New OS for PowerPC Macs

[narke]

Hi,

I'm writing an operating system for PowerPC Macs in order to be able to reuse them with an OS adapted to the modern world as far as possible. It will be a desktop OS taking into accout the limitaions of RAM and CPU speed of PowerPC Macs.

If you are interested you can back the project or ask me questions.

Here is the kickstarter campaign link: https://www.kickstarter.com/projects/1691639619/einherjar-operating-system-for-powerpc-macs

And github: https://github.com/narke/Einherjar

 

[Schmoburger]

Will follow this post I think out of curiosity.

I should point out that there are already options available in the world of Debian-based OS's that are well supported and perform very efficiently, and a fair bit of software available and maintained for OSX from independent developers such as the TenFourFox port of Firefox for instance. Tge big issue with most of these programs is that they don't work on OS versions prior to 10.4, which performs rather poorly on slower machines such as G3 and G4-based ones, has very poor graphics support for stock hardware as it depends upon the functions of newer stuff to run optimally, and due to all the bloat, it is very RAM-hungry. I know a lot of people who would love to use OSX still on their old hardware if possible, but due to the poor performance of 10.4 and 10.5 on lesser systems, they are forced to either backpedal to OS9 to regain software support (as there is plenty of stuff still out there for OS9 due to it's cult following), or look at other alternatives usually in the form of Linux if they wish to maintain a usable machine. If not, they must glug along with 10.4, which is painful on say a 400MHz machine, or simply retire an otherwise good machine... 10.3 on the other hand was fantastically fast and stable, but  for whatever reason it hasnt aged well with regards to continued software development in it's obselescence, unlike it's successors.

I have just yesterday gone down the latter route by installing MintPPC Linux on my Sawtooth, as it is almost as fast as  Panther, yet is still fully functional, unlike Panther in 2015.  However it is really in no way like OSX in it's UI... In fact is is very close in basic UI function and design to Windows. I am fine with this, perfectly fine with it, but when I remember how nice OSX used to be on such a machine, I do kind of yearn for it, and I know I'm not the only one. Not only that, in switching to Linux, anything Mac will no longer run on it, so I have to get other programs to do the job, and well... some just dont do it as elegantly as their Apple counterparts.

My point in a nutshell? Well, basically, there are quite a number of people that would probably be very receptiive to an OS that is similar in function to X, and able to use OSX software, but which does not have all the unneccessary bloatware of the semi-supported variants... Obviously for legal reasons you cant just chop and change OSX into something else, but if you were to make something that more closely resembles it, performs a whole lot better, and has some cross compatibility with currently available OSXPPC software, it would probably be a hit, and probably drive further development as well.

On a more general note of encouragement, I know there are going to be those who beleive the idea of developing anything for 10 year obsolete hardware is a moot point, and that the PPC platform has it's critics... whilst these critics do in some cases make some valid points and judgements, the fact remains there are people still developing software for Amiga and Atari after a couple of decades due to the demand from enthusiasts and afficionadoes of the platform, and there are more than a fleeting minority who reap the benefits of the same kind of continued unofficially sanctioned development for the "obsolete" PPC platform. As I have said many times, whilst 64-bit PPC is not necessarily a monstrously powerful platform by todays standards, and it did have it's implementation follies that probably went a long way to putting the nail in it's coffin very succinctly, it is far from useless... Whilst a Powermac G5 for instance may not have a million cores, it isnt slow, and the last versions addressed the issues of reliability that plagued the earlier designs, and is still more than capable of meeting and even exceeding the needs of many users. The only reason it is obsolete is because Steve Jobs decided it was and actively went out of his way to bury the G4 and G5-era of PPC in Apple's past by dropping support for such machines arbitrarily in software updates one after the other, and encouraging rather forcefully the major software houses to do the same when developing new versions of their titles. So for what it's worth, I do give this idea a tick of approval if it is able to get off the ground.

 

[techknight]

 and it did have it's implementation follies that probably went a long way to putting the nail in it's coffin very succinctly, it is far from useless... 

wut? BIG WORDS!! I HATE BIG WORDZZZZ

 

[Gorgonops]

Just going to toss this out there: Here's the website for the parent project. This OS has been a research toy since 2001, and it still doesn't have any more than a prototype GUI on its best supported platform, i386. There doesn't appear to be essentially any third party applications and the OS architecture differs substantially enough from UNIX (no "fork" or "exec") that making ports of most Open Source software is going to be rather difficult. (Not impossible to be sure; my ignorant reading of how the OS handles processes makes it sound vaguely similar to Windows and the POSIX compatibility layers for that... kinda work, but it certainly isn't going to be a simple "just recompile it" port. The system is also going to be lacking X which means any GUI software is going to pose some interesting challenges.)

In that light the schedule described in the Kickstarter seems... optimistic, but that's all I'm going to say.

 

[ScutBoy]

I looked at the Kickstarter as well.

Gorgonops, your assessment of the schedule is much kinder than mine. I started out at "effin crazy" and went downhill from there.

 

[johnklos]

Interesting and ambitious. I don't mean to sound like a skeptic, but this appears to be a multiyear project at minimum.

Even if the OS was ready and framebuffer access were functional, you'd still have to port the kind of software that people would be interested in running, like Firefox, which by itself is a huge pain in the ass. Large frameworks like Qt are going to require tons of work.

Personally, as I start to move away from modern OS X (I gave them a chance, and 10.11 just didn't cut it), I'll move to more minimalistic software, like Midori instead of Firefox. Do you think there will be much interest in a more minimalistic system as opposed to yet-another-GUI-running-Firefox? Because even if this new OS were up and running right now, it wouldn't be able to make up for the shortcomings of bloated "modern" software like Firefox.

 

[Gorgonops]

Gorgonops, your assessment of the schedule is much kinder than mine.

Well, yeah. I was sort of going with the "if you can't say anything nice don't say anything at all" strategy.

What we *really* need for PowerPC Macs is a working port of TempleOS... no, wait, I suppose we pretty much have that with MorphOS already.

 

[bse5150]

TempleOS is somewhat of a legend in the operating system community. Its sole author, Terry A. Davis, has spent the past 12 years attempting to create a new operating from scratch. Terry explains that God has instructed him to construct a temple, a 640×480 covenant of perfection. Unfortunately Terry also suffers from schizophrenia, and has a tendency to appear on various programming forums with a burst of strange, paranoid, and often racist comments. He is frequently banned from most forums.

lol

 

[Schmoburger]

TempleOS is somewhat of a legend in the operating system community. Its sole author, Terry A. Davis, has spent the past 12 years attempting to create a new operating from scratch. Terry explains that God has instructed him to construct a temple, a 640×480 covenant of perfection. Unfortunately Terry also suffers from schizophrenia, and has a tendency to appear on various programming forums with a burst of strange, paranoid, and often racist comments. He is frequently banned from most forums.

This has made my morning

But yes... I do tend to agree with the sentiment leaning towards biting off more than one can chew on a secondary glance at where things are at. Not intending to discourage development in any way, but well... One has to ask whether it will be a particularly worthwhile cause by the time it comes to workable fruition. I certainly do not have any intentions of ditching my PPC machines anytime soon, but I am sure that the numbers will dwindle slowly over the next 10 years... If for no other reason that hardware failures. Realistically, The G5 is the PPC that is worth coding stuff around, however I can iimagine the only G5-based machines left in the world in 10 years time will be Quad and dual-core PowerMacs... I'd say almost all of the earlier Powermac G5's and the iMacs would have found there way into rubbish bins because they were loads of crap from day dot. This makes it a better prospect to code stuff largely around 32-bit processors, meaning the capabilities of the remaining G5 machines will be not entirely made use of to the fullest potential, but well...How many developers will be enthusiastic enough even in 5 years time to keep coding new software from scratch, or going through hoops to rework existing code to port to a brand new operating system for a processor architecture that was considered expired 10 years ago by the majority of the world?

And all this is assuming that somehow development could leap into overdrive and get somewhere in that time which, well... let's be realistic, seems unlikely even at the hand of a handful of committed and semi-competent developers without corporate backing or financial drive. That's another thing also... who is paying for it all? It's all well and good to spend forever building an OS, but realistically in human terms, time is money. There is only so much time one can dedicate to something that doesnt keep lights on and mouths fed, and it's unrealistic to expect any major financial input from... well... anywhere. Much as the idea of a new OS for PPC seems cool, i doubt anybody would want it enough to pay for it when funvtional, pretty, quick and stable options already exist for free and have massive support. Fair enough, people still actively develop Linux, but well... there is a lot of cross-compatibility in the nix ecosystem, and Linux in all it's flavours has been developed collaboratively by hundreds, thousands even of developers worldwide for over 20 years. The hard work was done a very long time ago, and the idea has always been in a lot of cases to keep it simple and uniform rather than do things wildly differently... When you do things out of left field it best be something monumentally better than what curently exists, or you are likely to end up with a big white elephant.

 

[tendim]

I'm writing an operating system for PowerPC Macs in order to be able to reuse them with an OS adapted to the modern world as far as possible. It will be a desktop OS taking into accout the limitaions of RAM and CPU speed of PowerPC Macs.
If you are interested you can back the project or ask me questions.

I am a complete OS geek, and the timing of this post is most fortuitous as I've just started taking a deep dive into the NetBSD world.  But  have to ask: why another OS?  Linux has a strong following as an alternative as it is.  If you are looking for a non-Linux project to back, why not work on NetBSD instead?

 

[olePigeon]

I feel sorry for Terry.  Obviously a brilliant mind if it weren't for the schizophrenia.    I sure hope there's an actual cure for it soon so people like him can continue to do amazing work without all the problems.

 

[commodorejohn]

I am a complete OS geek, and the timing of this post is most fortuitous as I've just started taking a deep dive into the NetBSD world. But have to ask: why another OS? Linux has a strong following as an alternative as it is. If you are looking for a non-Linux project to back, why not work on NetBSD instead?

I can't speak for the OP, but: because *nix is boring and old hat. Whether it's Linux or BSD or something even more esoteric like OpenSolaris, "Unix clone" is the absolute most vanilla, uninteresting choice you can make when deciding to jump into OS development. At that point, you really have to ask yourself why you're bothering at all. Whatever we may think of the OP's plans or timetable, at least it doesn't look like he's going to inflict Yet Another Linux Distro on the world.

 

[Paralel]

I feel sorry for Terry.  Obviously a brilliant mind if it weren't for the schizophrenia.    I sure hope there's an actual cure for it soon so people like him can continue to do amazing work without all the problems.

Actually, the unique wiring of his brain may be what allows him to come up with interesting solutions to mundane problems, his brain just took it one step too far and it became a disability. There have been many instances where "fixing" a psychiatric disorder deprived the person of their exceptional talent as well. This is a serious problem when treating artists, musicians, writers, etc... that have mental illness. You have to balance solving the issue with not impacting their ability to create. It can be exceptionally difficult, sometimes impossible.

 

[narke]

Thanks for your reactions.

Having a critical view about the schedule is a fair point but this is a challenge which I would like to take.

As to why not to use Linux, I like OSdev and would like to have a low footprint desktop-based OS for PowerPC Macs.

And concerning applications, I don't plan to port resource intensive ones like Firefox or QT but instead NetSurf and FLTK, for example.

 

[CC_333]

I like the idea!

Is it to be some kind of command line-based thing, or will it be functionally more similar to the Classic Mac OS (i.e., more graphical)?

Having something Classic-like with preemptive multitasking, protected memory, and support for more modern applications and peripherals would be incredibly excellent. Perhaps these could be incorporated in a modular fashion, so they can be added or subtracted depending on the needs of the user?

Good luck! I hope you accomplish your goal!!

c

 

[Schmoburger]

I feel sorry for Terry.  Obviously a brilliant mind if it weren't for the schizophrenia.    I sure hope there's an actual cure for it soon so people like him can continue to do amazing work without all the problems.

The guy really is an absolute genius in every sense of the word... Even when you parouse his often extravagant ramblings, whilst he has some views that might seem challenging to those who consider themselves normal, it is clear when you read through the brash, confrontational manner in which he conveys his ideas, It's easy to see that he is a man who does an incredible amount of thinking, and has some fundamentally sound ideas about things as a whole. His thought process seems to operate on a very predominantly logical plain, with little regard for emotional considerations, yet still with a certain creative flair... which is what makes everything he does seem so bizzarre. But it works for him, at least as a developer if nothing else

It is a damn shame that his brilliant mind has, as Paralel put it, at some point or another entered the realms of being a disability, but again... treatment would entail treading a very thin line with people suffering this kind of torturous mental illness. on either side of this line, you have the status quo (The condition and the unbound brilliance it entails in the case of some. but also all the vices that come with it), or the distinct possibility of a normallish life being had at the expense of one's unfathomable talent possibly being subdued along with the illness.

Nonetheless, it is a damn shame that anyone must suffer the worst of schizophrenia or any other such severe mental disorder.

 

[olePigeon]

Actually, the unique wiring of his brain may be what allows him to come up with interesting solutions to mundane problems, his brain just took it one step too far and it became a disability. There have been many instances where "fixing" a psychiatric disorder deprived the person of their exceptional talent as well. This is a serious problem when treating artists, musicians, writers, etc... that have mental illness. You have to balance solving the issue with not impacting their ability to create. It can be exceptionally difficult, sometimes impossible.

Most of the brilliant people I've met are noticeably autistic to some degree, and I wouldn't suggest to change that for anything.  You're right, it's that kind of wiring that makes them so smart and their thinking so creative.  Maybe a cure is the wrong word, but perhaps a way to manage the more extreme and troubling aspects of it to make their lives more comfortable without hindering their thinking process.

 

[narke]

I like the idea!

Is it to be some kind of command line-based thing, or will it be functionally more similar to the Classic Mac OS (i.e., more graphical)?

Having something Classic-like with preemptive multitasking, protected memory, and support for more modern applications and peripherals would be incredibly excellent. Perhaps these could be incorporated in a modular fashion, so they can be added or subtracted depending on the needs of the user?

Good luck! I hope you accomplish your goal!!

c

Thank you.

I though about having Mac OS 9 styled GUI but I wonder if Apple wouldn't sue me for that, seeing all these patent troll stories that wouldn't surprise me.

Mac OS 9's GUI is one of my preferred but I settled for Enlightenment's look & feel.

 

[Paralel]

Apple is known for suing even over its really old stuff, I wouldn't chance it if I was you.