A look into Apple's Darwin strategy and its impact on Linux
A new-age strategy is falling into place.

02-12-2.000

 

It's fun when some pieces of news just click together into an idea. It's funnier when you like the idea, of course. And that's what happened this week with two pieces on John Carmack, the id guru, and Apple's new Finder.

Ever since it was first known that the new Finder in Mac OS X was going to follow in NeXT's steps, rumor and wishful thinking alike were that along with the new, "hard", Unix-ising system there would be a more traditional desktop view to maintain the Mac's advantage as a more intuitive, low-learning-curve option. The Aqua presentation froze those expectations and came close to panicking the Mac user base, except where people thought it a water-testing exercise by Jobs.

Now the MOSR rumor is that the newest Developer Release has a Finder which can be used in two ways: as the NeXTish "File Viewer" featured in San Francisco... and with a "View as Desktop" option that keeps the fundamentals of the Mac interface unchanged.

The other piece of news came from X Appeal, the site that keeps me up on Darwin and related Open Source technologies. John Carmack has hacked a Unix graphic environment onto the threadbare Darwin OS. It was something we all saw near, the coming of age of Darwin as an usable OS, and that Carmack (of all people) has suddenly brought much closer.

The significance of these informations is enormous for Apple. Let's assume both of them are true. And let us think why they should be.

Apple's reasons for a twin-track Finder are very clear: they would allow consumer users to keep using the Mac in the way that made it famous without losing any of its new technologies, while at the same time allowing more experienced or demanding users to manage their systems in a more efficient, fast and frills-free manner. Apple could keep its customer base and start clawing at some other system's users thanks to a very rounded, polished, well-defined and documented, cutting-edge OS that could do everything any other Unix could, with better interface and lots of applications (Mac, NeXT, BSD) to keep up with NT.

Carmacks reasons... we can guess at them too. Darwin is a rare specimen, a top-notch free operative system that evolves in symbiosis with a commercial version. It lacks a lot of frills, but has all the muscle. It's no newbie, since its basic technologies have been around for a while. It can be altered to your heart's content within the Open Source rules. And there's lots to be done in it, a lot of landmarks to claim. It's a hacker's dream, like Linux is or maybe was (before turning into a packaged money-spinner for a few companies). If only it were complete.

That was no small fault, and that's what Carmack's started to correct. If Darwin can be turned into an usable OS, its evolution will rocket ahead. It will have more than a theorethical interest: it will be the Linux project all over again.

And the nexus between the news should be quite obvious by now. Mac OS X is basically Darwin dressed to kill plus some very interesting add-ups. Darwin's qualifying for evolution (And that is how I read Carmack's practical statement) means two important things, in my opinion.

The first is that we'll soon see Darwin take up a place as a Unix system of its own, eventually to rival Linux on PPC systems and maybe further. The second is that Apple's Mac OS will take the place of Red Hat & co. in this new setup, only adding a significantly higher value through the development and integration of new technologies at which it's a proven leader. A systems administrator, a critical systems designer, anybody with very high OS expectations could perfectly use Mac OS X with its File Viewer (and related technologies) configuration, as a substitute to cooking his own Darwin... because Darwin's advantages would be right under the hood.

This strategy was the only serious reason for Opening the Darwin code, and I can only applaud at how it is being pulled off.

 

Back to the Menu