BeOS lives on, in the new Haiku OS Beta 4

Posted:
in macOS
Haiku OS, a modern clone of BeOS, is an interesting look back at what Apple once considered to advance its Mac operating system.




In 1995, Apple's head of Apple France, Jean-Louis Gassee left Apple to found a new company, Be Inc.

Be went on to develop the BeBox, a PowerPC-based computer, and the BeOS -- an ultra-fast modern media OS featuring fast disk I/O, rendering, and kernel.

Apple, and BeOS

In 1997 when Steve Jobs returned to Apple, and Apple was in need of a new operating system, the competition was between NeXTStep developed at Jobs' other company, NeXT, and BeOS.

Apple chose NeXT, buying the company. Over the next few years NeXTStep was modified to become Mac OS X, and ultimately iPhone OS 1.

But today, BeOS lives on in the Haiku Project -- an open-source API-compatible port of BeOS. A dedicated group of developers have been quietly working on Haiku for over fifteen years, toiling in their spare time, and the project is finally coming to fruition.

The Haiku Project just released Beta 4 of Haiku OS and it promises to give both macOS and Windows a run for their money once released.

Haiku features all the same performance improvements of the original BeOS, and improved support for modern hardware. It's also astonishingly fast, booting and shutting down in just a few seconds on most machines.

Haiku B4 is compatible with most modern and older hardware including most Intel and Apple Silicon systems, Macs, PCs and ARM-based computers. Given its provenance, it will run quite well on, say, an older Celeron laptop from the Windows 7 era.

Since it's beta software still, you may want to install B4 on a spare machine or hard drive. In any event if you do install it, be sure to back up all your data first.

Read on AppleInsider

Comments

  • Reply 1 of 14
    dewmedewme Posts: 5,361member
    I don’t want to take anything away from the dedicated team working on this but it looks like it is about 25 years beyond its point of potential commercial relevance if their intention is truly to take away share from Windows or macOS, I.e., “give them a run for their money.” In macOS’s case, there is no money to be taken away since Apple doesn’t charge for macOS. Maybe the screenshots simply don’t do it justice, but the UI looks like vintage Windows 95 or Windows NT 4 era stuff. Lots and lots of flat haze gray dialog boxes and forms and big buttons.

    Then there’s the issue of a what is imho a vastly more realistic competitor: Linux. Forget about Windows and macOS. There are so many compelling flavors of Linux. It can tailored to meet a huge variety of client needs, from near real-time embedded, to appliances, to full blown desktop clients that arguably give Windows and macOS a run for there money, at least on the UX side, for some folks at least, and for people who want to squeeze out a few more years of productive life from older platforms with limited resources. Some of the newest Linux client builds are very visually appealing and are very easy to use.

    Shall we talk about applications? Regardless of how you personally feel about apps/suites like Microsoft Office and Microsoft 365, if your platform doesn’t natively support these business critical apps, you are always going to be fighting an uphill battle and considered a second class citizen in the personal computing market. Despite the stability, robustness, and highly compelling UX on modern Linux distributions, they are still unable to scale or break through the impenetrable Office wall that surrounds corporate computer purchase agreements. OpenOffice and iWork are not slouches by any means, but they’ve had had many years to try to break through and still the Office wall stands firm and remains the chief “validator” that defines who gets to live in the top layer of the OS hierarchy. 

    If this incarnation of BeOS is really as rock solid and performant as they claim, they may have better success going after a narrower market like providing a better and more “open” OS for gaming consoles, industrial HMIs, set top boxes, streamers, single board computers, and even TVs. 
    edited January 2023 muthuk_vanalingamtdknoxwatto_cobraFileMakerFeller
  • Reply 2 of 14
    jeromecjeromec Posts: 191member
    "Haiku B4 is compatible with most modern and older hardware including most Intel and Apple Siliconsystems, Macs, PCs and ARM-based computers"
    I could only din compatibility with Intel/AMD 32 and 64bit.
    Is there really Apple Silicon/ARM compatibility? Where can I find the appropriate version?
    watto_cobra
  • Reply 3 of 14
    avon b7avon b7 Posts: 7,666member
    I was a big fan of Be as a candidate for a new Mac Operating System. A candidate that would have needed massive investment and lots of engineering to make it serve its purpose as a consumer oriented system. 

    Apple went with NeXT and I was a massive fan of Avie and that system also needed a ton of money and other resources thrown at it and was to serve the following 15 years. The original OSX was in no way ready for prime time on release and it wasn't until Tiger and then Snow Leopard that things settled down. 

    So 15 years for Haiku to reach beta 4 may be a landmark of sorts but it can't challenge Windows or macOS where it matters. 

    More to the point, even Windows and macOS aren't really ready for the next 15 years at this point either. 

    The future is IoT and AI. Everything connected to everything with virtualisation of hardware, distributed security, distributed file systems etc. 

    I very much doubt that the innards of Haiku have been developed with that in mind and that is what is going to be needed. 

    muthuk_vanalingam
  • Reply 4 of 14
    maltzmaltz Posts: 453member
    Better headline:

    BeOS lives on
    in the new Haiku OS
    Beta version 4
    mknelsontdknoxStrangeDaysdjames4242stompywatto_cobrafahlmanFileMakerFeller
  • Reply 5 of 14
    Obviously if Apple would have adapted BeOS, it wouldn't look anything like what Haiku looks like today on a code-level.
    macOS and iPadOS are quite old by now, which is why Google's Fuchsia is so interesting: they started from scratch with modern concepts, replacing Android.
    I wonder if Apple at some point feels the need to do the same with their operating systems.
    williamlondonwatto_cobra
  • Reply 6 of 14
    dewmedewme Posts: 5,361member
    Obviously if Apple would have adapted BeOS, it wouldn't look anything like what Haiku looks like today on a code-level.
    macOS and iPadOS are quite old by now, which is why Google's Fuchsia is so interesting: they started from scratch with modern concepts, replacing Android.
    I wonder if Apple at some point feels the need to do the same with their operating systems.
    Operating systems haven’t fundamentally changed because the underlying physical hardware architecture hasn’t fundamentally changed. All manner of optimization has taken place in hardware implementations, but it’s all been done around the same fundamental architectural design that is many decades old.  

    One of the primary role of the operating system is to provide an abstraction, i.e., a virtual machine on top of the underlying physical machine. There must be a very high degree of synergy between the operating system and the underlying hardware architecture to deliver the best overall system utility and performance. With Apple designing its own silicon they can take this synergy to greater levels and take advantage of things like downclocking and efficiency cores to improve system attributes like battery life.

    I would think that Apple views its ongoing OS investments as being very relevant, dynamic, and constantly evolving to take advantage of the capabilities of newer hardware. Apple’s macOS, iOS, iPadOS, etc., may be chronologically “old” but each new release brings in a lot of new code and modern designs, along with some degree of refactoring. If Apple was resigned to sticking with the “old” they wouldn’t be dropping support for older platforms quite so often.

    On the other hand, there is still a lot of old code in there, as evidenced by the occasional latent security issue that is discovered. The big question for software architects and product owners is whether it makes sense to refactor some of the older constructs when you know that in all likelihood you’ll end up injecting some new anomalies. The old code may have been ugly, but like some ugly old roofs on houses, it worked and didn’t leak. If I had to guess I would say that Apple primarily refactors using a reactionary approach. If an anomaly or security issues pops up in a specific area they probably do a root cause analysis and consider whether refactoring is needed in specific areas.

    Who really knows? Apple may have a stealth project underway to build a totally new OS based on the assumption of some stealth new hardware architecture that diverges significantly from anything that’s already been done. After all, they now own the silicon and aren’t beholden to anyone. They own their own dev tools like optimizing compilers that allow them get the most from their hardware and software. They have a solid basis for putting together their own programming language to make all of the benefits of their hardware and software stacks more easily accessible to application developers. Few personal computer manufacturers have the flexibility that Apple has to change the fundamentals.
    edited January 2023 danoxJinTechtdknoxwatto_cobra
  • Reply 7 of 14
    thttht Posts: 5,441member
    dewme said:
    Who really knows? Apple may have a stealth project underway to build a totally new OS based on the assumption of some stealth new hardware architecture that diverges significantly from anything that’s already been done. After all, they now own the silicon and aren’t beholden to anyone. They own their own dev tools like optimizing compilers that allow them get the most from their hardware and software. They have a solid basis for putting together their own programming language to make all of the benefits of their hardware and software stacks more easily accessible to application developers. Few personal computer manufacturers have the flexibility that Apple has to change the fundamentals.
    Sounds like xrOS or rOS has taken up a rather large chunk of their operating system team resources, and if rumors are right, it's expanding into the frameworks and product teams now. I'd imagine this OS is under the similar kind of restrictions that Mac OS X was when they designed it for the iPhone. Latency is going to be a vomit inducing problem for the goggles, literally, and I wonder what they are doing to guarantee latency with it. Heck, some fraction of the market will be nauseous with zero latency.

    Anyways, I don't think there are any interesting issues or problems holding operating systems back at the kernel or userland level anymore. It's all in the app frameworks and UI now, and even there, there isn't that much. The biggest issues I have are still how to find all the stuff I have.

    So, something something automatic meta-data and natural language interfaces for search and help would be an interesting feature. That's "feature", not some fancy core operating system thing.
    watto_cobraFileMakerFeller
  • Reply 8 of 14
    JinTechJinTech Posts: 1,022member
    avon b7 said:
    I was a big fan of Be as a candidate for a new Mac Operating System. A candidate that would have needed massive investment and lots of engineering to make it serve its purpose as a consumer oriented system. 

    Apple went with NeXT and I was a massive fan of Avie and that system also needed a ton of money and other resources thrown at it and was to serve the following 15 years. The original OSX was in no way ready for prime time on release and it wasn't until Tiger and then Snow Leopard that things settled down. 

    So 15 years for Haiku to reach beta 4 may be a landmark of sorts but it can't challenge Windows or macOS where it matters. 

    More to the point, even Windows and macOS aren't really ready for the next 15 years at this point either. 

    The future is IoT and AI. Everything connected to everything with virtualisation of hardware, distributed security, distributed file systems etc. 

    I very much doubt that the innards of Haiku have been developed with that in mind and that is what is going to be needed. 

    Avie was amazing! I remember being hooked on his blog. I think it was an email? Even though I’m not a programmer, and know nothing about the backend, I’m fascinated with UNIX and the macOS. 
    watto_cobra
  • Reply 9 of 14
    rezwitsrezwits Posts: 879member
    Looks amazing, and pretty good for a Beta 4.  Definitely a VMWare install... to mess around with for a while.

    Always loved the icons!
    watto_cobra
  • Reply 10 of 14
    jeromec said:
    "Haiku B4 is compatible with most modern and older hardware including most Intel and Apple Siliconsystems, Macs, PCs and ARM-based computers"
    I could only din compatibility with Intel/AMD 32 and 64bit.
    Is there really Apple Silicon/ARM compatibility? Where can I find the appropriate version?
    There isn’t really an ARM version. It’s a work-in-progress: https://www.haiku-os.org/guides/building/port_status/
    watto_cobraFileMakerFeller
  • Reply 11 of 14

    JLG left apple France in 1990.

    Also haiku can be ran from a USB drive without installing to a hard drive, to check for hw compatibility.

    No one is really using it as their daily os yet,.most use it as their tinkering/secondary box. Other than the browser situation, the OS is quite usable and stable.



    The future is IoT and AI. Everything connected to everything with virtualisation of hardware, distributed security, distributed file systems etc. 

    Right, but surely you find your standalone computer terminal not responsive at some points? Haiku is aiming to have a stable and extremely responsive gui system. The future for BeOS fans keeps getting brighter with multi CPU, multicore and hyper threading. Considering the efficiency of haiku, and how palmsource and ACCESS have done little with BeOS, set top boxes and iot/embedded are a good spot for haiku. Someone will eventually come along and buy a license or usage agreement.
    edited January 2023
  • Reply 12 of 14
    mattinozmattinoz Posts: 2,316member
    Obviously if Apple would have adapted BeOS, it wouldn't look anything like what Haiku looks like today on a code-level.
    macOS and iPadOS are quite old by now, which is why Google's Fuchsia is so interesting: they started from scratch with modern concepts, replacing Android.
    I wonder if Apple at some point feels the need to do the same with their operating systems.
    Wasn't one of the stated goal of Swift core design team to make a language they could be used to author the full stack from low level complier tools and OS core components all the way up to scripting with the REPL. 

    Add the work of the Swift Server group and ever expanding team in the swift open source project they seem to be inching closer and closer very slowly towards a full OS that could rival DarwinOS even maybe replace it under Apple various OS's. That to me would be the what they are investing in but who knows they could other teams working other angles as well to suit various projects. 
  • Reply 13 of 14
    dewme said:
    I don’t want to take anything away from the dedicated team working on this but it looks like it is about 25 years beyond its point of potential commercial relevance if their intention is truly to take away share from Windows or macOS, I.e., “give them a run for their money.” In macOS’s case, there is no money to be taken away since Apple doesn’t charge for macOS. Maybe the screenshots simply don’t do it justice, but the UI looks like vintage Windows 95 or Windows NT 4 era stuff. Lots and lots of flat haze gray dialog boxes and forms and big buttons.

    Then there’s the issue of a what is imho a vastly more realistic competitor: Linux. Forget about Windows and macOS. There are so many compelling flavors of Linux. It can tailored to meet a huge variety of client needs, from near real-time embedded, to appliances, to full blown desktop clients that arguably give Windows and macOS a run for there money, at least on the UX side, for some folks at least, and for people who want to squeeze out a few more years of productive life from older platforms with limited resources. Some of the newest Linux client builds are very visually appealing and are very easy to use.

    Shall we talk about applications? Regardless of how you personally feel about apps/suites like Microsoft Office and Microsoft 365, if your platform doesn’t natively support these business critical apps, you are always going to be fighting an uphill battle and considered a second class citizen in the personal computing market. Despite the stability, robustness, and highly compelling UX on modern Linux distributions, they are still unable to scale or break through the impenetrable Office wall that surrounds corporate computer purchase agreements. OpenOffice and iWork are not slouches by any means, but they’ve had had many years to try to break through and still the Office wall stands firm and remains the chief “validator” that defines who gets to live in the top layer of the OS hierarchy. 

    If this incarnation of BeOS is really as rock solid and performant as they claim, they may have better success going after a narrower market like providing a better and more “open” OS for gaming consoles, industrial HMIs, set top boxes, streamers, single board computers, and even TVs. 
    Office365 web apps have worked on haiku since about 2008. They didn't run well, however.

    The native browser runs much better in R4.
    edited January 2023
Sign In or Register to comment.