Adios Avie

Friday, 31 March 2006

This week’s announcement that today will be Avie Tevanian’s last day at Apple has brought out the conspiracy theories in full force. E.g. this write-up at Metafilter:

Avie Tevanian to leave Apple. Long regarded as the brains behind OS X (and NeXT before it), Tevanian’s unexpected departure is “to pursue other interests,” and his last day is Friday, 1 day before Apple’s 30th anniversary. Noted tech curmudgeon John C. Dvorak recently claimed, to much ridicule, that Apple was going to ditch OS X and move to Windows. Coincidence? Or has Steve’s famous temper gotten the better of him again?

Now, calling John C. Dvorak a “noted” anything other than idiot is a sure sign that this blurb isn’t exactly meant to be taken seriously. But, still, I get the feeling there’s a bit of uneasiness out there about what Tevanian’s leaving means for Apple and for Mac users.

The answer is: not much.

To be sure, Tevanian was hugely influential in the creation and design of Mac OS X, and before that, NeXTStep. It was under his management that Apple shipped Mac OS X 10.0 through 10.3 (not to mention Mac OS 8.5 through 9.2, which were fine releases and sent the old Mac OS out at the top of its game).

But in July 2003, he stepped aside as Apple’s vice president of software engineering and took the new title of “chief software technology officer”. By all accounts of what he’s done since then, his title might as well have been “guy who’s worked his ass off and done a great job and now has decided to sit back and do nothing for a while”.

I asked a few engineer friends at Apple whether my perception was correct — that Tevanian has had one foot out the door ever since he stepped down from day-to-day management of Apple software engineering in 2003, and that the news that he’s leaving the company completely isn’t really a big deal at all. They all agreed, more or less, that Tevanian has had both feet out the door but just hadn’t yet turned in his keys. No one I spoke to at Apple has any idea what he’s been up to the last three years.

When he was vice president of software engineering, everyone knew what he was up to. One reason I think Tevanian was successful is that he was an engineer at heart, not a manager. He truly understood how Mac OS X was designed and how it works.

In short, Tevanian’s leaving has nothing to do with the switch to Intel processors, or the continued growth and success of the iPod, cockamamie theories that Apple is switching to Windows or that Microsoft is switching to Mac OS X, or anything else that might be construed as big news. It’s just a case of a very smart guy who decided a couple years ago to move on to something else, and tomorrow is moving day.

Bertrand Serlet, Tevanian’s long-time right-hand man, took over as vice president of software engineering in 2003, and seems to be doing a fine job. Mac OS X 10.4 was developed and shipped under his leadership, and it’s been every bit as successful as the 10.1 through 10.3 updates that shipped under Tevanian. Serlet also seems to be much more popular with the Mac developer community, or at the very least is not prone to pissing all over them.

I’m certainly on the record as not being a Tevanian fan. All the criticism I levied against him after he stepped down in 2003 still stands.

But I also have a lot of respect for him. Criticizing Tevanian is like criticizing a sports player who won a bunch of championships: the bottom line in sports is winning, and the bottom line at Apple’s software division is shipping great software. Tevanian was responsible for shipping 10.0 through 10.3 and they were all great achievements. That alone qualifies him as a winner.