hacker emblem
Search | Tags | Photos | Flights | Gas Mileage | Log in

The Case for Aqua on x86

Started: 2003-11-11 13:14:41

Submitted: 2003-11-11 14:52:09

Visibility: World-readable

Three weeks ago, in a marketing blitz worthy of a major motion picture release, Apple released Panther to the world, its latest major point release in the highly-successful MacOS X series. Since then, everyone with a Mac handy and US$130 to spare bought the bits-in-a-box, and most of them wrote about it on their blog. The consensus is that everyone likes it; some even suggest one could dump one's Wintel machine for a Mac. (Of course, Apple suggests this as well.) Especially among those who managed to substitute "major newspaper" or "syndicated column" for "blog", Panther's features, usability -- and especially stability and virus-resistance -- provide a compelling excuse to switch. Maybe even with eMacs starting at US$800.

Therein lies the problem. By definition, everyone who might be tempted to switch already owns a Wintel machine. There's even a good chance it's a fairly recent one. After buying a new eMac, potential consumers are stuck with an expensive -- and toxic -- doorstop.

For the past twenty-five years, the personal computer industry has been strongly segregated. Volvo-driving yuppies and Volkswagen-driving graphic artists bought hermetically-sealed Mac toasters that came with everything they needed; everyone else bought Wintel machines. What if that wall were to come crashing down in an instant?

What if Apple sold OS X for commodity x86 machines?

The idea of MacOS on Intel hardware isn't new -- back in the early nineties, Apple's "Star Trek" project (where no MacOS has gone before) had their OS running on the day's 386s, until the project was killed to cut costs. OS X's BSD base, Darwin, already runs on Intel machines. Panther ships with gcc 3.2. What would it take to port the user interface?

The free-wheeling nature of commodity x86 hardware means that anyone, their brother, and sleazy half-uncle can build bargain-basement computers assembled from bargain-basement components with bargain-basement warranties, support, and reliability. This competition drives prices down and therefore is good for the consumer, but it is a nightmare for anyone trying to write an operating system to run on everyone's computers. By selling toasters, Apple has it easy -- there are a total of eighteen Mac models sold within the past five years that will run Panther that Apple has to worry about supporting. It would surprise me if any one of Dell, Gateway, or Compaq sold only eighteen models within the past five years, let alone everyone's dubious uncles building their machines in their basements.

This is a formidable problem, although I believe not an insurmountable one.

What would this mean for Linux and the rest of the open source community? In the short term, Linux would loose mindshare as a desktop operating system for the masses. In the long term, however, I believe increased heterogeneity of the desktop operating system market would only benefit the Linux community. Apple has the mindshare and the marketing dollars to become a significant player by catering to our mothers and everyone else who wants their computers to Just Work. The quasi-open nature of OS X's support system will pay vast dividends down the road -- envision an army of proto-hackers cutting their teeth on bash, perl, and gcc.

What about Microsoft?

This is the forty billion dollar question. Obviously, Microsoft isn't going to take kindly to the fact that Apple is invading its home territory. For years, Apple has been dependent on Office to maintain its Mac sales; not being able to open "standard" office documents would be a show-stopper for any Mac. Microsoft has successfully leveraged its Office monopoly to bend Apple to its wishes; any direct competition from Apple would likely elicit similar threats. However, one key difference now exists: Microsoft is a convinced monopolist. Ashcroft may have let Gates and Co. get away with a slap on the wrist, but the conviction stands. If Microsoft dared to wiggle its little finger at Apple, Apple could seek immediate relief, something I imagine the courts would be happy to do. (Especially the evil and left-leaning Ninth Circuit Court everyone keeps chattering about.)

The real question, then, is if Apple wants to sell its OS to the unwashed masses. For years, Apple has sold toasters -- open the box, plug it in, and it works. No mucking with the inside required or allowed. Apple killed off its mid-nineties clone market because it went against its ability to sell toasters. The iPod is a small, portable toaster.

On the other hand, Apple has recently discovered the benefits of selling operating systems unbundled from their hardware. Anyone can walk into CompUSA and walk out with a boxed copy of Panther. Microsoft has been doing this for years, but Microsoft has never had a fixation with selling toasters. After selling the iPod exclusively for the Mac, Apple released a for-Windows version. The same thing happened a week before Panther's release with iTunes. Is the sole purpose of iTunes for Windows just to sell iPods? Recently it was revealed that iTunes doesn't make money, so maybe it is, or maybe it's just to get established in the market.

Apple going toe-to-toe with Microsoft would be one of the most interesting shakeups the personal computer industry has seen. I, for one, would like to see it happen.