Maybe some people actually appreciate the availability of source code for the entire OS.
Also, If you have standardised on Linux, it is nice to have a single environment on all machines.
The only way Linux will run well on PPC hardware is if people actually use and develop it. A lot of people are interested in this, and I am among them.
My old iMac worked just fine with Linux, while OS X barely ran at all on it.
There is a lot about Linux to like, and making it run well on PPC and support Apple’s popular hardware is as valid a pursuit as making it run well on x86.
By your logic, nobody should bother developing Linux for the x86 platform because commercial OSes are available for x86.
You have obviously completely missed the entire point and purpose of the Free Software movement.
It is not enough to have software. We want to have free software, and are prepared to work to acheive it.
I suspect that there is more to running Linux on PPC 970 machines that just running them on Macs. The article goes into quite a bit of detail about IBM’s plans and the ability to use the chip in embedded devices.
Well, I’m not sure, but I’d be likely to put linux on a new Mac if I had one. I simply like it, and I’ve found oddities in OS X that I don’t like when I toyed with it on a friends laptop (none of them were big, just stupid things like no pager – I know you can dl one).
There’s also great interest in seeing an independent hardware platform for G5 and discussions are going in the Amiga community on how to get there. One way could be to use the Momentum PPC 970 Evaluation platform to create a real motherboard.
People who run linux (myself included) enjoy the ability to run commodity (plentiful and inexpensive for the quality) hardware. Buying a macintosh from Apple, with OS-X preinstalled does not meet that criteria. People complain about the microsoft tax on x86’s, it’s worse on Apples. At least I can go to the local computer shop and build an amd/intel box w/o me being required to buy an OS I don’t want.
I would myself love to be able to use linux on a powerpc box, I don’t see any reason to spend the money in this manner to do it though.
How come there is a Mac on Linux solution which lets you run OS X at near native speed in Linux for PPC but there is nothing on the x86 side which lets you do the same for Windows? Is there some kind of limitation in x86 which prohibits this from happening? It would seem there must be a limitation b/c if there wasn’t you would expect the much larger x86 Linux community to have already come up with an answer for this. I know of VMware and Bochs but they don’t count since they are emulators and much slower. Any answer?
From a design/technical point of view, the x86 CPU line is an ugly looking set of creatures. The one thing x86 has as a plus, is that being the most common, using x86 still offers the widest range of operating systems/software that you could run on it.
But software development is moving towards cross-platform. If I would ever move away from a x86 based PC, then PowerPC chips would be a good candidate.
@BlahCrapMadeUpName:
“Something tells me something north of 99% of linux work is done on x86/compatible. Any confirmations?”
Who cares? As long as that work doesn’t involve inline assembly code, it furthers Linux on all platforms it runs on, including PowerPC, right?
Thanks for pointing Win4Lin out. I’ve heard of it before but always figured it was an emulator like the others before reading their web site. It doesn’t really compare to Mac on Linux since it only allows the aging Win9x line to run with it. I can see it mainly being used to run a few needed apps like what Classic in OS X is used for.
Whats really needed is for the mass producing x86 mobo manufactures to get on board and mass produce just as cheap PPC mobo’s… SolTek, MSI, PCChips etc would be the ones to truely make this a viable solution.
An ATX motherboard that I can use as a drop-in replacement for my x86 one.
In the past I’ve run YDL on my G3 PowerBook (256MB). With GNOME, speed was comparable with OS X. With IceWM it was blazing fast.
Right now, what’s keeping OS X on the PowerBook is easy printing to my Epson Stylus C64, as well as easy faxing right from the OS X print dialog. Once I find something, and have the time to fiddle with printing/faxing from GNU/Linux, I’ll probably go back to IceWM.
Another thing that would be *really* nice is to be able to use dual monitors (the one built-into the laptop, plus an external flat panel). I’ve gotten used to it with OS X, and I’m too spoiled to want to go back to just one screen!
Whoops. I’m job searching right now, and expected all of you to be mind-readers. I’m printing and faxing resumes/cover letters every day, and don’t want to switch OS’s only to have troubles printing and faxing.
Apple is not the only company that sells PPC machines. I can’t picture buying Apple hardware and not having OSX installed to at least a partition. You can get white box PPC machines for well under Apple’s $2000 price point.
You would have to hang to run a PPC server on OS X, than to run one on Linux. Let me get a little creative and follow osnews’ theme this year.
/clears throat
Running Linux on PPC is provided for those who have concluded that the alternative OSes that run on PPC aren’t “desktop ready” for them.
Finally, geeks have put Linux on supercomputers, washing machines, vending machine, the PS 2, the Xbox, the Gamecube, the iPod, AMD64, Sparc box, x86 boxes, arm boxes, wrist watches, clocks, Opetron, Tivo boxes, Dvd players…the list goes on.
Did you seriously think they’ll overlook the PPC platform?
Free/OSS software is simply about building a software platform free from commercial and other constraints – giving everyone who obtains it the freedom to use it, modify it, redistribute it, and learn from it.
Whats difficult to understand about that?
Sure, you could say that only appeals to ‘geeks’ – and that ‘geeks’ are an irrelevant minority, but then why would you worry about what we might collectively do to the job market?
Plus, the idea that OS X is ‘the better OS’ is just not right. OS X might have a rather beautiful GUI layer and some nice APIs, but all the ‘good bits’ of OS X are closed and proprietary.
You can’t modify it, you can’t redistribute it, and you can’t learn from it. It’s so much less than it could be.
Your priorities w/regard to software are obviously quite different from mine, and I respect that. I was simply trying to answer your question as to why someone might want Linux on PPC.
As an example of why I don’t embrace OS X as ‘The better OS’ – I’ve seen it before – I look at A/UX and I see OS X 12 years ago – 12 years ago Apple had a UNIX OS with seamless MacOS 9 integration – capable of running UNIX and MacOS binaries.
Maybe you don’t remember that, or never even knew that – A/UX is long dead, and because of commercial licensing, will never be seen again.
OS X can suffer exactly the same fate – Apple falls on hard times and all that work gets dumped, and lost.
Look at the BeOS – So much potential, and good ideas – only just now resurfacing in Longhorn/Tiger. What a waste of time and effort.
AmigaOS – the amiga was revolutionary in it’s time. Now where is it? It certainly didnt die for lack of user support.
or NeXT – NeXT stagnated for 10 years before becoming MacOS X. Getting bought out by Apple was hardly a forgone conclusion – the ‘better’ OS could very well have never been, and all that work flushed straight down the corporate crapper.
I’m certainly not the oldest person in the computer industry, but I’ve seen enough to come to the conclusion that Open Source is, as far as I am concerned, the only sane way to move forward.
And if it leaves a smoking crater where a once-mighty commercial software industry stood, then so be it.
I’d like to point out to the rest of you posters that this was the only person who caught the drift of my original post and responded in a fashion that allows for real discussion before posting my response to him. That being said, thank you anonymous! (And to doggedblues, I didn’t mean to be hostile, I’m sorry you read my post that way.)
I see software through the eyes of someone who wants to make a living from it. I see the value in learning from other people’s code from the eyes of a student. What I have yet to see is a feasible way of combining the two. This is the crux of my difficulty in understanding the FOSS world.
To illustrate:
My job during the summer is that of a chemist. The products I design have a code consisting of chemical reactions, additives, and various components. In this world, open source is VERY bad. What good is my product if anyone can copy it? How does this help me ensure my survival in this world if someone else is out there taking advantage of my work? If someone out there is giving away my formulas and chemical reactions for free, that puts me out of a job! What good am I to a company when they have someone else doing my job for free?
That being said, I don’t see how this is any different in the software world. I’d like to stop this thread though because it doesn’t belong here. I am more than willing to continue this dicussion else where though if anyone else is willing to present the other side of the coin to me.
I’d like to ask you all one question that no one seems to have answered of mine yet, is there an advantage to the PPC platform that makes it desirable for Linux development that x86 cannot fulfill? If so, what is this advantage?
This is a bit off-topic, but it’s related to Yellow Dog Linux and PPC.
Mac is indeed NOT the only platform to run PPC Linux. For example there is the Open Desktop Workstation (http://www.pegasosppc.com/odw.php) which is based on the Genesi’s Pegasos PPC-motherboard. This is a G4-based computer that comes pre-installed with Yellow Dog Linux. This product is endorced my IBM and Motorola.
>I’d like to ask you all one question that no one seems to have answered of
>mine yet, is there an advantage to the PPC platform that makes it desirable >for Linux development that x86 cannot fulfill? If so, what is this advantage?
x86 is an old design that has had bits bolted on – time and again
– to retain backward compatibility.
Progressively more of the design effort, with x86, goes into limiting the effect of previous design decisions.
It adds upto: a convoluted instruction architecture, more silicon required (with attendant manufacturing challenges) and higher power/ heat levels in use.
Sooner or later x86 will be superceded – I, for one, would never have predicted that it would get this far.
As the open source technology takes a stronger part of mainstream then the dogma of particular chipset weakens.
The PPC series are a candidate with lower power, higher throughput per CPU cycle. Years ago I thought the ARM chips might make a dent …. these things are hard to predict.
“What’s really needed is for the mass producing x86 mobo manufactures to get on board and mass produce just as cheap PPC mobo’s… SolTek, MSI, PCChips etc would be the ones to truly make this a viable solution.”
The problem is to find a market which will buy large numbers of PPC motherboards. This is somewhat harder than finding a market that will buy large numbers of Macs.
The craziest thing here is that we would need Windows for PPC
in order to get some interest in PPC and therefore get low cost PPC board for Linux.
To be honest when I see how difficult it is for IBM to produce enough chips for Apple, I really doubt their ability to supply a large number of mobos manufacturers.
“Also, If you have standardised on Linux, it is nice to have a single environment on all machines.”
Your right. This is very important. Linux is platform independant, giving you the freedom to choose what hardware you like. So an x86 Gentoo user can switch to a PPC Gentoo box, and have a system that looks and works like the one he’s used to. That’s pretty nice.
“It is not enough to have software. We want to have free software, and are prepared to work to acheive it.”
Even besides that, I want choice. I, for one, appreciate the fact that buying an Apple computer does not bind me to using the Apple OS. Sure, for now, I CHOOSE to run OSX, but I have my option. Whether or not I ever switch over to Linux, I’m glad to know I CAN.
“Your right. This is very important. Linux is platform independant, giving you the freedom to choose what hardware you like. So an x86 Gentoo user can switch to a PPC Gentoo box, and have a system that looks and works like the one he’s used to. That’s pretty nice.”
But if it looks and works the same, why would he switch? What would be the improvement to justify the cost?
“The craziest thing here is that we would need Windows for PPC
in order to get some interest in PPC and therefore get low cost PPC board for Linux.”
…or Linux could gain significant market share. Once you’re on Linux, it’s not a huge difference from the user perspective to move from x86 to PPC (or other platforms, so long as you or anyone has made support for it). This would be one of the major improvements of Linux taking some large market share: not only would it break Microsoft’s hold on the software world, it would break Microsoft’s hold on the hardware world.
and some nice APIs, but all the ‘good bits’ of OS X are closed and proprietary.>>
As is Apple’s right. They laid out the $$$$ to make it. And have no obligation to give it away.
You can’t modify it,>>>
Thank God. 10+ years and “Team Linux” *still* can’t come up with a desktop that just effin works or software that has a straightforward universal non CLI install.
If it ain’t broke, don’t fix it.
you can’t redistribute it,>>
That’s because it’s not yours. However, you can distribute Darwin all you like.
and you can’t learn from it.>>
Oh, I think it’s possible to learn from OS X. Namely, *nix doesn’t have to suck rocks. Or that you don’t NEED or have to want to learn reams of “cthulu” to make the OS really useful on your desktop.
I don’t want to learn to config a video card. I want to learn to do cool things in P-Shop and FCP and Dreamweaver and Acrobat. (Actually that’s what I need to learn for my job.)
It’s so much less than it could be>>
Looking at rose briar (beautiful yet prickly) that is Linux, I’ll say that less is more.
Oh, and since you can throw on X11 and fink over any Linux program and windowing environment you want, OS X actually gives you more than Linux.
It’s sad that ARM has only become an embedded solution, it’s really a beautiful architecture. If Intel/AMD would pour as much resources into making these things fast as the do on x86.. wow..
I don’t know about the redistribution thing, but you certainly can modify and learn from, if not OSX, at least Darwin: the source is there for everybody (Yes, I know, Quartz’s source is not there, and it’s the only one that interests Free Software people).
“But if it looks and works the same, why would he switch? What would be the improvement to justify the cost?”
I don’t know, because you like the design of the Powerbook, or because the G5, according to some studies, anyway, is the “best bang for the buck”. Maybe you want to dual-boot into OSX, or maybe you just already have a Mac lying around?
Either way, I wasn’t really trying to get into a “Intel vs. IBM”, “Apple vs. PC”, or “x86 vs. PPC” type of argument. I was merely stating that one strong point of Linux is that it has support for a wider variety of platforms than many operating systems. That gives you more choices and more options. Also, it’s a good thing for the PPC architecture that it’s supported by Linux. Again, it gives you more choices and more options. Having more choices is rarely bad, even if you choose not to choose them.
So, it’s good for Linux and good for the PPC architecture. What’s good for the PPC architecture is good for IBM, especially since Apple isn’t the only company that uses PPC chips. In fact, I sometimes suspect that IBM pushing for Linux on the desktop is part of a plan to open a new market for their PPC architecture without having to go through Microsoft. As IBM is one of the big supporters of Linux, and things that make Linux good for IBM are going to increase IBM’s pushing Linux as the OS of choice, it’s probably good for Linux in that round-a-bout way, too.
Now, if you’re specifically anti-Apple, anti-Linux, anti-IBM, or generally against having extra choices, then maybe it’s bad for you, I guess. Otherwise, I’m not sure I see a problem.
But if it looks and works the same, why would he switch? What would be the improvement to justify the cost?
Perhaps you’ve hit on the point the original poster was trying to make? Since Linux supports various hardware platforms, this opens you up to making your hardware purchasing decisions based on things like cost and performance, instead of being forced into a hardware platform based on the operating system you want to run.
Hi Al, my Pegasos II G3 cost me much less than $900. I spent ~$350 on the board and CPU card. I bought it on http://www.pegasosppc.com in January 2004 and configured it in a case with memory, hard drive, graphics card for another $150. I used a keyboard and mouse from an old PC. It has been running since. I have partitioned the hard drive with Debian, Gentoo and MorphOS. I run Mac-on-Linux on Debian, but is is officially supported by Gentoo also.
As a linux user on Apple hardware, what I would relly like to have is a IBM branded laptop based on PPC architecture.
It would mean hardware support for linux and no MS-tax.
IBM builds PPC CPU and supports linux, so why not sell a machine devoted to linux on they very own processors instead than intel’s ones?
Apple Ibook lacks wifi support (as it has none PCMCIA cardbus only a usb external adapter is viable: not that portable-friendly), Powerbook are now simply to expensive without suspend-to-ram, video card full support (I blame ATI and NVidia, kudos to Benh) and the need to buy a PCMCIA wifi card (with an external antenna) while you get a useless Broadcom internal MINI-PCI nic.
The author mentions how some things of PPC Linux (Mainly their Yellow Dog Distro) just don’t work. Without basic desktop needs like Airport or 3D hardware compatibility, PPC Linux, or at least Linux for Mac does not meet standards to be used as a desktop machine. However, because of their extreme power and cheap cheap price, Linux PPC boxes are perfect server material.
> I don’t want to learn to config a video card. I want to
> learn to do cool things in P-Shop and FCP and Dreamweaver
> and Acrobat. (Actually that’s what I need to learn for my
> job.)
Well, that says it all really.
You feel your desire to learn only about the things that you feel are important are a justification for limiting my choice over what I should be learning about?
Maybe I want to know how a video card works, or the details of how my windowing system talks to it – is that something I shouldn’t know, because you don’t understand it either?
‘OS X is good enough for me – it must be good enough for you’ – is that what you are trying to say here?
> As is Apple’s right. They laid out the $$$$ to make it.
> And have no obligation to give it away.
Agreed. 100%. Thats why LinuxPPC exists. I don’t actually want to have to pay Apple to obtain the right to use my computer. I’m quite happy to pay them for the computer itself.
> If it ain’t broke, don’t fix it.
It is ‘broke’ – you can’t redistribute it, you can’t modify it, and you can’t learn from it.
Its very simple. The Free/OSS communities want to build a software environment that is free to use, free to redistribute, free to modify and free to learn from.
Your priorities are obviously different, and thats fine.
I’m not suggesting LinuxPPC is right for you, i’m not saying OS X is bad, or wrong.
But it’s not Free Software, which is why LinuxPPC exists alongside it.
If LinuxPPC is ugly to you, if it doesnt work for you, whats the problem? Nobody sold you a computer and told you that you couldn’t run anything but LinuxPPC on it, did they?
You’re saying LinuxPPC is ‘broke’ because it doesnt have a polished GUI layer, and doesnt run your apps, as if those reasons for it being ‘broke’ were somehow more valid than my ‘not free software’ reasons for OS X being ‘broke’.
Anyone is free to fix LinuxPPC’s ‘broke-ness’, you can’t say the same for OS X.
My job during the summer is that of a chemist. The products I design have a code consisting of chemical reactions, additives, and various components. In this world, open source is VERY bad. What good is my product if anyone can copy it? How does this help me ensure my survival in this world if someone else is out there taking advantage of my work? If someone out there is giving away my formulas and chemical reactions for free, that puts me out of a job! What good am I to a company when they have someone else doing my job for free?
In the same light, where would Newton be if he had given away his Calculus and the works of Optiks. Oh, wait. He did give them away! I think the question should be: where would we be had he not given them away. The scientific persuit shouldn’t be in the name of personal gain.
IBM did make RS/6000 laptops and PPC Thinkpads many moons ago, but they didn’t sell well. They sometimes turn up on eBay and use 603e processors. They’re built for AIX but you can slap Linux on if you like.
BTW, the iBook does support WiFi. Get an Airport card. Simple as that.
Somethign tells me something north of 99% of linux work is done on x86/compatible. Any confirmations?
I am wondering when the new version of yellowdog is going to be released anyone know?
Maybe some people actually appreciate the availability of source code for the entire OS.
Also, If you have standardised on Linux, it is nice to have a single environment on all machines.
The only way Linux will run well on PPC hardware is if people actually use and develop it. A lot of people are interested in this, and I am among them.
My old iMac worked just fine with Linux, while OS X barely ran at all on it.
There is a lot about Linux to like, and making it run well on PPC and support Apple’s popular hardware is as valid a pursuit as making it run well on x86.
By your logic, nobody should bother developing Linux for the x86 platform because commercial OSes are available for x86.
You have obviously completely missed the entire point and purpose of the Free Software movement.
It is not enough to have software. We want to have free software, and are prepared to work to acheive it.
You should have read the “recent related articles to find the answer:
http://www.osnews.com/story.php?news_id=7762
I suspect that there is more to running Linux on PPC 970 machines that just running them on Macs. The article goes into quite a bit of detail about IBM’s plans and the ability to use the chip in embedded devices.
Well, I’m not sure, but I’d be likely to put linux on a new Mac if I had one. I simply like it, and I’ve found oddities in OS X that I don’t like when I toyed with it on a friends laptop (none of them were big, just stupid things like no pager – I know you can dl one).
There’s also great interest in seeing an independent hardware platform for G5 and discussions are going in the Amiga community on how to get there. One way could be to use the Momentum PPC 970 Evaluation platform to create a real motherboard.
http://www.970eval.com/
I don’t know if it would work, but it’s an interesting idea.
People who run linux (myself included) enjoy the ability to run commodity (plentiful and inexpensive for the quality) hardware. Buying a macintosh from Apple, with OS-X preinstalled does not meet that criteria. People complain about the microsoft tax on x86’s, it’s worse on Apples. At least I can go to the local computer shop and build an amd/intel box w/o me being required to buy an OS I don’t want.
I would myself love to be able to use linux on a powerpc box, I don’t see any reason to spend the money in this manner to do it though.
How come there is a Mac on Linux solution which lets you run OS X at near native speed in Linux for PPC but there is nothing on the x86 side which lets you do the same for Windows? Is there some kind of limitation in x86 which prohibits this from happening? It would seem there must be a limitation b/c if there wasn’t you would expect the much larger x86 Linux community to have already come up with an answer for this. I know of VMware and Bochs but they don’t count since they are emulators and much slower. Any answer?
you mean like win4lin …..
to bad it only runs win9x series and not 2k/xp
From a design/technical point of view, the x86 CPU line is an ugly looking set of creatures. The one thing x86 has as a plus, is that being the most common, using x86 still offers the widest range of operating systems/software that you could run on it.
But software development is moving towards cross-platform. If I would ever move away from a x86 based PC, then PowerPC chips would be a good candidate.
@BlahCrapMadeUpName:
“Something tells me something north of 99% of linux work is done on x86/compatible. Any confirmations?”
Who cares? As long as that work doesn’t involve inline assembly code, it furthers Linux on all platforms it runs on, including PowerPC, right?
Thanks for pointing Win4Lin out. I’ve heard of it before but always figured it was an emulator like the others before reading their web site. It doesn’t really compare to Mac on Linux since it only allows the aging Win9x line to run with it. I can see it mainly being used to run a few needed apps like what Classic in OS X is used for.
Whats really needed is for the mass producing x86 mobo manufactures to get on board and mass produce just as cheap PPC mobo’s… SolTek, MSI, PCChips etc would be the ones to truely make this a viable solution.
You took the words right out of my mouth.
An ATX motherboard that I can use as a drop-in replacement for my x86 one.
In the past I’ve run YDL on my G3 PowerBook (256MB). With GNOME, speed was comparable with OS X. With IceWM it was blazing fast.
Right now, what’s keeping OS X on the PowerBook is easy printing to my Epson Stylus C64, as well as easy faxing right from the OS X print dialog. Once I find something, and have the time to fiddle with printing/faxing from GNU/Linux, I’ll probably go back to IceWM.
Another thing that would be *really* nice is to be able to use dual monitors (the one built-into the laptop, plus an external flat panel). I’ve gotten used to it with OS X, and I’m too spoiled to want to go back to just one screen!
> Once I find something
Whoops. I’m job searching right now, and expected all of you to be mind-readers. I’m printing and faxing resumes/cover letters every day, and don’t want to switch OS’s only to have troubles printing and faxing.
Apple is not the only company that sells PPC machines. I can’t picture buying Apple hardware and not having OSX installed to at least a partition. You can get white box PPC machines for well under Apple’s $2000 price point.
You would have to hang to run a PPC server on OS X, than to run one on Linux. Let me get a little creative and follow osnews’ theme this year.
/clears throat
Running Linux on PPC is provided for those who have concluded that the alternative OSes that run on PPC aren’t “desktop ready” for them.
Finally, geeks have put Linux on supercomputers, washing machines, vending machine, the PS 2, the Xbox, the Gamecube, the iPod, AMD64, Sparc box, x86 boxes, arm boxes, wrist watches, clocks, Opetron, Tivo boxes, Dvd players…the list goes on.
Did you seriously think they’ll overlook the PPC platform?
Free/OSS software is simply about building a software platform free from commercial and other constraints – giving everyone who obtains it the freedom to use it, modify it, redistribute it, and learn from it.
Whats difficult to understand about that?
Sure, you could say that only appeals to ‘geeks’ – and that ‘geeks’ are an irrelevant minority, but then why would you worry about what we might collectively do to the job market?
Plus, the idea that OS X is ‘the better OS’ is just not right. OS X might have a rather beautiful GUI layer and some nice APIs, but all the ‘good bits’ of OS X are closed and proprietary.
You can’t modify it, you can’t redistribute it, and you can’t learn from it. It’s so much less than it could be.
Your priorities w/regard to software are obviously quite different from mine, and I respect that. I was simply trying to answer your question as to why someone might want Linux on PPC.
As an example of why I don’t embrace OS X as ‘The better OS’ – I’ve seen it before – I look at A/UX and I see OS X 12 years ago – 12 years ago Apple had a UNIX OS with seamless MacOS 9 integration – capable of running UNIX and MacOS binaries.
Maybe you don’t remember that, or never even knew that – A/UX is long dead, and because of commercial licensing, will never be seen again.
OS X can suffer exactly the same fate – Apple falls on hard times and all that work gets dumped, and lost.
Look at the BeOS – So much potential, and good ideas – only just now resurfacing in Longhorn/Tiger. What a waste of time and effort.
AmigaOS – the amiga was revolutionary in it’s time. Now where is it? It certainly didnt die for lack of user support.
or NeXT – NeXT stagnated for 10 years before becoming MacOS X. Getting bought out by Apple was hardly a forgone conclusion – the ‘better’ OS could very well have never been, and all that work flushed straight down the corporate crapper.
I’m certainly not the oldest person in the computer industry, but I’ve seen enough to come to the conclusion that Open Source is, as far as I am concerned, the only sane way to move forward.
And if it leaves a smoking crater where a once-mighty commercial software industry stood, then so be it.
I’d like to point out to the rest of you posters that this was the only person who caught the drift of my original post and responded in a fashion that allows for real discussion before posting my response to him. That being said, thank you anonymous! (And to doggedblues, I didn’t mean to be hostile, I’m sorry you read my post that way.)
I see software through the eyes of someone who wants to make a living from it. I see the value in learning from other people’s code from the eyes of a student. What I have yet to see is a feasible way of combining the two. This is the crux of my difficulty in understanding the FOSS world.
To illustrate:
My job during the summer is that of a chemist. The products I design have a code consisting of chemical reactions, additives, and various components. In this world, open source is VERY bad. What good is my product if anyone can copy it? How does this help me ensure my survival in this world if someone else is out there taking advantage of my work? If someone out there is giving away my formulas and chemical reactions for free, that puts me out of a job! What good am I to a company when they have someone else doing my job for free?
That being said, I don’t see how this is any different in the software world. I’d like to stop this thread though because it doesn’t belong here. I am more than willing to continue this dicussion else where though if anyone else is willing to present the other side of the coin to me.
I’d like to ask you all one question that no one seems to have answered of mine yet, is there an advantage to the PPC platform that makes it desirable for Linux development that x86 cannot fulfill? If so, what is this advantage?
A petition for nvidia to port their x86 drivers to PPC:
http://www.petitiononline.com/nvppclin/petition.html
Methinks it’s an nvidia/Apple thing rather than any technical limitation. Although, perhaps there’s x86 assembly for performance…
This is a bit off-topic, but it’s related to Yellow Dog Linux and PPC.
Mac is indeed NOT the only platform to run PPC Linux. For example there is the Open Desktop Workstation (http://www.pegasosppc.com/odw.php) which is based on the Genesi’s Pegasos PPC-motherboard. This is a G4-based computer that comes pre-installed with Yellow Dog Linux. This product is endorced my IBM and Motorola.
>I’d like to ask you all one question that no one seems to have answered of
>mine yet, is there an advantage to the PPC platform that makes it desirable >for Linux development that x86 cannot fulfill? If so, what is this advantage?
x86 is an old design that has had bits bolted on – time and again
– to retain backward compatibility.
Progressively more of the design effort, with x86, goes into limiting the effect of previous design decisions.
It adds upto: a convoluted instruction architecture, more silicon required (with attendant manufacturing challenges) and higher power/ heat levels in use.
Sooner or later x86 will be superceded – I, for one, would never have predicted that it would get this far.
As the open source technology takes a stronger part of mainstream then the dogma of particular chipset weakens.
The PPC series are a candidate with lower power, higher throughput per CPU cycle. Years ago I thought the ARM chips might make a dent …. these things are hard to predict.
“What’s really needed is for the mass producing x86 mobo manufactures to get on board and mass produce just as cheap PPC mobo’s… SolTek, MSI, PCChips etc would be the ones to truly make this a viable solution.”
The problem is to find a market which will buy large numbers of PPC motherboards. This is somewhat harder than finding a market that will buy large numbers of Macs.
The craziest thing here is that we would need Windows for PPC
in order to get some interest in PPC and therefore get low cost PPC board for Linux.
To be honest when I see how difficult it is for IBM to produce enough chips for Apple, I really doubt their ability to supply a large number of mobos manufacturers.
PB
Assuming we have a ready supply of Motherboards…
Where do we get processors from?
I thought that for the AmigaONE board, designing it to use the ready supply of Apple Macintosh processors (ZIF and newer) made sense.
It would reduce the cost of the logic board to PC level costs.
Most of the cost of the Pegasos II and AmigaONE Boards is the Processor chip.
If I could buy a $150 – $200 bare logic board, and a surplus $50 – $200 PowerMac Processor (400mhz – 900mhz), I think a LOT more boards would sell.
THEN, I’d pay $100 for AmigaOS or MorphOS, or a PPC Linux Distribution.
When a board with a processor is $900+
That’s not going to be attractive…
“Also, If you have standardised on Linux, it is nice to have a single environment on all machines.”
Your right. This is very important. Linux is platform independant, giving you the freedom to choose what hardware you like. So an x86 Gentoo user can switch to a PPC Gentoo box, and have a system that looks and works like the one he’s used to. That’s pretty nice.
“It is not enough to have software. We want to have free software, and are prepared to work to acheive it.”
Even besides that, I want choice. I, for one, appreciate the fact that buying an Apple computer does not bind me to using the Apple OS. Sure, for now, I CHOOSE to run OSX, but I have my option. Whether or not I ever switch over to Linux, I’m glad to know I CAN.
“Your right. This is very important. Linux is platform independant, giving you the freedom to choose what hardware you like. So an x86 Gentoo user can switch to a PPC Gentoo box, and have a system that looks and works like the one he’s used to. That’s pretty nice.”
But if it looks and works the same, why would he switch? What would be the improvement to justify the cost?
“The craziest thing here is that we would need Windows for PPC
in order to get some interest in PPC and therefore get low cost PPC board for Linux.”
…or Linux could gain significant market share. Once you’re on Linux, it’s not a huge difference from the user perspective to move from x86 to PPC (or other platforms, so long as you or anyone has made support for it). This would be one of the major improvements of Linux taking some large market share: not only would it break Microsoft’s hold on the software world, it would break Microsoft’s hold on the hardware world.
It existed once, back in the old Windows NT 4.0 days.
and some nice APIs, but all the ‘good bits’ of OS X are closed and proprietary.>>
As is Apple’s right. They laid out the $$$$ to make it. And have no obligation to give it away.
You can’t modify it,>>>
Thank God. 10+ years and “Team Linux” *still* can’t come up with a desktop that just effin works or software that has a straightforward universal non CLI install.
If it ain’t broke, don’t fix it.
you can’t redistribute it,>>
That’s because it’s not yours. However, you can distribute Darwin all you like.
and you can’t learn from it.>>
Oh, I think it’s possible to learn from OS X. Namely, *nix doesn’t have to suck rocks. Or that you don’t NEED or have to want to learn reams of “cthulu” to make the OS really useful on your desktop.
I don’t want to learn to config a video card. I want to learn to do cool things in P-Shop and FCP and Dreamweaver and Acrobat. (Actually that’s what I need to learn for my job.)
It’s so much less than it could be>>
Looking at rose briar (beautiful yet prickly) that is Linux, I’ll say that less is more.
Oh, and since you can throw on X11 and fink over any Linux program and windowing environment you want, OS X actually gives you more than Linux.
Thanks Al, that’s exactly what I was looking for when I was asking about PPC versus x86!
It’s sad that ARM has only become an embedded solution, it’s really a beautiful architecture. If Intel/AMD would pour as much resources into making these things fast as the do on x86.. wow..
ISA wise I think it’s even a lot nicer than PPC.
ARM > MIPS > SH4 > PowerPC > x86
(where > means is nicer than)
I don’t know about the redistribution thing, but you certainly can modify and learn from, if not OSX, at least Darwin: the source is there for everybody (Yes, I know, Quartz’s source is not there, and it’s the only one that interests Free Software people).
“But if it looks and works the same, why would he switch? What would be the improvement to justify the cost?”
I don’t know, because you like the design of the Powerbook, or because the G5, according to some studies, anyway, is the “best bang for the buck”. Maybe you want to dual-boot into OSX, or maybe you just already have a Mac lying around?
Either way, I wasn’t really trying to get into a “Intel vs. IBM”, “Apple vs. PC”, or “x86 vs. PPC” type of argument. I was merely stating that one strong point of Linux is that it has support for a wider variety of platforms than many operating systems. That gives you more choices and more options. Also, it’s a good thing for the PPC architecture that it’s supported by Linux. Again, it gives you more choices and more options. Having more choices is rarely bad, even if you choose not to choose them.
So, it’s good for Linux and good for the PPC architecture. What’s good for the PPC architecture is good for IBM, especially since Apple isn’t the only company that uses PPC chips. In fact, I sometimes suspect that IBM pushing for Linux on the desktop is part of a plan to open a new market for their PPC architecture without having to go through Microsoft. As IBM is one of the big supporters of Linux, and things that make Linux good for IBM are going to increase IBM’s pushing Linux as the OS of choice, it’s probably good for Linux in that round-a-bout way, too.
Now, if you’re specifically anti-Apple, anti-Linux, anti-IBM, or generally against having extra choices, then maybe it’s bad for you, I guess. Otherwise, I’m not sure I see a problem.
But if it looks and works the same, why would he switch? What would be the improvement to justify the cost?
Perhaps you’ve hit on the point the original poster was trying to make? Since Linux supports various hardware platforms, this opens you up to making your hardware purchasing decisions based on things like cost and performance, instead of being forced into a hardware platform based on the operating system you want to run.
Hi Al, my Pegasos II G3 cost me much less than $900. I spent ~$350 on the board and CPU card. I bought it on http://www.pegasosppc.com in January 2004 and configured it in a case with memory, hard drive, graphics card for another $150. I used a keyboard and mouse from an old PC. It has been running since. I have partitioned the hard drive with Debian, Gentoo and MorphOS. I run Mac-on-Linux on Debian, but is is officially supported by Gentoo also.
Whoops, and YDL too, but it still needs some work to run as well as Debian or Gentoo in PowerPC.
As a linux user on Apple hardware, what I would relly like to have is a IBM branded laptop based on PPC architecture.
It would mean hardware support for linux and no MS-tax.
IBM builds PPC CPU and supports linux, so why not sell a machine devoted to linux on they very own processors instead than intel’s ones?
Apple Ibook lacks wifi support (as it has none PCMCIA cardbus only a usb external adapter is viable: not that portable-friendly), Powerbook are now simply to expensive without suspend-to-ram, video card full support (I blame ATI and NVidia, kudos to Benh) and the need to buy a PCMCIA wifi card (with an external antenna) while you get a useless Broadcom internal MINI-PCI nic.
The author mentions how some things of PPC Linux (Mainly their Yellow Dog Distro) just don’t work. Without basic desktop needs like Airport or 3D hardware compatibility, PPC Linux, or at least Linux for Mac does not meet standards to be used as a desktop machine. However, because of their extreme power and cheap cheap price, Linux PPC boxes are perfect server material.
> I don’t want to learn to config a video card. I want to
> learn to do cool things in P-Shop and FCP and Dreamweaver
> and Acrobat. (Actually that’s what I need to learn for my
> job.)
Well, that says it all really.
You feel your desire to learn only about the things that you feel are important are a justification for limiting my choice over what I should be learning about?
Maybe I want to know how a video card works, or the details of how my windowing system talks to it – is that something I shouldn’t know, because you don’t understand it either?
‘OS X is good enough for me – it must be good enough for you’ – is that what you are trying to say here?
> As is Apple’s right. They laid out the $$$$ to make it.
> And have no obligation to give it away.
Agreed. 100%. Thats why LinuxPPC exists. I don’t actually want to have to pay Apple to obtain the right to use my computer. I’m quite happy to pay them for the computer itself.
> If it ain’t broke, don’t fix it.
It is ‘broke’ – you can’t redistribute it, you can’t modify it, and you can’t learn from it.
Its very simple. The Free/OSS communities want to build a software environment that is free to use, free to redistribute, free to modify and free to learn from.
Your priorities are obviously different, and thats fine.
I’m not suggesting LinuxPPC is right for you, i’m not saying OS X is bad, or wrong.
But it’s not Free Software, which is why LinuxPPC exists alongside it.
If LinuxPPC is ugly to you, if it doesnt work for you, whats the problem? Nobody sold you a computer and told you that you couldn’t run anything but LinuxPPC on it, did they?
You’re saying LinuxPPC is ‘broke’ because it doesnt have a polished GUI layer, and doesnt run your apps, as if those reasons for it being ‘broke’ were somehow more valid than my ‘not free software’ reasons for OS X being ‘broke’.
Anyone is free to fix LinuxPPC’s ‘broke-ness’, you can’t say the same for OS X.
“‘OS X is good enough for me – it must be good enough for you’ – is that what you are trying to say here?”
No. He’s saying “OS X is good enough for me. If it’s not good enough for you, use Linux.”
Your entire post is saying “Linux is good enough for me. If it’s not good enough for you, use OS X.”
In other words, you are (both) stating the bleeding obvious.
My job during the summer is that of a chemist. The products I design have a code consisting of chemical reactions, additives, and various components. In this world, open source is VERY bad. What good is my product if anyone can copy it? How does this help me ensure my survival in this world if someone else is out there taking advantage of my work? If someone out there is giving away my formulas and chemical reactions for free, that puts me out of a job! What good am I to a company when they have someone else doing my job for free?
In the same light, where would Newton be if he had given away his Calculus and the works of Optiks. Oh, wait. He did give them away! I think the question should be: where would we be had he not given them away. The scientific persuit shouldn’t be in the name of personal gain.
IBM did make RS/6000 laptops and PPC Thinkpads many moons ago, but they didn’t sell well. They sometimes turn up on eBay and use 603e processors. They’re built for AIX but you can slap Linux on if you like.
BTW, the iBook does support WiFi. Get an Airport card. Simple as that.
J…