…something touched me deep inside, the day the music died...

I have (had) a (very) few things in common with Steve Jobs: we were born within a year of each other, we were both adopted, we were both cancer survivors (until last week), and we both worked in the computer industry.

We differed in so many different ways: Steve dropped out of a liberal arts college, embraced the counter-culture, and peppered his career with daring moves. I conformed: starting at the U.S. Naval Academy, transferring to an established (if somewhat innovative) engineering school, married young, and moved between "safe" software jobs over the years.

I had been working in software for seven years, mostly in CAD and computer graphics, when I first encountered an Apple computer — a pre-release Macintosh — as an employee at Lotus Development. I was skeptical for a short time (a few weeks), but I eventually concluded that the Mac represented a compelling vision of the future of computing, and embraced it; or, as some would say, started drinking the Kool-Aid.

I had the privilege of "meeting" Steve twice, in small-group situations. The first time, he and John Sculley met Mac developers at Lotus, shortly after Sculley had been hired by Apple. The second time, Steve visited Lotus as CEO of NeXT (with Conall Ryan, an ex-Loti who was NeXT's VP of Marketing). During that visit, Steve expressed an interest in seeing the product I was working on (Lotus Agenda) implemented for the NeXT platform; I was hooked. I eventually landed a job working on Lotus 1-2-3 for Mac at Lotus, where I was able to pursue this passion, and it was one of the most satisfying projects of my career.

After those experiences with Apple and Steve at Lotus, the Mac (and NeXT, to the extent that I understood it) became my personal reference points for how software should work for users. This orientation was not always conducive to career development. I did land a position with Symantec developing Symantec C++ for PowerPC (a descendant of THINK C for the Mac, which I had used for years), which was probably my dream job of all time; but the Mac tide was ebbing, and I had to find work in domains where my Mac experience and values were not, well, valued.

Four years ago, I landed a job to lead the implementation of a Mac version of a successful Windows product. The project was not a spectacular success, for various reasons, but I was back to the Mac, and anxious to stay. As corporate interest in that product waned, I moved on to a similar situation (a company with a successful Windows product and a desire to succeed on the Mac). I am back in my element, and determined to implement something Steve would approve of.

I am crushed by Steve's passing; in private, I have cried, uncontrollably. We have been deprived of years of his masterful guidance. I have great confidence that Apple can continue to execute as well as it has in the past, but it is undeniable that something is lost without Steve's oversight. Nonetheless, I cast my lot with Apple. (Literally: I have Apple stock I bought in 1997, and I'm still holding.)

I have only a few years left to work in this industry before retirement (I hope), and Steve's advice (as expressed in his Stanford commencement speech) rings in my ears: "Your time is limited, so don't waste it living someone else's life".

I am not going to waste the time I have left.

Four and a half to five years ago…

I speculated that Apple's Boot Camp would (OK, could; I hedged) help Apple boost market share significantly. My rationale was that Mac hardware was suddenly potentially useful to Windows users, as part-time or even full-time Windows machines, and that even a small percentage of Windows users buying Mac hardware would represent a substantial increase in Apple's market share, even if it added only a couple of percentage points to Apple's roughly 4% market share.

Anecdotally, I've seen this happening. At my last two employers, Macs, usually laptops, have regularly been purchased with the intent of running Windows at least part of the time, even by IT types who profess to hate the Mac. Many of these Mac owners started using Mac OS, sometimes for the first time in years, others for the first time ever. I haven't done a comprehensive survey, but it's my informal observation that many are "switching" to Mac OS.

Quantitatively, my expectations have been wildly exceeded. This week, some reports put Apple's market share at over 10%, making it the third largest PC vendor in the US. Apple's share price is nearly 10 times what it was in June of 2005, its market cap puts it at the top of the S&P 500, and it has more cash in the bank than (almost?) any other company in the world.

To be sure, the Mac is not the only driver of this growth; the iPod, iPhone, and iPad are all game changers. But I assert that they are all inter-related; there has been much talk about the iPod (and later iPhone) "halo effect", but it all adds up to Apple's tight focus on their product line.

I see more upside. The iPad is in its first generation, and I expect it to follow the same kind of trajectory the iPhone did. The Mac (and OS X) are back on Apple's agenda next week, and I see a lot of potential there. Mac OS X is way ahead of its developers; mainstream applications that need to support OS X 10.4 (Tiger) are still unable to exploit much of the potential that Apple built into Leopard (10.5) and Snow Leopard (10.6). I have high hopes for OS X 10.7, although I don't have many predictions to offer. A new file system seems like a possibility: ZFS was rumored for Snow Leopard, and it wouldn't surprise me if Apple has something new here this time around.

And there must be something for the new datacenter in North Carolina to do...

A Few Short Observations About WWDC '10 (one hint: don't sell Mac short)

Of course, I'm under non-disclosure, so I'll be brief (and careful):

- The conference was light on Mac OS X, but most app developers haven't adopted some of the most powerful aspects of Snow Leopard yet: Grand Central Dispatch and OpenCL to name a couple. There is a lot of goodness yet to be extracted from OS X by developers; Apple doesn't need to rev it yet.

- Apple is investing big-time in Dev Tools (Xcode), to the benefit of both iOS and OS X. All Apple devices (and OS's) are positioned to take great advantage of emerging technologies.

- Having seen Apple's development and debugging tools for iOS devices (iPhones and iPxds), I can't see how Android can its multiple manufacturer's can keep up. Sure, they can build a lot of devices and throw them at the wall to see what sticks, but will they ever be able to fine-tune these resource-constrained devices to the same extent that Apple can, with it's tightly-managed platform and well-matched tools? We'll see...

Apple, Flash, and Platform Control

Gruber @ DaringFireball, as always, has already nailed this on the head, and Mark Bernstein (whom I've never encountered before), has driven it home, but I think I have $.02 and a little history to add.

How Should Mac Apps be Distributed?

John Gruber has a thoughtful piece on Daring Fireball (easily my favorite blog), reflecting on another "thoughtful piece" by Alexander Limi; the topic is how Mac apps are installed on the Mac and how the process might be improved.

I have two thoughts on the matter, which touch on topics raised by John but which I had formed before reading his piece. (You'll have to take my word on that.)

1) The Mac should have an interface dedicated to finding and launching applications, probably accessed from the Apple menu. The interface should take guidance from the iPhone home screen(s), displaying all of the applications accessible to the user (regardless of where they are installed), organized to the user's preferences. (Maybe a third-party app opportunity. Hmmm.) It could be layered on top of the existing scheme, so none of the flexibility of the existing model would be lost.

2) How about a Mac app store? For those well-behaved Mac apps that don't require an installer, this is a no-brainer, and I suspect many of the factors that dictate an installer (contributions to the Frameworks, Application Support, or other common destinations) could be accommodated fairly easily. This would eliminate most of the room for user error cited in the two blogs.

Might be a good use of that new datacenter Apple is reportedly building...

Microsoft can catch up?

In this interview, Steve Ballmer suggests that Microsoft can catch up to Google in search if it only persists, as it did with Windows:

A. No. No, Windows 95 was a basically mostly interesting and successful concept that came to full fruition, right? This is not Windows 95. It’s more, I don’t know —

Q. Windows 3.1?

A. 3.0 maybe even, right? It’s more, hey, we’ve had some early tries, kind of like you might call Windows 1, and I think there was something called Windows 2 in there, and Windows 386 in the late ’80s, but it’s far more like Windows 3. People say, aha, I see the vision. It pays off but it won’t fully pay off the vision in its first incarnation.

But it’s like Windows — the most important thing I just said in all of that is it’s important like Windows is important. It’s important like something that we really care about, we really think about, we’re going to stay persistent with, we’re going to invest in.

If you stop and think about it, Windows 95 came 12 years after we started working on Windows. We’ve been working on search five years. I’m not saying it should necessarily take 12 years, but in a sense what we’re trying to do is accelerate the pace, and see if we can’t get there.

This is bogus to the point of being laughable. Google (the company) dominates search. In the 80's and 90's, PC operating systems were dominated by... Microsoft.

There is a common perception that with Windows 95, Microsoft finally "caught up" to Apple and Mac OS; Ballmer seems to be referring to this.

But the reality is that the Mac was never ahead of Microsoft's OS in terms of market share. The IBM PC, running Microsoft's DOS, blew past the Apple II in 1982-1983, well before Apple rolled out the Mac. PC's outsold Mac's all along; I remember all too well, as a Mac developer (by choice, when I could) working for Lotus during those years. MS-DOS had dominance when Windows 3.x was finally able to start displacing it; that's why these were Lotus' "golden years", as a seller of MS-DOS software.

So if Windows 95 represented Microsoft "catching up" to Apple, I'm left to conclude that Ballmer is asserting what a few of us know all along in the 80's: the Mac OS was superior to Windows.

Can the Mac survive success?

[Old entry that apparently never made it to the blog. Still relevant.]

This column, way more articulately (if somewhat obscurely), raises a point I have hoped to worry about for years.

From the beginning of the Macintosh, Apple promoted user interface guidelines to encourage developers to accept and adopt common look-and-feel for common operations. The premise was that users of applications should be presented with familiar interfaces for commonly performed operations, so that the learning curve for adopting a new application would be as shallow as possible.

A simple case in point is file-open and -save dialogs. Apple strongly promoted adopting a standard user interface to this common operation, so that users of new applications would not have to learn new ways to perform a common operation. This was drilled into early Mac developers, and accepted as common wisdom.

By contrast, when MS-DOS developers became Windows developers, there was less of an ethic for consistent user interfaces. I distinctly remember Windows developers at Lotus seeking to implement "better" file-open and -save dialogs than those standard to Windows.

This sounds trivial, but it's fundamental: in those formative days, Mac developers defined themselves by how well (and, sometimes, creatively), they could adopt the platform standards, and Windows developers focused on how they could perform common operations "better" (i.e., differently).

As time went on, the Mac community, users and developers alike, embraced the guidelines and held developers accountable. I remember this as a developer of Lotus 1-2-3 for Macintosh, presenting the product to users at MacWorld Expo and listening to their critiques. I saw it again as the architect of Symantec C++ for Power Macintosh, representing the product to our customers (other developers), and being held to the same standard.

The Macintosh community, users and developers alike, mandated adherence to platform standards, and became informed judges of what was acceptable innovation. From my point of view, this worked to benefit of all Mac users.

Today, the Mac is being adopted by large numbers of former Windows users, and there is a lot of pressure to accommodate those users by addressing their sensitivities. In other words, there is a point of view that the Mac platform should adjust itself to the expectations of the legions of former Windows user who are now adopting the Mac.

I understand the logic, but I think it's a horrible idea.

People who have been attracted to the Mac have, by definitions, accepted a certain degree of change. I don't think we, as Mac developers, should give into the temptation to accommodate the expectations of neophytes.