The latest issue of the Haiku’s newsletter has arrived, read it here.
New Haiku newsletter Available
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
What do you guys think… will we see Haiku R1 in 2k5 or 2k6?
With the progress of the app_server and the kernel getting farther along in development I think that we should see something usable soon. As far as r1 thats a lofty goal, reatos has taken a long time to get where it is and I think that haiku will also take some time before binary compatibilty is 99% of the way there.
I also think, that Haiku R1 will not be released before 2008 ir even later. Creating a stable Operating Systems is certainly a damn hard job and the programmers are doing this job just in their spare time.
There will be various alpha releases where you don’t have 100% of intended final features, with quite a few bugs. Maybe even a beta with quite a few bugs, but I’m not going to hold my breath with the expectation that a full 5.03 compatible system that functions at a binary compatible level, compatible warts and all, will be done before the end of this year. And what will exist is likely to be suboptimal for speed, as they’re (hopefully) more concerned with making it work correctly before optimizations.
The funny thing is that some of the incompatibilities before a final R1 release may be from situations where what was intended in BeOS 5.03 is actually made to work in Haiku, but that makes it incompatible, thus breaking many applications that rely on those bugs of documentation.
But in reality, Haiku R1 is merely a stepping stone to the long-term future viability of the system, and something that could be listed under the “necessary evil” category of creating something to start with that’s a known quantity. If there was no desire to worry about backwards compatibility, perhaps more would be done now, perhaps less, as it’d be easier in many ways to define whatever exists as the correct result, rather than programming against an existing behavior of a complex system that’s not 100% documented, or not 100% documented correctly. A huge amount of the time involved to getting a suitable final Haiku R1 release will likely be due to testing for backwards compatibility, and tweaking Haiku to behave as the older applications were implemented to expect. And people wonder why Microsoft, with all their resources, can take so long to put out new releases…. well, consider the amount of software in comparison, and their aim (thus far) to maintain a high degree of backwards compatibility from one system to the next.
There are also issues with Libc being a bit old. The need for changing compilers will probably become more aparent as we get closer to R1. My guess is that there will be a compatiblilty layer that instead of haiku becoming 100% compatible there will be two sets of librarys, the beos compatuble set and the haiku native set. My guess would be that all new apps should be writen for haiku and that certian essantial apps that are not open source such as gobe productive will be supported with the backwards compatible set of libs. I dont think it will take till 2008 for haiku to be usable, and by that I mean installed on a hard drive and the thing boots up and runs like it should. I would guess more like next year. just my two cents.
I believe that we’ll have a functioning Haiku, maybe not 100% but it’ll be more than useable, well its useable now because you can surf the net in Lynx =P Anyway I can imagine that Haiku should be finished before 2006 finishes.
I think, the best best way to get a working release of Haiku as fast as possible, is to support the developers with donations or to help them finishing their work by coding ourselves.
The most important thing is, that the Haiku Projekt gets more skilled programmers! And, in deed, this is a very promising and interesting project, that should attract muxh more programmers.
The newsletter is fancinating. Truely. But where is the news?
I see many peoples claims that Haiku development is slow.
So, why don’t you help them ?
(you can’t develop ? donate money !)
<< But where is the news? >>
Well, feel free to join in and make some !
I won’t comment on when it will be done, but I do want to say that still, to this day, binary compatability has still *not* been the big issue that everyone expects. We have many apps that run on both R5 on Haiku now, including Lynx and our app_server tests. The binary is literally copied from R5/Bone/Dan0/Zeta to Haiku and it works. If any of the devs are having a tough time with it, they certainly haven’t told *me* about it. So far, it just works. Obviously we haven’t tested every application out there, but for the things that we have tried it is not an issue. We aren’t likely to change compilers or libc for R1, either.
Having said that, the one thing that no one could have predicted when we started OBOS/Haiku was Zeta and the popularity of Bone/Dan0. People in the Be community are facing something that we never had to before – different versions. Not the DLL/.so pain that Windows and Linux users have to face, but still some hint of what that could be like. It should be obvious to everyone in the community how important it is that this not get out of hand. We work with Zeta and Zeta works with us. Interoperability is critical and it is something that can’t be ignored.
Yeah, the development is slow. I don’t think the issue is having people develop more for it or donate more. I believe that deep in our hearts, we know BeOS is gone and isn’t really comming back.
I for one would rather see BeOS DIE AND STAY DEAD than a few scattered projects work on something that should have been left alone. IIRC, that was the wish of J.L.G. to begin with.
It’s sort of interesting and nostalgic to see stuff like OpenGL support showing up in the news, but the time for that was at least 5 years ago if not one or two more. Let it die.
Penguin Go Back to The North Pole.
BeOS Community is alive and well.
Check bebits and all BeOS forums and you will see the latest happenings.
BeOS is simple and does what needs to be done without bloat.
Yeah, Ive got a BeOS machine running, and I do see Bebits et all because of that. It doesn’t change anything. It’s nice for nostalgia, but the “cool new things” it has really should have been there before Be Inc. closed its doors. This is all stuff that everybody else already had for years. It’s pretty sad when 3 out of 5 of the top downloaded (or 10 out of the top 15) apps on Bebits are drivers.
Let it die?
What the Hell… Zeta is a commercial product, ALIVE, selling thousands of copies. Haiku is a parallel project that is fascinating thousands of users over the NET…
When I see this, I can think everthing but the death of BeOS…
A Dead Os Selling over 80,000 copies.
“BeOS” is dead. We shall see no new releases nor updates nor fixes of “BeOS” from Be Inc. Loved their logo, but Palmsource couldn’t care less.
What is alive, then? Zeta, Haiku and friends of course. continuing to build on what was GOOD abut BeOS, despite the fact that Linux distros, BSD distros, MacOS X (especially), and Windows have caught up and in some areas overtaken the BeOS 5.0.3 we knew and loved and put up with, warts and all. I mean, you’re gonna complain, berate the GNU and Linux projects for copying old UNIXes and the BSD’s and building on THEIR ideas. Berate the BSD’s for copying … themselves. Berate ReactOS for trying to build on what worked in Windows NT. Hell, berate Microsoft for copying ideas from VMS in the original Windows NT.
If something was good, worked, you LIKED it, but it’s no longer available to you, why NOT try to bring it back if you can? The Haiku team is and good for them. I’m sure, in a few more years, Haiku will be a VERY slick and useful hobby OS, with Syllable, SkyOS, AmigaOS, reactOS, and others. They’ll get to putting in all the stuff that Be never got a chance to and more, and likely borrow ideas from other successful projects to improve their own. I’d bet my dime that Yellowtab would shift to Haiku (when it’s ready, netting Haiku some funding and freeing yellowtab to leave the whole “who has the rights to what” silliness behind them).
Hell, we ALL win in the end.
–JM
>>>A Dead Os Selling over 80,000 copies.
That number is not that believable if you look at the historical record. At the height of Be Inc.’s business, Be Inc. never had more than 2 million dollars in annual revenue. At $100 retail, Be Inc. never sold more than 20,000 copies a year — at the height of BeOS’s popularity.
BeOS not dead. For us musicians the mainstream OS’s are a pain to work with. Nothing has given me more hope than being able to work on an OS that just gets out of the way and allows me to fully utilise my computing power.
BeOS gave me this hope and Haiku will realise it in conjunction with YellowTab.
Although it looks like things are goin to break in different versions, the differences are minor – a bit like Win2000, WinXP type of situation.
I think the obvious direction for Haiku is to aim for 100% Dano/Zeta compatibility. It is quite easy to provide backward compatible R5 stuff (net_server mainly) on top of that. So in the end everything will be quite consolidated, perhaps more so than in Windows or GNU/Linux distros.
I don’t think that BeOS derivatives will die any time soon. I’m hanging out for Haiku because I want to. Damn man, why so negative?
In the short time that I played with R5, it felt solid to me. Don’t get me wrong, I use Linux for my server, but never for the desktop – it feels really disjointed.
That said, a lot of the power that I have grown accustomed to in Linux (mostly bash CLI) I really miss in Windows. The tools are available, but not by default. It is my hope that Haiku will help to bridge this gap.
I could care less about digital cameras and MP3 players (I own neither) I want stabilty and power – by default – out of my PC.
I’ve nothing to do with the Haiku team but i find it unfair that it doesn’t have it’s own topic and icon artwork on this site..
BeOS is history and Haiku is a real, serious, living project.
Be (inc) and BeOS failed in many ways and it’s not fair to bring their logo when the Haiku team is working hard to make this great os a reality and to give it a future.
When you look at the existing topics page, you’ll see my point.
http://osnews.com/topic.php
long life Haiku
Don’t start it again.
You probably missed big war here, when Eugenia replaced BeOS icon for all related topics with OS Zeta icon.
Inlc. BeOS and Haiku topics.
Imho, current icon is most reasonable choice. As old Amiga icon used for all derivatives etc.
Imho, current icon is most reasonable choice.
Totally agree. And kind of practical too. There are still a lot of Be-BeOS users like me who follow news for Zeta, Haiku – nice to have them grouped by common denominator
I don’t want to start a war here, i know the subject is sensible..
“You probably missed big war here, when Eugenia replaced BeOS icon for all related topics with OS Zeta icon.”
Yes i missed it, and yes i think this was unfair and illogic.
“Totally agree. And kind of practical too. There are still a lot of Be-BeOS users like me who follow news for Zeta, Haiku – nice to have them grouped by common denominator
”
So following your logic, every linux distro should be grouped by their common denominator, Linux
But look, there is the Linux topic and the Ubuntu one (how long has this project been going?) !
Apparently there isn’t any debate about this!
It just doesn’t make sense anymore. BeOS doesn’t make news anymore. And this is a site about operating system news so why would it be confusing for anyone to have a NEW topic for a NEW system.
I used to be a BeOS user, and i follow news…
This is bad publicity for those people that have nothing to do with Be Inc. that’s all i wanted to say..
So if a os project tries to be compatible with Windows, you’ll have every news under the Microsoft topic… Nice.. Thanks OSNews, you make things clearer.
friendly,
alex