Submitted by Dan Kusnetzky, Apple program vice president, spoke to PC Pro News at IDC and said that Apple has no plans to support booting into Vista on the new Intel Macs. However, a Microsoft spokesman said that Microsoft would have no problem granting a Windows licence to Apple, in exactly the same way it currently provides licences to Dell and HP: “Microsoft would support Apple the same way it supports every other PC manufacturer.” But Dan Kusnetzky said it would be difficult to know who would support that machine if Windows was running on a Mac. However, “no doubt someone will work out how to run Windows on the Mac, even if Apple doesn’t technically support that.”
I can’t see Apple agreeing with this. If OSX was sold seperately to be installed on any x86 system then yes they may consider becoming like HP and Dell that provide consumers the choice between Windows or Linux be preinstalled. Since the OSX EULA is entirely set out to ensure OSX is not distrubted on systems not sold by Apple then Apple won’t be interested in providing their customers this option.
Then you are really getting into interesting territory. After all the only magic in MacOS is at the UI level, the kernel has just been shown to be as insecure as WinXP. I really wonder how fast OSX kernel compares to WinVista in speed since MS has years and years of expertise in optimising for x86.
Edited 2006-03-12 19:24
After all the only magic in MacOS is at the UI level, the kernel has just been shown to be as insecure as WinXP
When? By whom?
There’s a hell of a lot of “magic” in Quartz, which I believe is fairly closely tied to the kernal.
And more importantly, do you think replacing the entire kernal would be that easy? Plus all the BSD tools that many people rely on would just stop working. And Apple would have to persuade Microsoft to provide a PPC versionb of the kernel, becuase Apple certainly aren’t going to suddenly stop PPC versions of the software, just because they now have some computers that run on Intel chips.
No it’s not, that’s the point of having a micro-kernel. If it’s closely tied to the kernel than Apple has got a largely monolithic kernel
.
I think there’s a lot of magic in quartz that’s dependant on driver functionality which runs as “servlets” (or whatever they call them) for the kernel.
Actually, Mac OS X’s kernel is in fact more mono than micro.
http://apple.slashdot.org/comments.pl?sid=149636&cid=12544699
> There’s a hell of a lot of “magic” in Quartz, which I believe is fairly closely tied to the kernal.
It’s a freaking Microkernel!. There’s NO magic inside the kernel.
> lus all the BSD tools that many people rely on would just stop working
As for all BSD tools, Windows has it – it’s called Services for UNIX. Windows networking stack is based on BSD.
1) OS X isn’t a microkernel, but BSD layer is in the kernel. You’re right that Quartz is in userspace, though.
2) The Windows networking stack is not BSD based. They used to use a BSD-based stack from a company called Spider, so there are some BSD-licensed utilities in there, but the stack is Microsoft’s own.
3) UNIX apps on Windows is possible, but Windows is generally different enough that things don’t work smoothly.
4) Ripping out BSD and substituting the NT kernel would be next to impossible. The OS X GUI is tied to the system-level userspace (stuff like the LWWP and Quartz Compositor), which are closely tied to kernel-level APIs like the IOKit. Large parts of the userspace use Mach IPC and other features, making it difficult to port the OS X GUI to FreeBSD, much less Windows NT.
And BSD has had absoluetely no experience on x86?
There is more “magic” than the UI level, and OS X is not “as insecure as WinXP”.
False statistics do not a point make.
There is more “magic” than the UI level, and OS X is not “as insecure as WinXP”.
Actualy, as much as I don’t like MS (hell I hate Apple even more), I say bull to that.
Both OSs are more or less secure, trouble is how they default their desktop usage (this is the reason of one being secure and one not in common opinion). OSX is directed to more secure usage, while XP is directed (in point of security view) to as stupid approach as possible.
But parent was not talking about desktop usage, he talked about kernel.
If one secures WinXP as he should, usage without problems is almost guaranteed. The fact that 100% of viruses and mallware is Windows-ware should not count here either, OSX just isn’t interesting platform for that. One who wants to make damage, is always inclined to greater damage, and Apple 2% is not wished result here.
Personaly, I preffer Linux over anything. But I wouldn’t dare to say Linux is the most secure here. Sure it provides solutions like Xen, AppArmour or SELinux, UML, chroot, least-privilege-model, now even Trusted-X, etc. But how would Linux fare under the same stress as Windows? I can only assume better, because they have advantage of seeing mistakes on Windows before even getting to them. And the more security features it gets before hits actualy start to show, the better it will be and the better it will fare comparing to others.
False statistics do not a point make.
And comparing apples to orages doesn’t make it either.
> But how would Linux fare under the same stress as Windows? I can only assume better,
Sorry but we’re talking WinVista. Even Win2K3 is more robust than Linux – moreover Linux kernel is pretty lame for backwards compatibility issue because you can’t take drivers compiled on Linux 2.6.15.1 and run them on Linux 2.6.15.2.
Sorry but we’re talking WinVista. [skip here] moreover Linux kernel is pretty lame for backwards compatibility issue because you can’t take drivers compiled on Linux 2.6.15.1 and run them on Linux 2.6.15.2.
[your versioning nonsense] You don’t have a clue about linux, aren’t you? [no need to answer this, answer is obvious]
Did you actualy read what I wrote, or you’re just jumping to find one part where sentence clashes with your bias, but without reading it completely?
Or to translate my question. Do you like fighting over who’s got bigger dick and pissing contests like that so much?
Even Win2K3 is more robust than Linux
Since this is nothing but a lame trolling comment not having a shred of connection to my post, I’ll just ignore that. And please, you ignore this part of my response too.
Edited 2006-03-12 20:35
> your versioning nonsense] You don’t have a clue about linux, aren’t you? [no need to answer this, answer is obvious]
Right here right now – let’s settle this. You prove to me that the nvidia driver compiled on Linux 2.6.15.1 can work on 2.6.15.2 without recompiling and I’ll stop posting here on OSNEWS.
If you lose, you stop posting here forever.
Deal?
Dude, I know what I’m talking about so you better back away from this argument.
Right here right now – let’s settle this. You prove to me that the nvidia driver compiled on Linux 2.6.15.1 can work on 2.6.15.2 without recompiling and I’ll stop posting here on OSNEWS.
Why would I do that? I never said that you can do this.
As much as I’m pro Linux, I’m also realistic.
If you go trough my post history on OSNews you can also find my little war with few real zealots about this.
Let me repost it again, or at least the point of it:)
Monolithic kernel and lack of stable ABI in kernel is double edged sword.
Good sides:
1. not bounded to specifics forewer
2. larger gains possible because devs are not bound to ancient history
3. monolithic is faster
Bad sides:
1. companies don’t like the moving ground
2. see 1.
3. for each change in main branch, complete kernel has to be recompiled
From a user perspective:
1. there are repositories
2. usualy those repositories follow both, drivers and kernel modifications
3. when you update kernel you usualy update driver too. So there is no more recompiling for users anymore.
As you see, flaws of linux approach are mostly company related. And since my usage is mostly commercial, I could say I’m part of the party that suffers the most from this fact. But even I can’t remember one single example where these facts would hurt me.
If you lose, you stop posting here forever.
Why? No one will loose:) I won’t argue you don’t have it right:) I will just argue that your viewpoint is immature:) And this is the reason for me pointing out that you don’t know much about linux:)
Deal?
Nah:) It would be a waste to lose you. You’re funny:)
Dude, I know what I’m talking about so you better back away from this argument.
No, you don’t:) Or at least your viewpoint is completely missed:) You can’t say that something is bad if it provides enough benefits back +1-1=0
Edited 2006-03-12 21:33
> Let me repost it again, or at least the point of it:)
> Monolithic kernel and lack of stable ABI in kernel is > double edged sword.
>
> Good sides:
> 1. not bounded to specifics forewer
So that’s another way of saying, I don’t know jack about designing operating systems but lets see what I can hack up this week.
> 2. larger gains possible because devs are not bound
> to ancient history
Imagine how this would play out at MS/Apple/Oracle/IBM/Sun?. You release a patch of your system that toally breaks everyhthing that came before.
> 3. monolithic is faster
OK no arguments here.
> Bad sides:
> 1. companies don’t like the moving ground
> 2. see 1.
> 3. for each change in main branch, complete kernel
> has to be recompiled
Why just companies. Users don’t like moving ground. One day Nvidia works and the next day it doesn’t. I can assure you that through all the SP1s and SP2s and Patch Tuesdays of Windows XP, I’ve never had to reinstall my Nvidia driver.
> From a user perspective:
> 1. there are repositories
> 2. usualy those repositories follow both, drivers and > kernel modifications
> 3. when you update kernel you usualy update driver
> too. So there is no more recompiling for users
> anymore.
Except these user repositories don’t cover all kernel drivers and modules – how about ones from VMWare or Win4Lin or OSS? These people are so religious that they say you want us to maintain proprietary modules – tell the companies to open source them and they jolly well know that the proprietary guys will not.
Edited 2006-03-12 22:01
ok let see what you hacked up:)
Imagine how this would play out at MS/Apple/Oracle/IBM/Sun?. You release a patch of your system that toally breaks everyhthing that came before.
I already said it is bad for company haven’t I???? Zero points, can’t count repeating my words as your knowledge:)
btw. Look how RHEL updates kernel for its distro. Oracle won’t stop working unless you start using your own vanilla compiled and patched kernel.
Why just companies. Users don’t like moving ground. One day Nvidia works and the next day it doesn’t
If users installed driver from repo. kernel won’t update until dependancies for all packages will be satisfied:) Meaning until driver is repo, no kernel update for you. But after update, everything still works:) Zero points.
Except these user repositories don’t cover all kernel drivers and modules – how about ones from VMWare or Win4Lin or OSS? These people are so religious that they say you want us to maintain proprietary modules – tell the companies to open source them and they jolly well know that the proprietary guys will not.
Yep, I forgot to mention VMWare, but it is part of the bad for commercial I specified. One half of the point. There’s still no repo that would support VMWare updating modules. Again, if you go trough my history, you’ll see VMWare was one of the general points of flamewar:)
Soooo, 0.5/3. How did you fare in your opinion?
p.s. I should probably take off 1 point for completely missing the debate you started in the first sentence. “I don’t know jack about designing operating systems but lets see what I can hack up this week.”
Well, this week you talked about the affects of non stable ABI on proprietary world (and this is just one of zillion point in designing operating systems, a drop of water in the sea). And not even a word about “designing operating systems”. Ok, lets say you had a bad week without stripping you one point:)
Now my question:) Answer would be appreciated
Why do I get the feeling that you’re trying to fight a holy war without a willing adversary?
Or at least, I won’t fight, I’m just amused by your comments 
I’m no zealot, I’m not saying everything has to be opensourced (hell, I don’t do that for all of my software). I’m not saying Linux is better than Windows or OSX (and whenever I did I always specified personal opinion and I really do that always), mostly I’m agnostic because my job forces me to think crossplatform, but for my misfortune I’m always ready to answer to everybody. All of Windows 2k3 and all Vista betas have more or less just discouraged me to use them on a larger scale, I use 2k3 only when needs say “no different solution”, and Vista I’ll probably avoid as long as possible.
While you on the other hand (or translated into human language), were specifiying Vista is the next gods gift for humanity, and Win3k sweeps the floor with Linux. Both are only biased, untruthfull and over zealoted remarks.
I can assure you that through all the SP1s and SP2s and Patch Tuesdays of Windows XP, I’ve never had to reinstall my Nvidia driver.
Lucky you then. Countless people had problems wih network and mouse drivers when updating on SP2. (search on google) Just as many apps stoped working (ms even provided 2 page list of them). The only app I use on Windows stoped working on SP2 update. Talk about this funny fact:)
Edited 2006-03-12 22:39
That’s a really lame argument. Apple didn’t guarantee that drivers would work between OS X releases either (until 10.4). Nobody seemed to have noticed. The only difference is Linux has a mechanism to detect the driver version and prevent a module from loading but mysteriously failing, and OS X doesn’t.
Driver compatibility is a big deal on Windows, where the drivers come from third parties, but on OS X and Linux, where almost all drivers are first-party, it’s really not important.
Right here right now – let’s settle this. You prove to me that the nvidia driver compiled on Linux 2.6.15.1 can work on 2.6.15.2 without recompiling and I’ll stop posting here on OSNEWS.
The level of maturity around here rivals a Middle School playground sometimes. Why don’t you simply just say that your daddy could beat up his daddy?
But parent was not talking about desktop usage, he talked about kernel.
You’re absolutely correct. Hardening is much different than the kernel on its own. Still, OS’s are installed with default kernel configurations as well that run deeper than the desktop.
If one secures WinXP as he should, usage without problems is almost guaranteed. The fact that 100% of viruses and mallware is Windows-ware should not count here either….
I agree. I’ve never said that XP can’t be made more secure, just that, as you said, the default configuration (which the majority of users never change) is overwhelming vulnerable.
And comparing apples to orages doesn’t make it either.
Comparisons can, indeed, make a point. The desktop does have an effect on the kernel with regards to interaction. Default installations have an impact on both the desktop and the kernel. There are a large number of variables involved in maintaining system security, and pigeon-holeing one aspect is akin to saying that living leads to cancer; the whole picture needs to be looked at.
Comparisons can, indeed, make a point. The desktop does have an effect on the kernel with regards to interaction. Default installations have an impact on both the desktop and the kernel. There are a large number of variables involved in maintaining system security, and pigeon-holeing one aspect is akin to saying that living leads to cancer; the whole picture needs to be looked at.
Parent was talking about Vista not XP.
I’m not saying Vista hasn’t got any problems, but as far as security goes. With Vista, Windows finaly got least-privilege and drivers running in userspace. Which takes away most of the problems with current Windows defaults.
And I’m a linux guy my self.
Parent was talking about Vista not XP.
Incorrect. stephanem was talking about XP security:
After all the only magic in MacOS is at the UI level, the kernel has just been shown to be as insecure as WinXP.
stephanem only brought up Vista with regards to the responsiveness, or speed, of the kernel:
I really wonder how fast OSX kernel compares to WinVista in speed since MS has years and years of expertise in optimising for x86.
I do agree that Vista will be leaps and bounds ahead of XP as far as security goes, but my initial comment remains.
I believe the OS X kernel is very efficient. I’m horrible with operating systems; I can crash them easily, and when I’m messing around with Linux I usually mess it up so I can’t boot X anymore.
I have to reinstall Windows XP every six months because it slows down, and just generally becomes unstable. I used to have an iBook, and man no matter what I did I just couldn’t screw up OS X. After a year the system was still working wonderfully.
I think Apple is just fine and dandy sticking with present OSX kernel. If these were the OS 9 days, then maybe I could see your argument, but saying that Apple should use the Vista kernel with OSX being so stable and mature now is just crazy.
No, the OS X kernel is very innefficient. It’s a micro-kernel. It’s an elegant, debuggable, modular, beautiful, slow solution.
I imagine Windows kernel is fairly efficient, although I think the massive amount of code dedicated to UI responsiveness may make it a lot less efficient, but that’s a sacrific intentionally for responsiveness at the cost of throughput.
Linux is efficient. I’m sure most BSD kernels are efficient.
Apple should probably stick with their kernel, but if they did move it’d be idiotic to license the kernel from Microsoft. That’d simply be financially stupid (waste of money). It’d be strategically dangerous (depending on Microsoft, a competitor). And it’d be ignorant, of the other two kernels more suited to your present system.
OS X on top of NT would be nuts. OS X on top of OpenBSD’s kernel might make a shred of sense though. But OS X on (is it mach?) where it’s at now is a working solution and I don’t think I’ve heard about many kernel security issues yet!
[QUOTE]OS X on top of OpenBSD’s kernel might make a shred of sense though. But OS X on (is it mach?)[/QUOTE]
OS X is already on top of the FreeBSD Kernel.
[QUOTE] It’d be strategically dangerous (depending on Microsoft, a competitor)[/QUOTE]
AFAIK, Microsoft has some money inside Apple. I have never ever been able to understand how this exactly work, but I’ve read from reputable sources that Microsoft indeed does have some money inside apple.
So you’re trying to say that because you can cause Linux to not boot X the kernel is somehow ineffecient? I’m sorry, but I don’t think you have a clear concept of what the kernel does or what would make it ineffecient (and yes, the OS X kernel is ineffecient imho).
I stated that I believe the kernel is efficient, and ended that statement with a period. The following did not relate to that statement at all. If I was continuing that statement, I would have used something called a semicolon. I’m sorry but I don’t think you have a clear concept of how English grammar works.
“I really wonder how fast OSX kernel compares to WinVista in speed since MS has years and years of expertise in optimising for x86.”
Really? It doesnt show.
This really is such a non-news item. The headline might as well say “MS Willing to Contine to Find New Avenues Toward Increasing Revenue”.
Can anyone explain why MacOS X 10.4.5 runs perfectly on my IBM laptop?
Yes, it is Maxxus version but explain why it runs *perfectly* on my IBM laptop – including support for wLan, BlueTooth, network cards, USB, special keys, special IBM function keys, the trackpointer etc.
This isn’t anything added by Maxxus. It must have been included by Apple. This is certainly nothing of use if you have a Mac, especially support for the trackpoint, IBM-keys etc.
MacOS X 10.4.5 is way more compatible with my IBM laptop than Windows XP where I have to install separate network drivers, wireless drivers, monitor drivers, keyboard drivers etc.
Why?
I believe there is a “standard” for shortcut keys, there is a de-facto mapping for the “spare” keyboard scancodes that shortcut keys often use. Maybe that explains it?
As for wLan, BlueTooth, USB, etc. These are all fairly common components with a fairly small number of common chipsets.
IBM’s “special keys” talk direct to the hardware for volume, brightness, wireless antenna control, etc. No OS intervention needed (its the same on a lot of laptops where their main audience isn’t the home user, for some odd reason…).
The trackpoint is a PS/2 device – well, it might be USB by now – and as such is completely standard too.
Can anyone explain why MacOS X 10.4.5 runs perfectly on my IBM laptop?
Simple fact.
You’ve got almost the same chipset components as MacBookPro. And since Darwin (hardware up to 2006 in this version) is not Apple specific, Apple alone doesn’t take away non-Apple drivers (They are in Darwin already. But you could expect removing like that in future if Apple would start to think that it hurts them in any way).
Windows (where kernel is based on hardware up to 2002) on the other hand supports much wider variety of hardware. And this is why you need to install separate drivers.
It is not that I preffer Windows, I hate them, but at least mini realistic viewpoint is the least that one can show.
Now your IBM laptop was made, when? You bet, MS can’t do mumbo-jumbo, jump a few hoops, run few times over the fire and make all drivers for next few years ahead. Apple with their Darwin and new releases every few months can.
Because your laptop is nearly identical to a mac on the hardware front. Macs use no special hardware. Really only they motherboards are re-shaped, but the chips on them are the same as PC counterparts. With PPC boxes the chipset was a bit more special, but now they use normal intel chipsets.
Also, remember IBM/Lenovo doesn’t make the thinkpad, and Apple doesn’t make the macs. 3rd party companies do. In the case of laptops they all go back to a couple companies in asia that make them for everyone. Asus makes most of apples laptops, and one other company does the rest, IBM/Lenovo uses the same companies. Those companies won’t change internals much between different brands.
If its anything OSX already supports, or is something that uses a standard protocol that OSX supports, its going to work.
For things like trackpoints, I think they use just the standard mouse interface, so what ever in the hardware identifies it as a mouse, it treats it just the same.
If it was a special add on. Well, then there is hope that the trackpoint may come to macs, which would be great, since finger slidy slush pads suck.
Wishful thinking is fine, even when it does clutter up the Internet, but know something about Apple before making up futuristic operating systems. They’re nothing if not proud.
The only way Windows would ever show up on an out-of-the-box Mac is if a third party reseller were authorized by both Apple and Microsoft to make a dual-boot machine. Apple might agree (there’s a company that does it with Linux, though their name escapes me), but Microsoft doesn’t allow OEMs to distribute dual-booting machines.
All Microsoft Man has said here is that they’d happily go along with it if Apple dropped their flagship product to allow more Windows distribution, so as Pseudo Cyborg pointed out, he has said nothing at all.
When Apple is sitting on the OS everybody wants to run, why on earth would they license the OS from Microsoft? Clearly, it should be the other way around…
When Apple is sitting on the OS everybody wants to run, why on earth would they license the OS from Microsoft? Clearly, it should be the other way around…
—————
I have NO desire whatsoever to run OSX. I’ve been using Macs longer than I have PCs, and I still never feel very confortable when using the Mac OS. For some reason Windows (95 and up) and to some extent KDE just feel much more intuitive to me.
Now if Apple were to offer a Windows machine… I’d definately consider a laptop from them the next time I am in the market. I still wouldnt buy a desktop from them (or anyone else for that matter). I prefer to build them myself.
Apple is sitting on the OS everybody wants to run….
We used to think this, and maybe at one time it was true, but the evidence is that they would sell a lot more Macs if they would run Windows. Not to Mac devotees, but to the rest of the market. And in a way the Mac devotees don’t matter. They are not going to stop buying Macs with OSX on them, just because other people buy them with XP on them.
Apple’s real asset at this point in computing is that its a designer hardware brand. The only way to exploit this is to sell the software people want on it.
Not that there is any danger of Apple going down this path. There has been no radical connected thought about the computer business model in Cupertino since 1985, and there is no sign of any starting now.
MS really wants Mac OS 10