s and its turned mainstream in the late 90’s (a variant (unfortunately) that really sucked)
It takes time to gain acceptance. The concept of a Media OS that is built from the scratch for Media is still unique. Light and fast are still unique and if you look at the roadplans of windblows, OS X, or the progression of linux none suggest the same set of attributes that made beos special. The creative world still lacks an OS optimized for their needs, one which is small profile.
I applaud OBOS’ efforts especially their commitment to open source and i don’t care if it takes five years.
Truth be told, no one can really predict when we will be “done”. That is the nature of (small) open source projects. Axel, Marcus and Ingo could all get together for a beer and a meteor could take them all out. How much would that set us back? Or Intel could call me and tell me that they want to fund us with $20 million and we could be done in a far shorter period of time than anyone things.
But I would like to speak (type) briefly about alleged mismanagement and David’s departure. I will not discuss the specifics of his leaving – I think that is uncalled for. But I will say this – Peter doesn’t really know what went on. Axel, Marcus, Ingo, myself and a few others do. And no one could further off than what was posted here. David’s decision was his own, and for his own reasons. As far as mismanagement, I will freely admit that this is my first gig at managing an OSS project. My email address has never been a secret. If you have the secret to good OSS management and want to share it with me, well, the link is just above. Go for it. If not, well, then that should indicate something.
Look – to set the record straight on a few things. We have accomplished a whole heck of a lot. Most of the kits are in alpha. Some we aren’t going further on because they have proprietary BeOS dependencies – the Input Kit is a good example; the app_server/input_server protocol is not published. The big three are really what are left – Media, Interface and Kernel. That shouldn’t surprise anyone. It certainly doesn’t surprise us. And all three are under development by the best coders that OBOS has. What more can you ask for?
Finally, as far as Zeta… AFAIK, they have licensed some (not all) of Be’s code. Which is cool. But that puts them in a situation. First of all, they have debt, in a sense. To VC. You know – the same people that changed the whole nature of Be, Inc. Be was very cool pre-IPO. They spoke freely and were fun. Post-IPO, is was not as much fun, but I was hooked on the product. 🙂 Furthermore, many BeOS users and devs have been burned by a company (Be). Let’s pretend for a moment that we dumped OBOS and worked for Zeta for free. We would all be “contaminated” by seeing BeOS code. Then, what happens if Zeta runs out of money? BeOS dies. Pure and simple. And how much will money effect Bernd’s decisions (by necessity)? This is not to say “don’t buy it”. I will and you should, too. But, understand what you are doing and what we are doing and why. OBOS is about ENSURING that BeOS never dies. As is Zeta. But with truly open code, I think that people have more assurances.
Peter Schultz said: “Had the others stayed out of David’s way, there would have been no problem.”
Well, you are talking bullshit.
There have been problems because of David’s way, and because of that others stepped in his way, including me. But nobody forced him to quit, it was his free decision. So please stop talking about things that are not of your business.
And please don’t call David Reid a genius. He’s a talented coder, and certainly already achieved a lot of things, but he is not a genius.
And regarding that YellowTab / Zeta thing, I won’t buy it. I was promised to be given a beta a few times now, but it never happend
Look, I don’t mean to be too mean, but I just have to congratulate ryan for what I consider to be the dumbest post in recent OSnews history. Talk about mis-using analogies!
CDMA: invented in the 1940’s, gained broad appeal in the 90’s
CDMA’s capabilities are adequate for the needs of the decade in which it was deployed. It’s just a protocol. Do you really think that the BeOS’ capabilities (assuming its development continues at its current rate, and the same for its competitors) will be sufficient long in the future? Heck, for almost everyone, they’re insufficient now.
Unix (jesus how old is unix 30 years)It is just starting to gain acceptance on the desktop.
The Unix of 30 years ago is most certainly not gaining acceptance on the desktop. The Unix design, coupled with thousands and thousands of improvements, is. Desktop users aren’t moving to Linux and Mac OS X because they finally realized that the scheduler is cool and that grep rocks their world; it’s because Unix provides a strong underlying architecture, allowing their OS of choice to do all sorts of neat things while remaining stable and so forth. They’re not just using Unix; they’re using systems with value built upon Unix. The BeOS is gaining such improvements at a rate that I would estimate to be 1/100th or so of that of its competitors. Go grab a PDP-11 and compare it to Mac OS X on a modern machine. Then try to tell me that they’re the same product with a straight face.
Punk rock: started in the 70’s and its turned mainstream in the late 90’s (a variant (unfortunately) that really sucked)
Punk rock is not a platform; an operating system is. I can go out and buy a punk rock CD without changing my stereo, my speakers, or anything else. I cannot change my platform without replacements for all of my software, and possibly some hardware changes. This is a horribly mismatched analogy.
For me, the best part is that by your logic, the Coleco could be heading for some real mass-market acceptance in a few years. I’ll bet Sony, Nintendo, and Microsoft are shaking in their boots.
The concept of a Media OS that is built from the scratch for Media is still unique
Please explain to me what about the BeOS’ design makes it so different, and more importantly, so much better, for media than other operating systems currently on the market. That is, I would appreciate your providing specific examples of design (rather than just marketing hype) and concrete examples of why it benefits media reation/playback/whatever. In other words, “because it’s pre-emptively multithreaded!!!” is not an acceptable answer.
or the progression of linux none suggest the same set of attributes that made beos special
Again, could you please explain to me what these are, and why they benefit the user so much? I ask that you answer in terms real, practical value. In other words, “replicants are cool!!!” is not an acceptable answer.
Nearly every single post to every single story on this site is bullshit, Marcus.
David is, by definition, a genius, and so are you. Deal with it!
So what if I don’t know the exact problem. I got the gist of it and now you guys are trying to gloss it over. Rather than work things out, _YOU_ALL_ started a big fight. It is nobody’s ultimate fault, but now a good coder is gone. I understand your defensive position, as you’re still trying to maintain your project, but it does not take away from the fact that _YOU_ALL_ made a horrible mistake. If it’s not my business, then you should remove your sources and your developer relations from public scrutiny. That is the only possible way that can make it not my business.
Happy coding!
> David is, by definition, a genius,
Holy cannoli! You’re right. I just checked Webster’s, and there, in the entry for genius, is a picture of David. Also, coincidentally enough, I looked up “virile” and there was a picture of *me* there, though I don’t recall giving them permission to publish that compromising photo…
> Nearly every single post to every single story on this site is bullshit, Marcus.
Noone forces you to browse this website.
> David is, by definition, a genius, and so are you. Deal with it!
I too thought I was a genius long ago… then I discovered I was only human
Get a life.
> So what if I don’t know the exact problem.
If you don’t know you don’t need to talk about it.
Only parots talk about things they don’t understand.
As I understood when this happened, David left because he couldn’t stand someone working in “his” part of the CVS tree. Damn, cvs is a tool for cooperation, besides, there are means in cvs to lock files if you really want no other to touch them for a while. Anyway it’s how groupwork works.
I didn’t say I don’t like the site, now did I?
David left because there was no reasonable way to work out the conflict within the group. I talked with a high level OpenBeOS developer at some length just after it all went down, and know enough to feel perfectly confident about what I’m saying. The fact that one after the other of you keeps coming up with shit like “only parots talk about things they don’t understand.” Show that you’re grasping for straws. You have no sound counter statement so you resort to insult.
Your group is weak because there are no guidelines for resolving or even avoiding conflicts such as the one we’re discussing. Instead of acknowleging and fixing the problem you just attack me, and your idea on completely locking others from a file is certainly not good. You’re going to have to try harder in order to avoid such conflicts in the future. I can’t imagine why any group would want to take such risk. There’s a major difference between chaos and anarchy. If the project grows and continues to mature, you will encounter such chaotic events again.
First, you were the first to attack here.
Second, I think the “bazaar”, as would say ESF seems to work quite well up to now.
Of course you won’t stop people from disagreeing, that’s human behaviour. That’s why there are stupid wars out there. that doesn’t mean each time ppl disagree they need to resort to that, or just quit. Humans can also talk I heard.
s/ESF/ESR/ of course
Peter Schultz wrote: “David left because there was no reasonable way to work out the conflict within the group.”
I say: David left because he had no interest in solving conflicts. Because he was unable to handle even minor conflicts in a normal way. He took the cheap way, and just quit. This didn’t show that he’s a genius.
You also wrote: “I talked with a high level OpenBeOS developer at some length just after it all went down”
I say: Nothing went down. That’s bullshit, once again. You don’t know what you are talking about When David left, what remained were large parts of a ported BSD stack, that workes unreliable because of the different locking needed by BeOS, compared to FreeBSD, and a few other errors. This project is neither in chaos, nor anarchy.
And which high level developer are you talking about? Looks like you can’t even have the heart to tell us his name. Perhaps you didn’t comprehend what he was going to tell you.
Nobody of us is grasping for straws, and I believe not even David would agree with what you are saying here.
Unix (jesus how old is unix 30 years)It is just starting to gain acceptance on the desktop.
UNIX was never aimed at any desktop. UNIX gaining desktop acceptance has been as silly a concept as UNIX gaining acceptance on the C64.
Punk rock: started in the 70’s and its turned mainstream in the late 90’s (a variant (unfortunately) that really sucked)
Where do you live? India? Punk was commercialised and mainstreamed to death in 1977.