Andy Hertzfeld has stories to tell. Dozens of them. And if you ever owned a 128K Macintosh, aspired to own one, or admired the work behind that extraordinary box, Hertzfeld's new book Revolution in the Valley is a charming and picaresque trip through his personal experience in helping give birth to the Mac.
The book is an outgrowth of Hertzfeld's Folklore.org Web site, which he started in July 2003 to relate the pieces of the past that have never been told, or at least not told at length. The site itself is a demonstration of software he's developing to let people tell stories collectively through recounting and annotation. Because Folklore.org continues to operate on the same basis, if you find errors in the book or take issue with Hertzfeld's interpretation, you can visit the site and comment on the particular anecdote.
Hertzfeld has had an interesting career since leaving Apple after the first Mac shipped in 1984; he also has just a handful of scores to settle. Most of the time, he comes to praise, not to bury. The book revolves around the nitty gritty of producing a computer that had to pull off many dozens of unique tricks in hardware and software to work at all. Apple previously and even simultaneously suffered notable failures in putting too much innovation in one box - the Apple III, the Lisa - and being able to deliver at a reasonable price and performance.
(Don't flame me, Lisa fans: Steve Jobs raided Lisa team members and innovation to squeeze into the Mac, helping to doom the earlier machine. As Hertzfeld recounts, Lisa architect Rich Page screamed during an early Mac/Lisa cross-team briefing, "You guys don't know what you're doing! ... The Macintosh is going to destroy the Lisa! The Macintosh is going to ruin Apple!" And for you Apple III fans... what am I saying? There are no Apple III fans. Although I did spend some time entering data into an Apple III around 1980, however, it didn't give me any profound insight into the machine.)
Hertzfeld didn't compile a straightforward narrative for the book, and it shows its roots as anecdotes and short stories on a Web site in two ways: first, it meanders quite pleasantly around amusing stories, doubling back into a past that's already told to extract another nugget. Second, Hertzfeld used some of the comments left on his Folklore.org site to annotate his book, including those that contradict or critique his memory. The book would have benefited from more of this back-and-forth, actually, as a number of comments on the Web site are quite pointed, poignant, or just credulous about the accuracy of certain stories.
The Hacker Hero -- The book does have a hero and a villain, and a few lesser good and evil figures. The hero is Burrell Smith, the hilariously weird hardware genius who came up with many of the strangest and most successful ideas of squeezing more performance out of the Macintosh motherboard. He also should earn Mac owners undying love for trying, unsuccessfully, to insert an expansion port and upgradable RAM into the first Mac.
Jobs and the Mac's conceptual father Jef Raskin agreed that the Mac shouldn't have a slot because it would add cost and complexity. Smith was told by Jobs that "there was no way the Mac would even have a single slot." But Hertzfeld notes that "Burrell was not easily thwarted... After talking it over with Brian [Howard], they decided to call it the 'diagnostic port' instead of a slot, arguing that it would save money during manufacturing if testing devices could access the processor bus to diagnose manufacturing errors." But the engineering manager Rod Holt spotted the subterfuge. "That thing's really a slot, right? You're trying to sneak in a slot! ...Well that's not going to happen!"
Ah, well; we had to wait until the Macintosh II for a full-fledged slot. And, surprise, a company founded by none other than Burrell Smith - a little firm named Radius - took incredible of advantage of that slot to offer advanced graphics cards that helped establish the Mac's early preeminence in desktop publishing and illustration. (When I worked at the Kodak Center for Creative Imaging in the early 1990s, we had at least a few hundred thousand dollars in Radius cards and monitors.) Hertzfeld notes on his site in response to a comment that Smith is now quite private and has been retired from commercial work since leaving Radius many years ago.
Other members of the team have also left the technology realm. Bill Atkinson, for instance, became a full-time photographer after many years of intense work. I met Bill in 1991 at the Center for Creative Imaging where he was attending a special design invitational along with John Sculley and a host of designers, photographers, and illustrators. (That's where I overheard a Kodak employee, while demoing a terrible piece of software to John Sculley, explain how keyboard commands were better than mouse commands. "No," Sculley said quietly, "they're not.") Hertzfeld's picture of Atkinson shows him as rather prickly and sensitive, although that's partly because Atkinson's role in the Lisa was largely ignored, and he didn't want to be pushed to the sidelines again.
Other minor heroes include Bud Tribble, who at the time was pursuing a medical degree while writing memory management software. (Tribble left Apple, later joined Hertzfeld at Eazel, and eventually returned to Apple a couple of years ago.)
The Manager Villain -- You're expecting me to say Steve Jobs, right? Wrong.
The villain of the story is Bob Belleville, the Mac's engineering manager for Hertzfeld's last couple of years at Apple. Hertzfeld seems least fair in presenting a pretty one-sided and nasty picture of Belleville. He may have been a poor manager or out of his depth - I don't know whether that's accurate - but he's the least fleshed-out person in the book. Everyone else emerges as quirky and interesting, even when they're screaming at Hertzfeld. Belleville is his bete noire, and a nasty cipher.
Also interesting is that Jef Raskin appears quite positively in the book. Raskin has spent a lot of time since leaving Apple well before the Mac shipped trying to prevent Apple and others from erasing his name from the history books as the conceptual originator of the Macintosh's core concepts. Raskin deserves to be placed front and center as the person who pulled together ideas that he had been writing about and lecturing about since the 1960s into a single project with funding. The fact that Jobs stripped him of control and his role, and that the ultimate Macintosh has significant differences from what his general vision and specific hardware choices were, shouldn't lessen the appreciation of his role.
Hertzfeld's recollection of Raskin is as a fun and creative manager with an imperious and professorial manner who helped bond a team together around a common and unique vision. Without Raskin, as Hertzfeld relates it, the Lisa would have been Apple's flagship with incremental improvements, rather than revolutionary ones. Jobs's spearheading of the Mac led it to success because he was constantly overriding and micromanaging the project for good or bad - but the project received staff, resources, and his laser-beam attention.
Steve Jobs ultimately drove Hertzfeld to distraction, and also appears as a paper-thin caricature. But that may be the only Steve Jobs that anyone who works with him gets to know. Jobs pushes his staff to work crazy hours, makes last-minute changes, and pursues insane technical decisions. When Smith shows a blowup of the blueprint of the latest motherboard layout, Jobs says, "That part's really pretty... but look at the memory chips. That's ugly. The lines are too close together." When an engineer points out that no one will see the board, Jobs replies, "I'm gonna see it! I want it to be as beautiful as possible, even if it's inside the box. A great carpenter isn't going to use lousy wood for the back of a cabinet, even though nobody's going to see it." (It's clear Jobs was never a carpenter.) He makes the team design a pretty board, and when it doesn't work, they revert back to the functional design.
More typically, Jobs pursues dead ends, such as an Alps-designed 3.5 inch floppy disk drive; fortunately cooler heads at Apple maintained a back channel to Sony (who provided the final 3.5 inch drives), which involved sometimes hiding a Japanese engineer in a closet in an Apple building when Jobs unexpectedly popped by.
On the other hand, Jobs does make a number of key decisions along the development process that make the Mac what it was, from case design to aspects of its performance. The man couldn't stop poking, but he did bring out the best in his engineers, a trait that he has apparently retained to this day.
Hertzfeld describes how he and a few other key Apple people had dinner with Jobs after Sculley organized the board coup that removed virtually all of Jobs's control of the company, despite being the titular chairman of the board. It's the most human picture of Jobs in the book. And it's clear from the story that Jobs was never going to be in a position to be fired by anyone ever again.
Bill Gates also comes across as a villain, appearing frequently in the guise of Coyote, twisting words and using his magic bag of tricks to seize patents and ideas.
Hertzfeld's Journey -- It was exciting to read Hertzfeld's first-person accounts of developing the software for Thunderscan, a scanner-head replacement for the ImageWriter's print head built by a company that needed his help in making it fast and slick; and Switcher, the original context-changing tool for running multiple programs at once on the Mac.
I remember the excitement of owning my first Macintosh Plus, and remember buying an upgrade toolkit with more RAM (static strip, long Allen screwdriver, and case cracker) - and then seeing the glory of the signatures on the inside of the case as I put in a whopping four megabytes of RAM.
I can't say that Hertzfeld doesn't have an ego, but most of the stories he tells are about other people. He doesn't put himself front and center except in some of the most painful incidents, which typically involve Steve Jobs either demanding something of him or putting him in a position where other people are asking him not to listen to Jobs, his nominal uber-boss.
Hertzfeld ends the story before joining Radius, helping to found General Magic, and then being back with many original Apple developers at Eazel. We don't quite know how the last 20 years treated him because the universal interest in Apple doesn't necessary extend to those other firms. And perhaps the statute of limitations on telling the blunt truth (as he sees it) extends back 20 years.