Ive been flooded by e-mails and phone calls from people giving me their two cents on how they think the future of desktop computing is going to turn out now that Apple is moving Mac OS to the Intel platform.
I had a feeling this might happen when I opined that the Mac operating system on Intel poses a real threat to the future of the Linux desktop on Intel. Since I cant write to everyone, here are my responses to some of the most popular comments.
One of the most common themes Ive been hearing is that Apple will not—will not, I tell you!—be letting its precious operating system ship on anything except its own branded hardware.
Yes, that appears to be what Apple is planning, but I dont think itll be able to pull it off.
The Intel-based PC architecture is an open one. How is Apple going to keep people from installing Mac OS on non-Mac PCs? Proprietary firmware? A BIOS, a la Phoenix Technologies TrustedCore, which can enforce DRM (digital rights management)? Pick a method and I foresee hackers cheerfully breaking it.
In addition, open-source programmers are working against proprietary BIOS. While AMD has backed open BIOS, Intel has been fighting the Free Software Foundation on this issue.
This has been a slow-burning fire, but if Apple does indeed try to lock its operating system with an Intel-based proprietary hardware configuration, I can easily imagine public flame wars with the open-source community that Apple cant afford.
I see two possibilities here. The first is that Apple sticks to its guns. In this case, the result will be a gray market in non-Apple-Mac-OS PCs, in which Apple eventually gives up and, at the least, licenses Mac OS to a few select hardware OEMs. Anyone want a Mac OS HP Pavilion? How about a Sony Vaio Mactel?
Or, Apple sees this as a battle they cant win, and does all of the above without a fight.
Ive also had people tell me that they dont think Apple will have as much trouble with device drivers as Linux has, either because Apple will use only a small subset of all the possible equipment you can stick on a PC, or because Apple will be able to get better deals from hardware vendors.
Theres no question that Linux still has a long, long way to go with device drivers. Its just that Macs have even farther. Macs support only a very limited range of hardware.
In the past, if the equipment wasnt built by Apple, or had its making overseen by Apple, the odds were the device wouldnt work on a Mac. Linux still has big driver problems. Mac OS on Intel will have enormous, but conquerable, driver problems.
On Intel designs, Apple will have to contend with a much wider variety of external equipment. From Webcams to scanners to, heck, even mice, Apple will have to deal with tens of thousands of new devices. Even if “Mactels” turn out to be hermetically sealed boxes, Apple and friends are going to be spending a lot of time working on device drivers. And, lest we forget, historically, Apple hasnt gotten along well with equipment vendors.
On the flip side, Ive gotten messages saying that Apple cant possibly be successful running Mac OS on Intel.
What part of the Jobs keynote speech did they not hear?
It must have been the part where he said that Apple has had Mac OS running on Intel chips as well as PowerPC for the last five years.
Its not perfect yet, but Apple isnt even shipping the first ones until 2006. By then, I expect—and so do the 500,000 or so developers that Apple claims—the Mac OS will be doing just fine on the Intel platform.
Ive also been hearing from a lot of people who insist that it doesnt matter what Apple does, since Linux will survive no matter what the companies do. Would someone please send these people in their basements the memo from a few years ago that Linux is a business operating system and not just a hobby?
In a somewhat similar vein, Ive been told that Mac OS wont pull in Linux desktop users. Well, probably not, but Im not worried about them.
Its all those Windows users who could come over to Linux as they finally get sick and tired of Windows endless security problems and a lamed, late Longhorn. I had really hoped that the Linux desktop could scoop up a lot of those users. Now, unless things change, I see them going to Mac OS instead.
One reason why they might do so is that I predict that between WINE, CodeWeavers CrossOver Office and Intels Virtualization Technology, I see many Windows applications running natively on Mac OS on Intel.
Once on the Intel chip family, the biggest problem with running Windows applications on Macs—emulating the 86 instruction set—will disappear. The Linux desktop will also gain Windows users from better Windows applications with Unix support, but I fear that Mactel developers will reap the most benefits from it.
So, to get back to the original point of my column, if you want Linux to play a serious role on the desktop in the years to come, the time is now to get serious about creating an outstanding Linux desktop operating system. Mac OS X is coming to Intel and, unless things change, its likely to be unchallenged as the best Intel-based desktop operating system around.