We all cheered when Apple began experimenting with community-driven, open source development for its flagship operating system, Mac OS X. But if those experiments are now drawing to a close, should anyone really be surprised?
In his column earlier this year, Tom Yager noted how Apple seemed to be back-pedalling away from open source. Seen through that lens, the recent news that the OpenDarwin Project will be closing its doors looks like just another sign of the times.
Among other things, OpenDarwin aimed to create a functional, fully open source alternative to the Mac OS X kernel, starting from the Darwin source code provided by Apple. But, four years after its inception, it hadn’t made much progress. Why not? According to its maintainers, “availability of sources” and “difficulty building and tracking sources” were among the roadblocks — not to mention “interaction with Apple representatives”.
I can’t say what those interactions might have been, but not wanting to share its toys is classic Apple behaviour. There was a brief period in the 1990s when Apple toyed with licensing the Mac OS to third-party hardware vendors, but Steve Jobs put a stop to that when he returned to the company in 1997. The Mac OS was Apple’s ball, and Steve took it and went home, where it remains.
It seems Steve Jobs doesn’t think he needs anyone else’s input to properly serve Apple’s market. And here’s the thing: maybe he’s right.
Let’s face it; Apple under Steve Jobs has been so successful precisely because it’s so good at what it does. Arguably, each new Macintosh PC and laptop has become successively better since the original G3. And Mac OS X still wins accolades, even from die-hard open source advocates, despite the fact that the parts that really make the Mac OS what it is — including the GUI layer, the Carbon and Cocoa APIs, QuickTime, and so on — were never part of Darwin and were never open sourced to begin with.
So who cares if Apple retreats from open source? There are two ways to build software products, just as there are two types of organisations. You can build your software through a loosely knit community process, democratically accepting input from all sides, or you can do it in a more tightly controlled, top-down fashion.
One need only look at what kind of company Apple is today to guess which method suits it better. Steve Jobs, for all his anti-authoritarian swagger, has far more in common with his good friend Larry Ellison — or even with Bill Gates — than he does with Richard Stallman or Linus Torvalds.
Of course, there is a certain amount of hubris associated with such a top-down approach. It means that all the risk is placed squarely on Apple’s own shoulders.
If the judgment of Steve Jobs and his lieutenants remains sound, Apple will doubtless continue its string of successes. If not, they will have no one but themselves to blame.
Meanwhile, the real open-source developers are out there in their countless thousands — experimenting, collaborating and, slowly but surely, toiling away to perfect their code. They don’t have Apple’s resources or its marketing, or its R&D budget.
But the entire software industry is rapidly being forced to confront the fact that open-source developers are definitely in the race. And, like the tortoise and the hare, they’d better not catch Apple napping.