No Intel On OS X Part I: Economics 101

by , 2:30 PM EST, November 21st, 2000

The "OS X on Intel" movement is unquestionably an exercise in futility, to say nothing of a fine example of outright foolishness. Apple is no more likely to port OS X to Intel than Microsoft is to make Windows open source. There are many reasons for this.

First, the publicity the movement has been given makes it seem far larger than it really is. While there have been many articles on it, each refers back to the same "www.osxonintel.com" web site. If this were really the widespread outcry it’s been made out to be, surely there would be more than one web site focusing on the topic. (There are several web sites about "pocket rockets," little toy rockets made from matches and aluminum foil. If frivolity like that can produce more than one site, surely something as vital as which OS you run could do the same.) The obvious conclusion is that that one site, and the people running it, are the only ones seriously advocating that OS X be ported to the Intel platform.

The 20,468 "signatures" (the total at the time of this writing) aren’t all that impressive, either. To begin with, the site is subject to the "vote early, vote often" flaw most Internet polls suffer, raising the question of how many different people have "signed" it. (I myself "signed" twice. It requires unique e-mail addresses, and nothing more. The addresses don’t even have to be valid.) But assume that each "signature" (mine excepted) truly comes from a different person. This is still less than one-third of the number of OS X betas actually sold in about the same amount of time, since mid-September, despite the fact that "signing" costs nothing, while the beta costs $30. Moreover, the "signatures" are coming from the Wide World of Intel, which outnumbers the Macintosh easily twenty to one, while the betas are going only to a subset of the Macintosh community. What evidence is there, then, of a large enough market for OS X on Intel to justify Apple producing the product?

The OS X on Intel advocates might cite the business maxim that each letter of complaint received represents another ten unhappy customers not writing, and say that, similarly, each "signature" represents a larger number of potential happy buyers. This overlooks one critical point: the only software initially available for OS X on Intel would be the adequate-at-best programs that are part of the OS X distribution. No current Windows software would run on it, since OS X is incompatible with Windows, and Apple wouldn't be offering a "Windows Classic" environment as part of the package. (They can do it with OS 9 because they own OS 9. They have no such advantage with Windows.) Much current Linux software wouldn’t run on it, because it’s written for some Aqua-incompatible Linux GUI. (It’s true that Linux command-line software could be ported, since both Linux and OS X are Unix variants. However, if you’re confining yourself to command-line-run software, you might as well just run it on Linux, which is free, rather than OS X, which won’t be.) No current Macintosh software would run, either, unless Apple included a PowerPC emulator. Even if they did, however, the target market—Intel owners—would have to go out and buy Macintosh versions of their favorite programs. That would raise the effective cost of OS X by hundreds of dollars (assuming Macintosh versions could be had at all), and thus shrink the potential market for OS X on Intel considerably.

Another question is, why do these people want OS X to run on Intel? Perhaps they see it as a much-easier-to-use alternative to Windows. But the Mac OS has always been regarded as easier to use. Why wait until OS X to clamor for a port, especially as the consensus so far seems to be that OS X is actually less easy to use than OS 9. Why wasn't anyone demanding "OS 9 on Intel"? Perhaps they want the greater stability that OS X offers. Well, they can have that: Darwin, the part of OS X that provides that stability, already runs on Intel (or, at least, will do so much sooner than OS X as a whole), is open source, and is there for the downloading. Well, then, it must be the glitzy, "lickable" look of Aqua they want. But if that’s what they want, there are Windows equivalents to Kaleidoscope, which allow users to tweak the interface for a desired look. (For example, there is WindowBlinds [http://www.windowblinds.net/], which offers the Aqua-esque "Liquid2".) Perhaps it's just that OS X is, at least for now, "cooler" than Windows and Linux.

What I think is really being said is this: we want OS X, but we don’t want to have to buy a new computer to run it on. OS X on Intel offers its advocates only one benefit that OS X on PowerPC doesn’t: it runs on the machines they already own, and is therefore a much smaller investment. After all, those 20,000 OS X on Intel supporters can have OS X, and can have it today. They need only buy an iMac ($800) and the beta ($30). If OS X ran on PCs, they could skip buying the iMac and save $800. Now, there’s nothing wrong with them wanting OS X to run on the machines they already own. Wanting what you don't have is a fine American tradition. But expecting Apple to make it happen is nothing more than an exercise in wishful thinking.

For the fact of the matter is that Apple makes its money selling hardware. Assume, for the sake of argument, that 10% of the price of an Apple product is profit. Then, on a $1500 G4, they’d make $150; on a $100 copy of OS X, they’d make only $10. (Even if OS X CDs were pure profit, selling a G4 would still earn Apple more money than selling a copy of OS X.) Apple is better off selling Fred P.C. Owner a new G4 than they are selling him a copy of OS X alone. It follows, then, that OS X exists not to sell OS X, but to sell Macintosh brand computers to run OS X on. OS X is designed precisely to foster the very OS lust that motivates the advocates of OS X on Intel, to instill in them the feelings of “They have it. We want it.” that they have expressed, and to get them to satisfy that desire by buying a Macintosh. Allowing people to run OS X on the machines they now own would defeat that purpose. If people don’t need a Macintosh to run OS X, they won’t buy one, and Apple would lose the hardware sales that bring in the bucks, money that sales of OS X CDs would not replace. For Apple, putting OS X on Intel, allowing it to run on anything but a Macintosh, is nothing short of corporate suicide. Sorry, people, but it’s not going to happen.

Apple