OSNews

I forget which version it was, but I have the CD from when they were supporting various PowerPC computers other than the BeBox. It will be great once they get x86 finished far enough to warrant working on other platforms and we see PowerPC support once again.

That would have been Preview Release 2, most likely. It was widely distributed as a free-to-use version.

re: 4.5 CD ... me too!

Hah! I 'see' your Beos 4.5 and raise you a Beos 3.1! ;D

I up your 3.1 and say I have 3.1 and the BeOS Bible. One hell of a book

I think we have a winner

Looks like you can get the BeOS Bible on amazon for about 5 bucks including shipping.

I see you 3.1 and BeOS Bible, and up you BeOS Bible, BeBook, Be Advanced topics, Practical filesystems design, GoBe Productive 2.0, Corum III, BeOS Pr2 running on a dual processor Mac. As added collateral, BeOS 5.0, 4.5, a 4.0 demo disc.

In 2007 I could have upped even more, I used to own a BeBox 66Mhz and Dr 8.1, AA, and Pr 1.

I could also burn in hell for mentioning my Zeta collection (Neo - 1.2, though I got them as a developer for free or minimal update of 10 euros.)

I do actually still have my BeBox here

I have a Be Inc logo badge (pin) too... I forget who gave me it, but it is an official one, not a repro.

Edit: I also used to have JLG's signature on the BeBox welcome letter ;-)

Edited 2012-11-13 16:04 UTC

First comment on OSnews wich I have been reading for years
I raise you with Civ:CTP unopened, PersonalStudio 1.5 running on a Pentium4, various japanese books like the "BeBox GuideBook", BeSpecific discs, shirts, a dual133 BeBox running DR7 (soon replaced by DR6) and an Edirol DV7-DL coming my way soon.
OT: henderson101, don't forget me on bebox.nu, I sent you a PM there.

I burned a nightly release two weeks ago, I could have waited, thanks to the Haiku team for a very good job!

you lucky bastard
i once tried to buy a used bebox, but thanks to the shitty banks in austria couldn't finish the deal >:(

does anybody know what happened to bernd korz?

You can find his blog here http://www.berndsworld.com/

Hogwash, OPENSTEP was a much better os.

All I have is a boxed copy of Zeta that I got (from that guy who organizes BeGeistert) for helping with porting Python.

But by then we won't be using 1 GHz ARM with 512 MB of RAM.

Well, I certainly will! I think a Raspberry Pi 512MB version overclocked to 1GHz running Haiku-ARM would be nice.

The BeagleBoard-XM also meets those specs, and uses a newer ARM chip that might be easier to target. There's also the PandaBoard if you want even more power.

..OR , for a rather more money at $199
--the dual core 1.2Ghz cotton candy?
http://store.cstick.com/cotton-candy.html
--i wouldn't bite at that price, maybe if they can knock $60+ off

or more budget MK802 with single core 1.5ghz @ $74 http://www.engadget.com/2012/05/18/mk802-beats-cotton-candy-to-mark... , now at the Pi-like price of ~$50 on ebay.

----
Off topic but just upgraded my sheevaplug to ubifs debian squeeze after bricking it, god bless jtag ports. and then installed subsonic.war under tomcat6/oracle embedded(ARMv5 headless)java1.7.0_06
and my 1.2ghz arm sheevaplug (after a night indexing) is now happily serving a 1TB music library(attached via usb NTFS-3G/FUSE) to several simultaneous clients over my home network and to mobile client apps over the internet. -I'm amazed what can do with literally just a few watts of cpu power these days..

I'm absolutely loving the small computing revolution! My long term goal is to be "x86 free" for general computing, not for any philosophical or moral reasons but for practicality. Ironically, my last holdout will probably be my ~1999 AMD Duron system that currently serves as both a native BeOS 5.0 Pro workstation and a Windows 98 classic gaming system. Once Haiku goes beta or release on ARM (yes, I'm aware that could be many years from now) I'll look at chucking that dinosaur for good.

I also want to break into the microcontroller world beyond the simple tinkering I've done with TI kits, but that's purely hobby stuff and can wait until I've gone down to one employer.

2000 at the earliest, Duron launched halfway through the year (hey, if you can't be pedantic on ~tech websites, where can you be?! )

And now I wonder if my dual Pentium II 266, if with BeOS, is worth some bragging rights ...well, I suppose it would be one nice BeOS machine, back in its heyday.

Well, that's why I said approximately (~) 1999. I was going by the fact that it had a Windows 98 CoA sticker; I figured if it was made in 2000 it would have Windows Me instead. I got it in a trade several years ago so I didn't know the history of it.

I once had a dual PII system, I never tried BeOS on it but I'm sure it would have been pretty good! It was a Compaq professional workstation with Windows 2000. I gave it to a friend who needed a low cost file server; since it had SCSI she could just throw the drives from her dead server into it.

No, I'd much rather see multi user support. I'm not planning on running it on ARM anytime soon.

Multi user support is just not BeOS though. /jk

It won't happen before R1, as R1's goal was to be on the same level as R5.03 of BeOS.

Edited 2012-11-13 12:12 UTC

Right, understood. I was replying to a post that was asking for features post R1.

But that doesn't mean it will be a priority. I'd rank ARM higher than multiuser. I'm not really sure I've ever needed multiuser for BeOS. It was only ever me using the machine, it's not like anyone else was. Plus, ARM is already in development... and is reasonably close to working prior to R1.

Virus and exploits?

I surely don't want exploited applications to have full access to the complete hard disk.

That has nothing to do with multiuser though. That has to do with privilege level. The way both Windows 7 and Mac OS X (and a few Linux desktop distros) allow privilege escalation is the way forward, not adding extra accounts and complexity. (Though Windows 7 is a bit weak, not requiring a Password all of the time.) So as an example, I have a single account on my Macbook under Lion, but I get asked to provide credentials whenever I do anything "dangerous". Under Snow Leopard, I went one further and created an admin account and revoked admin rights from my default account, but that was total overkill.

Multiuser is something of a fail... outside of the University mainframes when I was in college, I've NEVER seen or needed multiuser support in any capacity, be it home or work, in over three decades in the computer industry. Not to say they shouldn't work on it, just that there are much more important features they should work on first.

Note that privilege/protection level is separate from multiuser. For normal usage, the single user shouldn't be running at a "root" level. This is a security issue more than a multiuser issue.

If you have processes running with different user accounts, you have multi user.

And you need to do that because...? Especially on a Desktop machine which runs no server processes.

Well not everybody like the linux way in doing stuff; in this haiku seems a good plan B so far, hoping that also AROS (that am advocating) will come to be usable for modern computing needs soon too.

IIRC AROS doesn't even really plan introducing proper memory protection? (since it would break compatibility too much?) Can an OS without that be really modern?

Oh well, at least still fun; and I think I like your avatar, is that some cat marvelled with a running instance of AROS? :p

But then again I never mentioned Linux.

I don't see how they can have a "common cause". AROS is a recreation of the AmigaOS and Syllable is an extension of AtheOS. Haiku, of course, is a recreation of BeOS. Each has a different API and code base.

To each their own, I guess. I prefer to focus on the positive side, in that Haiku still makes progress after all these years. The quality of the nightly releases over the last year is simply amazing considering the size of the team and the struggles of recreating an entire commercial OS from scratch.

I realize that Haiku may not have as much relevance today as it did five years ago, but I have faith that the R1 version will be a viable platform for those who want something different.

On a personal level, I'm looking forward to seeing new and updated DAW software for the platform, as that was its biggest draw for me back when Be Inc. was still alive.

That common cause being...? As many others point out, Linux is not what these developers want. The API of all three are incompatible (though Syllable mimics the BeAPI quite a lot.) They all strive for a different goal. What commonality are you seeing here?

The developers are of course free to do what they want with their time.

Me I just want a good OS. It doesn't have to be what AmigaOS was in my case.

For me personally I would likely had benefited if they all had worked together on a common OS because well, then maybe I would had used that instead of not using either of them as is today. Though Icaros desktop seem very interesting and something I kinda would had wanted to run by now.

You can't fit a square peg in to a round hole. All three projects, plus ReactOS, are trying to target a different goal. Any similarities AtheOS had to BeOS have been squished quite a lot by Syllable's direction towards their weird little VM language intepreter pet project. The AROS is an Amiga alike OS. ReactOS want's NT compatibility. Haiku wants BeOS compatibility. BeOS has absolutely nothing to do with AmigaOS, bar some UI ideas they borrowed. It borrowed as much (if not more) from MacOS classic UI.

Again, where is the commonality? There is none. The only thing 3 of the projects have in common is that they want to recreate another OS in a way that is (somewhat) compatible with the original OS. Syllable is the only one that has no legacy leanings, but I wouldn't hold Syllable up as a model for OS development, not since Vanders etc left the project.

I guess one way too see it is that Syllable seemed like something which kinda wanted to take the good stuff / hints from BeOS and AmigaOS and use them anyway. But it's not really neither.

Haiku is focused on being what BeOS was. Or at least was focused on that. Rather than being what BeOS could had been, BeOS would had been now or an OS with the same benefits as BeOS but something new.

Similarly AROS always was about being like AmigaOS on x86. Not a modern OS with the benefits of AmigaOS only better.

I liked AmigaOS but I'm not using and wouldn't really want to use AmigaOS now almost 30 years later. I don't really see why that should be your goal. I'd rather have something new.

Your interest and opinion may be different.
There's of course MorphOS and SkyOS to. And IMHO that's a lot of talent and ideas and work put into all those and more products without neither one becoming a larger player as is. They would of course had reached longer if they all had worked on the same product , though then maybe variation and different ideas is better and as some people for whatever reason has mentioned not everyone want Linux and while that was never what I said not everyone might had wanted that Syllable+Haiku+AROS+SkyOS+MorphOS-developers product. Me I might had been interested

Edited 2012-11-13 16:05 UTC

SkyOS? It died nearly 10 years ago. Even the last post message on their homepage is from 2006.

And Syllable is as good as dead too. As good as all devs are gone, and the improvements in the last 5 3-4 years, are as good as zero. There are no devs to fix the most simple bugs. Just look at their changelog, and you will see how few and how trivial stuff you find in there, things that are even not woth to be mentionated.

Haiku on the other side, has all changes to grow out of the "hobby os" image. It's needed to bring the packagemanager into a user-userable state, enable hw acceleration, enable window composition, update the GUI to a modern one with transparencies and so on... (and add some more drivers) and you will all agree that it's a quite modern os.
In terms of useability I really think, that haiku has a chance to catch up and to over run most linux-based OSs in the next few years.

Yeah, 2013 will be the year of Haiku on the desktop... wait... that sounds familiar...



...meanwhile in the real world - will 2013 be the year of linux on the desktop? Probably not... and there is even less chance of Haiku becoming a main stream OS. By the time haiko makes beta people will be beaming to other countries instead of flying and everyone will be using computers with HUDs over there eye balls apple will probably call their version of it iBalls.

Sorry but your dreaming if you think haiku is going to go in a different direction to sky os, riscos or any other niche OS.

Edited 2012-11-13 17:25 UTC

Do you really believe that is the goal though? I always thought it was going to be niche at best, but the best damned niche OS there was. I expected it to put Linux to shame on the desktop too.

I have never used it myself, but what I see in films on youtube reminds me a bit of riscos... especially when you have menus opening within menus within menus...

I think in terms of interface riscos HAD one of the best interfaces for its time... but its age now shows. I think when/if heiko becomes beta it will be showing its age before it's really been born.

When I look at what gnome shell looks like and does or macos, or windows 8 I can't see haiku being anything more than something that gets mentioned on sites like this.

Maybe all these niche os developers hould get together and actualy build together an OS that stands a chance of being used by more than a handfull of people. Instead of starting something that will ultimately be left uncompleted

Edited 2012-11-13 18:01 UTC

Nothing like RISCOS at all. I've used both and RISCOS is horrible.



I'll have to disagree. Mac OS 8 was one of the best overall user interfaces, bar the menu being stuck at the top of the screen. RISCOS was ugly, crash prone and the user interface paradigms were very unnatural. The file type system (also a Mac OS feature, also irritating) was stupid. File load/save was horrible. I really hated the RISCOS user interface with a vengeance and I was very glad when RISCOS finally died from the mainstream.




Why? I don't get why people keep saying this. Because it doesn't have alpha blended compositing and GL rendered windows? Seriously?



If you really don't understand the motivation and you don't understand the reasoning, I suggest you pack up, cut your losses, delete you OS News account and go home. Geekism is not for you ;-)

If these people enjoy what they're doing it's not a waste.

You're not MOS6510!?! O_o (but at least you have a nicer-looking, non-blurred Avatar avatar)

After seeing his I now understand what you mean with blurry!

It's the same as going from a retina iPad to a non-retina one I guess.

Or... going from how Avatar "really"[1] looked like, to being drawn/slaughtered[2] with bilinear filtering.

1. because, really, Avatar was typically shown on old CRT televisions - which introduced a certain amount of ~blurring/filtering themselves. I guess you'd just have to make a photo of Avatar from such display, if you'd care about it ;p

2. in the infamous style of the early Glide-accelerated or N64 games, where low-resolution textures were being turned into soap.


PS. So, when will you finally correct it?

Edited 2012-11-13 22:17 UTC

Perhaps I will!

What is your avatar? I think I can spot a hand in it.

I'm eating a sausage with mustard and bread (and in another hand, a beer)

I can't see what is human and what is food. Apart from the hand.

IT'S PEOPLE! SOYLENT GREEN IS MADE OF PEOPLE!!!

(come on, you surely saw that one coming ;p / increase gamma?)

Stop whining, better help the developers with your programming skills or money. Their very small developers team don't owe you anything.

Beauty is more than skin deep. The UI is rather plain looking by today's standards. But, the UI is not the most compelling feature for using Haiku. To quote from Haiku's site,

"Those of us who have used BeOS/ZETA in the past know the goodness of how extended attributes are for certain file types. The approach of having emails messages and contact information in individual files with attributes makes your data very portable and accessible even at the file manager (Tracker) level. You can, for example, switch your email client, and still have access to all your emails and contacts, as the data remains always the same."

You can read the rest at http://www.haiku-os.org/blog/koki/2007-05-08/settings_beos_style.

Many of the OS's capabilities are due to the 64bit journaling filesystem, which is capable of automatically indexing data and cataloging metadata.

I wrote that, back in my Haiku days. :-) My geek days are over, but I still have a sweet spot for BeOS and Haiku. Congrats to all the Haiku folks for their latest release!

BeOS was a kind of revival of the Amiga spirit.

An operating system for multimedia applications, which explored the hardware capabilities much more than the contemporary desktop systems.

Nowadays it is too late, but it could have become Mac OS X.

OSX is bloated and slow, though very capable of handling multimedia

Really? Running Windows 7 on the same hardware, OS X gives Windows a real run for its money.

It wasn't really. Amiga was just another OS. Most of the key developers at Be, especially to begin with, were Mac based or ex-Apple. Benoit Schillings, Steve Sakoman, Erich Ringewald and Bob Herold were at any rate. I'll quote myself here: http://www.osnews.com/thread?514205

For me it was, because of its multimedia pedigree.

That is more co-incidence rather than design. Be didn't go with multimedia right away, and the OS was written almost from the moment the hardware (Hobbit based) was in a position to be used. When the OS was being written, it was Benoit, Bob and Erich... all Macophiles. Erich also worked on Pink, so that has a lot more to do with the direction.

http://en.wikipedia.org/wiki/Taligent#Pink_and_Blue

An I get down voted for being right? Democracy.. lol.

I know we Europeans find it hard to believe, as Amiga was ubiquitous here - Amiga was fairly niche in the US. It certainly wasn't perceived as a serious machine by many consumers:

http://en.wikipedia.org/wiki/History_of_the_Amiga#Amiga_in_the_Unit...

Did anyone actually use the Amiga for anything but gaming, demos and tracking?

Wasn't Video Toaster a killer app for a while?

Incorrect: http://en.wikipedia.org/wiki/Killer_app#Selected_applications_for_c... ;p (hey, if Wiki doesn't mention it... )

No, but seriously, the Amiga application in the above article (Deluxe Paint) is probably more like it - Video Toaster was quite a niche (of a niche; even if relatively inexpensive in its field at the time)

Yes, many TV shows used it for video production.

Babylon 5 was one of them.

For more information see Notable historic uses at http://en.wikipedia.org/wiki/Amiga.

Most of those do fall under "demos and tracking" though, when more broadly understood ("animations and music", not just from people who called themselves "demoscene").

And yeah, Babylon 5 effects... which left something to be desired (at least SeaQuest CGI aged really well, thanks to naturally "foggy" environment)

Why not try it yourself and find out? It's not like it will cost you any more than a few minutes of your time and bandwidth to download a file, a blank CD-R (or CD-RW, or USB flash drive...), and the electricity needed to power your computer when running the OS...

If you don't like it, just reboot, chuck the CD into the trash and forget all about it.

The full C++ API was also a dream. Well thought out and complete in a time where the Windows API was a cobbled together piece of sh*t.

Multi-threading, filesystem events etc were all there and working. There was also the mandatory call IsComputerOn so your program could know if the computer was running. Paired with IsComputerOnFire you could handle significant failure conditions right within your code.

The C++ was actually a blessing and a curse. It tied Be to a very specific compiler for a very long time. It was hard to expand the API without a lot of planning or creation of second versions of the classes (they seem to have been doing this for the BeIA project as there's a second namespace with different classes in the Dev kit I've used.) Adding methods to existing classes was costly as they reserved only a number of "slots". Using them up would cause a lot of issues.

The API being multi-threaded was cool, but it forced a lot of uncomfortable synchronisation on to the developer. There was no middle ground - you either adapted or your app sucked. It had the potential to make really simple operations very complex.

djohnston pretty much has it covered, but personally I don't like the UI all that much. I tend to span the deskbar and turn on single window browsing in Tracker whenever I do a new installation.

What drew me to BeOS in the first place (besides being an alternative to Windows 98 and GNU/Linux) was the amazing multimedia software that ran on it, as well as the simplicity and power of the file system. I could achieve a very productive and comfortable workflow that simply wasn't possible on any other OS at that time.

Since then I've found OS X to be close to my ideal both for audio work and for my general workflow, and recently KDE on GNU/Linux is manageable for the latter. But I have a feeling that a mature Haiku will be what I've really been looking for.

Oh, back in the late 90's Windows just terminally sucked. Mac OS Sucked. Linux on the desktop wasn't pretty ( visually or functionally). But BeOS... was beautiful and it worked. I only used it on x86, but it fully supported my hardware with zero configuration-- including my tv tuner ( which barely worked in windows without crashing it every 10 minutes). It was awesome. Plus it gave you a haiku in the web browser when an error occurred, that was so cool. Never got around to exploring the api much, just a few hello world programs, plus a port of my senior research program that modeled various high energy particle reactions.

That made me wonder: is that what your avatar sort of shows, is it somehow related?

No, the avatar is a balloon animal. Clowns make them for the delight of children of all ages. Its the logo for my defunct charity "People for the Ethical Treatment of Balloon Animals" or PETBA for short.

Yes, it did always look that way to me (too bad it's defunct ;> ) ...but, on this occasion (the above quote), it made me recall some diagrams of particle accelerators or molecules.

I enjoyed working with the database-like file system. Scripting and creating some dynamic website code in Python was fun too... I just loved how it all worked, looked, FELT...

When Haiku can do everything R5 (with the beta networking update) could do I will definitely consider moving back to it even ditching my games on Windows (or at least making Windows secondary).

The only API work I did was to add syntax plugins for an editor... so that doesn't really count.

I think it's that there's no simple way to deal with xz'd tarballs on Windows.

OK, so I checked and a 2 year old version of 7-Zip handles them just fine... (would be a bit weird if it didn't, since XZ appears to be a stripped down variant of 7z)

7-Zip. http://www.7-zip.org/

FAT is probably simply good enough, supported by everything (also millions upon millions of existing embedded, low-powered stuff), hence convenient.

It's not like MS doesn't support 3rd party file systems under Windows - all you need is a driver. Auto-launch its installer from a mini-partition masquerading as a CD upon hookup, and your device doesn't even need to have any trace of FAT ...but, somehow, nobody cares to do that. So we're "stuck" with FAT (at least MS even contributed some rather decent standard in the area, Media Transfer Protocol)

IIRC win2k also supports zip folders, BTW.

Yes, this topic has come up before (and did so again in the last couple days)...

We do push the xz versions of our releases to the mirrors along with the zips, but we haven't yet provided direct links to them for simplicity on our get-haiku page. Perhaps we can rework the download page in the coming days to make this easier for users.

You are correct - it would save > 100mb per download, which would be a nice gesture.

Thanks for the feedback, and we'll try to improve this soon.

Also, for anyone reading this: Keep an eye out for another update on the website in the next 24 hours.

The sarcastic bastard in you should send them couple bucks in donations or stfu. It will come very handy when the beta time comes next year. Thank you very much.

That wouldn't be very bastardly of me now, would it?

Maybe he did - 12 years ago. Maybe 12 years is a long time to wait? Not to belittle the project, but it still only really does what the BeOS on my PowerMac does, and that is 12 years old. I have BeOS installs that are older than the Haiku project that still do as much, if not more.

Having said that, I have the utmost respect for the Haiku team, I fully support their effort and eagerly await R1. What I don't respect is people like yourself, who come and abuse old timers because you think you know better than us. Believe me, I've used BeOS for a lot longer than 12 years, having owned a BeBox and used the DR releases when "new".

I have a better idea... why not just close down the comments section and replace it with a Paypal link for every story?

There is no reason for him to STFU - he is commenting in the comment section of a website. What's the problem - that he is being snarky? So what? I laughed at his comment. It improved the quality of my life.

12 years since OpenBeOS project was started. Serious code committing started a bit later, though.

It ain't easy to rewrite BeOS from scratch having very small dev team, working purely on fan-drive and dedication, especially when everyone else just don't give a damn about your project. That's fine, there were times when Linux was as obscure as Haiku now, you know, without millions of dollars, big sponsors, huge army of developers and mainstream acceptance.

And, if I remember correctly, the serious coding began again after Thom published an article on this site that tended to prod the developers into increasing the coding activity.

Can I ask how you got Clementine to work? I would love to run that on Haiku.

You might not feel confident unless you understand that Haiku's alpha version quality is far better than other softwares' beta or even release versions.

Well, then don't switch. I've never understood this weird idea that your choice of OS/desktop/application should only ever come down to which one has the most features, as though the design or feel of it doesn't matter. Some people prefer the way that Haiku is designed and how it feels to use it. They'll use it even if you wont.

All that blurriness makes my eyes hurt.
Nice that netpositive finally have a good rendering engine though.

Is the blurriness you're referring to in regards to the resolution of the video? It's an HD YouTube video. Please select a higher resolution (1280x720).

Oh yeah, that helps.

I hope you'll post your issue on the Haiku bug tracker (if you decide to become a little less lazy).

I love reporting Haiku bugs. The developers are nice to work with. It may take a couple days or a couple months, but the bugs I report always get fixed.

Maybe.. but there are so many factors. I could be that I was booting from an SD card, could be that the Wind has 2GB RAM stick, with 1GB also being soldered on the mother board (disabled), could be that the SD card was crap (was free with a Samsung Android phone.)

I'll try booting with VMWare... (though can't do that from same image... FAIL!).. err.. this laptop after I next reboot..

Edited 2012-11-13 13:54 UTC

Similar hardware here (Aspire One) and alphas 3 & 4 boot from a usb stick dded with the anyboot image. Crashes booting from usb cd-rom.
After alpha3 there's been a minor gma950 regression as the native resolution must now be set manually, 1024x600 @ 70Hz for some reason.

Mine is booting from an SD card using the Anyboot image. Might just be the SD card's fault. I'll probably try another at some point.. or try writing the image on the same device (used another laptop and the suggested windows based DD alternative.) The Wind is reasonably old now (2008?) and my kids tend to use and abuse it, so it might just be a hardware issue. I think the AHCPI is on too in the BIOS, so I might try turning that off...

I have it running in VMWare 5, hosted on Windows 7 64 bit.

But I bet you haven't booted the VM from an SD card, right? That's the issue...

Edit: decided to use the Anyboot to create a new VM with VMWare as the image booted fine using it as a "CD". Runs nice!

Edited 2012-11-14 12:02 UTC

No, it was running from the hard disk.

I mean - I wrote the image to an SD card and ran it native on the laptop. VMWare player won't boot from a physical hard drive AFAIK.

In my case I was reading the ISO image from the hard-disk.

Then I installed the live ISO into a newly created VMWare machine.

Should have been a bit more clear.

Aha!! This is the thing blocking the boot process, this describes exactly what I'm seeing:

http://dev.haiku-os.org/ticket/9145

Okay - I debugged the boot and got to KDL whilst it was trying to do something with the registrar/filesystem/mimetypes. Corrupt FS? No idea. Not going to go any further as it doesn't really interest me much. "cont" in KDL started the boot back up, but went to a black screen as mentioned above.

I booted the VM image in VirtualBox and....nothing but the blue desktop. No deskbar, no nothing.
Guess I'll give the ISO a try.

Edit: Spoke to soon. Enable NX, more ram and video memory and now it runs.

Edited 2012-11-14 04:03 UTC

Wow! Chris Herborth! cool :-)

Lies! I was never cool. ;-)

My copy of Corum 3 tells a different story ;-)

First they have to make gcc 4.7 available in the system as default compiler.

Qt5 is more C++11 friendly.

the default one (that you can set with setgcc gcc4) is 4.6.3 right now, if i remember correctly. Upgrading to 4.7 is not a big deal. The haiku devs dont like to upgrade the compile short time before the release, but I'm quite sure, it wont take much time, and gcc will be updated in the nightlies to 4.7.x
That's what i like at haiku, that fact, that they have a quite up-to-date compiler. You can even use clang if you like.

I'm currently working on this a bit right now. As things stand, the Haiku sources need some changes here and there to compile with GCC 4.7.2. I'll be sure to get them all sorted out before I commit the updated buildtools. That said, the jump to GCC 4.7.2 isn't quite as trivial as I initially expected, so it may take me some time to make sure I get it into a working state that can be committed.

Luckily, for quite a few people their PC is now mostly a WWW terminal, plus the basics of ~office suite, media player, IM, BT. Haiku should cover such needs relatively easily, eventually.

But I doubt it has much chance at mobile. First, it's a very desktop-focused OS. Second, that battle is probably already won, or nearly so (it's an open question if there's really a place for a third mobile ecosystem - and if yes, that will be probably MS...)

I think we will see a majority of the "basic users" moving to tablets in the next few years.



Yes, there are two leaders in the mobile world at the moment. But I don't think that it's impossible to gain market share there. People are not looking at phones and tablets the same way they look at desktops. In fact, they seem to be switching between ecosystems without much thought about it as long as the basic apps are there. And getting the basic apps on a mobile device at this stage is much easier than on the desktop given the simple nature of them.

I don't really see how haiku is any more desktop oriented than Windows 8 is. Sure, you'd need to remake the app server and add the hardware support. But apart from that it should make a fine base for a mobile os I think.

It's desktop oriented in the sense that its very small dev team targets ~desktops almost exclusively (and it always targeted them, also as BeOS - its web appliances didn't go far, and were almost desktops anyway).

And in the sense that it's otherwise relatively unremarkable - what it would offer for mobile, that would make a tangible difference? (IIRC somebody once said "media handling" ...but present mobile OS do that good enough already)

As I implied earlier, it would have to be done by a dedicated team. I actually think that Be was on the right path with the focus shift, although they were much too early.



Not much really. Except for the fact that it's already a clean and lean little OS that is well suited for the task. Much more so than Linux in my opinion.
But other than that the question would really be "how can we make this different?". I think that there's plenty of room for innovation in that market if you are up to it.

That said, I'm not arguing that it's something that must be done. Just saying if anyone had the desire to make it big, that would pretty much be the only and last chance.

I don't know ...a variant of Linux is ready - it's actually shipping, and clearly good enough to be massively popular. I'd say that's being far better suited for the task than any hypothetical Haiku offshot. :p


I think that question isn't ultimately that important, WRT market forces (yes, we can say "unfortunately isn't that important" - but we, those who even know about BeOS or Haiku, hardly matter here). If even Microsoft being the "3rd ecosystem" is not certain, there's much less hope for any smaller players...

I suspect the innovation might happen, from now on, mostly within the established ecosystems - roughly like it was with the PC: look what immense improvements were brought by its economies of scale, in the last 3 decades (meanwhile, all the more ~proprietary or niche platforms died out - even Macs are just PCs underneath, really).
BTW, overlooking that was a major error of Be, with how they self-exiled themselves into ~premium markets (and so x86 port of BeOS came too late, when win95 & 98 already took over)

My point was that you need a pretty powerful device to get a decent Android experience.
I don't know how much linux is to blame for that from a technical standpoint, perhaps not as much. It could just as well be cultural thing. The end result is the same none the less.



I depends a lot on if it gets a powerful hardware manufacturer to stand behind it. That was the only thing that made Android a success. What would it be without Samsung?

No-one has the same kind of grip that MS had on the desktop. At least not yet. The situation is very different from that standpoint.
The fact that pretty much no one has a relationship to BeOS is just an advantage in my opinion really.

If I was to choose a foundation for a mobile OS today I think Haiku would be among my first choices. In fact, that was my first thought when I saw the iPhone for the first time. "Hey, Haiku would be perfect for that thing"

Again, I'm not saying that it would be an easy task, and I'm not saying that its even needed. But the chance of ever gaining a user share other than old BeOS nostalgics on the desktop is non existent. If someone would want it to become used by people it would have to be on mobile devices, televisions and such.

I don't think it matters much. As I hinted in the previous post, the mobiles are undergoing what the PCs did in the decade starting in mid-90s, a "Moore's Race". Lack of ~performance won't be an issue, you can say it isn't already (the lowly-end getting 1 GHz Cortex A8: http://en.wikipedia.org/wiki/MediaTek - this SoC manufacturer is what powers http://en.wikipedia.org/wiki/Shanzhai )

Plus, it's definitely not so much about Linux itself - there were pre-Android mobile phones based on it, with modest hardware, working fine (for example: http://en.wikipedia.org/wiki/Motorola_Rokr#E2 ). It's just that the world chose lately a slightly "heavy" layer on top of Linux, in the form of Android ...which also gives great capabilities, so it's a good trade I guess.


Oh I would think quite a few Android manufacturers would love that, since they could be more easily on top and/or not struggling for survival. But I doubt it would be much different for us - Android seems to be what the world wants, it would prop up some other manufacturer if Samsung didn't pick it up.

BTW, Huawei and ZTE are also riding successfully on the rise of Android, ZTE is by now probably the 3rd largest mobile maker.


But MS didn't start from such position, they earned it in early-to-mid-90s by being the best option (among the all-somehow-bad): http://www.osnews.com/thread?522221
Haiku now would be already in a similar position as BeOS then. Haiku/BeOS pre-iPhone, maybe ...but Apple bought Next, not Be, so that closed the possibility here. As for Android, WP - those are made for developer familiarity, where Haiku is also largely out of the question.


Maybe, maybe not - sometimes I see it doing well on low-powered ~netbooks (what LXDE apparently also targets): mostly a WWW terminal, plus some ~office, IM, media player.

As for ~mobile ...IMHO you overlook what an immense amount of work would need to happen before Haiku would be useful on such - and without any guarantee it would be even on par with read-for-the-taking Android
(or... why not OPIE, or GPE, or Maemo? All open source)

Edited 2012-11-16 06:17 UTC

Well, in the grand scale (looking at the estimated time period between the big bang and the heat death of the universe), even our civilisation lasts a mere blink of an eye... :p

True...

Funny, it boots on a wide array of hardware for me - including an i5 Thinkpad T420 and my core2duo box here...

Hopefully you're using the Anyboot image dd'd to that USB stick of yours, and not the raw image.

And - you have opened several tickets demonstrating the issues you're running into I assume.

Try disabling ACPI. Works for me on most hardware that doesn't boot haiku right away.

"Haiku's ACPI support, which is enabled by default, might cause problems on some hardware. ACPI can be disabled in the boot loader's safe mode options menu."
https://www.haiku-os.org/get-haiku/release-notes