logo
Published on developer.* Blogs (http://www.developerdotstar.com/community)

Andy Hertzfeld's Apple performance review

By Edward G Nilges
Created 2005-02-04 23:02

I first read of Andy Hertzfeld’s “bad performance review” in 1986, in a book of interviews “Programmers at Work” edited by Susan Lammers and published by Microsoft Press.

The story is quite a shocker, for the naïve at least (as I was in 1986). It has been reconfirmed with details by Mr. Hertzfeld in a new book he’s written, Revolution in the Valley, published this year by O’Reilly.

Andy Hertzfeld was and is a genius who designed and coded key software for the first Macintosh. Whereas Steve Wozniak was a hands on guy, Steve Jobs never actually designed software or hardware, but instead exercised his abilities to recently create, through others, great Stuff from the original Mac to the Ipod and the mini Mac.

But based on what I knew of Apple culture in the 1980s, and based on my experiences with New Age thought in the 1960s and 1970s, I was already suspicious about Apple’s culture: it reminded me of how in the Sixties peace and love merely tie-dyed the basic contradictions between races and social classes in America.

The “Sixties” were a label for the anarchism of the lower middle class which replaced the Civil Rights movement and genuine change in the cities with a highly individualised search coupled with overemphasis on identity politics. As such, the “Sixties” left middle class, college educated males feeling somewhat left out, with their space-program technical interests uncelebrated in the 1970s orgy.

The wimmen were ripping off their bras and not shaving their armpits as they do so refrain today in Paris. The homeys were Superfly and the age old sexual threat to white males in America was in full blast, concealing from honky and homey the basic fact, that the latter were being sold down the American river. Techie males felt isolated, estranged, their plight only addressed by the appearance of steadily cooler and cooler gizmos such as the HP 65 and the Apple II, into whose arcana they fled.

The idea that a device could bring about social change was strange to culture theorist Theodore Roszak as early as 1986. In “From Satori to Silicon Valley”, Roszak said of the idea that distributed and powerful computers, which would set the pace of work, were the dawn of the real Age of Aquarius “how could anyone believe anything so unlikely?”, and any end user, being screamed at merely for changing her screen saver to a bunny, can relate.

Nixon's negative income tax, giving people money when their income is below a certain level, certainly would seem a better way to fund, to provide a base for the superstructure of, the Age of Aquarius. Surfers would surf in Pacifica, and be fed, and less overall competitiveness would foster fewer Surf Nazis.

This idea, as any Sixties survivor like me (who arrived just soon enough to lose his virginity in 1969 on the same weekend the astronauts landed on the moon) can attest, has its own naivety…for as my fat, and rather more prolix pal Theodore Adorno would say, the “whole” is untruth.

In 1970, I didn’t want a handout. I wanted instead to write an Algol compiler while smoking unfiltered Camel cigarettes. The very idea of sitting on a beach with my girlfriend in an unstained Malibu was a non-starter. She left, in a purple dress, Chicago for Malibu in 1969 with me standing on the platform feeling no end of a damned fool, for I knew that she was a (the) Goddess, who always wore a bikini under her miniskirt because she loved the beach.

(Ah yes, dry your whiskers: the ladies forget us, we not them).

It was indeed unlikely as Roszak said, that harnessing people to devices would bring about the Age of Gold, and the chicks of 1969 were onto something, as was my cat (who went crazy because I wouldn’t stop coding, and go to bed, and allow her to sleep on my face), and, a few years later, my ex-wife, who’d read to me from D. H. Lawrence in order to get me To Change.

But here’s Andy’s story, with digressions.

After working ninety hours a week on critical and brilliant software, Andy was called outside the Apple building by his supervisor-in-name, a Bob Belleville, for even in 1983, the year these events took place, Apple had formal performance reviews.

Not too many people are aware of the origins mechanisms including performance reviews and the related phenomenon of instructor evaluations by students.

As far as I can tell, performance reviews in writing and according to defined procedures were an artifact of civil rights and equal opportunity legislation of Lyndon Johnson’s Great Society. When I started looking for a job at 17 at my father’s not-so-gentle importunings, the Chicago Tribune had separate help wanted ads for men and for women, and favoritism along racial and class lines was rife. But in the 1970s, increasingly, I and other corporate types had performance reviews presented as fair, which usually (up until Reagan took office) were pretty fair.

Instructor evaluations by students have a more surprising origin as far as I can tell, for they seem to have arrived in America from China, by way of France.

In the middle of the 1960s, Mao instigated the Cultural Revolution in China in order, in part, to secure his power against competitors including Lin Piao and perhaps even Chou EnLai. One mechanism Mao used was to turn the students of China against middle authorities and teachers so as to deprive an adversary of a base in civil society, and students were increasingly during the 1960s encouraged to speak up, and to call their teachers either stupid, or counter-revolutionary, or both.

Julia Kristeva and other French intellectuals of the 1960s, following Simone de Beauvoir’s lead in the 1930s, visited China for lessons applicable to the West, and, as far as I can tell from the record, imported the idea of student evaluations (at first in the form of Sorbonne rhodomontade, later formalized) to the elite schools of France.

Which is where my confederates in Students for a Democratic Society and at Roosevelt University seemed to have imported the idea, and part of our demands in campaigns for popular professors (such as the left-wing historian, Staughton Lynd) was a formal system for instructor rating.

In 1971 a programmer was engaged by my university’s Anthropology department to write the code for this system, but a year or so later I discovered it was a complete mess, which assigned the worst rating to the instructor, silently, on invalid input codes on punched cards…in fact a precursor of the infamous “hanging chads” of 2000. I communicated this discovery to the Dean of Faculties, who dropped the program.

Today, the liberatory impulses that in fact underlie both performance reviews and instructor evals, fairness in the former case, fairness and egalitarianism in the latter, are at a discount in American society, and as a result both systems are monstrosities and were so even as far back as 1983.

In his 2004 book, Andy writes that Bob Belleville, his supervisor, called him outside the building and there said “well, Andy, you’re not going to like hearing this, but you’re a big problem on the software team”.

Andy was shocked since the “team” liked and respected him, as did, apparently, Jobs himself. Andy had made the Macintosh run as a computer and not as a pretty box.

He made a protest that many programmers will be familiar with making, that his technical work was first rate; many of us programmers, myself included, were attracted into computer science professor told us that our grades would be perfectly objective because our code either worked, or did not.

We’d escape what we saw as the falsity and compromise of careers as lawyers or even doctors. My understanding is that during the same epoch, men in Russia like Natan Scharansky, today an Israeli minister, fled at first into scientific, mathematical and computer careers to escape a corresponding falsity. The career of Mikhail Gorbachev, scorned as he is today, shows the limitations of a technical dodge, for Gorbachev seems to have taken the bull by the horns.

Bob gave Andy’s technical protest the response which will be familiar, to programmers who are honest with themselves, as a scripted response in a system unfamiliar with its very origin, and this was that Andy was “insubordinate” and did not work well with the “team”.

In Bob Belleville’s scripted response, “ideology” replaces technology in a post-enlightenment scenario, because the techie “feels” (with some justice) that he’s incorporated Enlightenment, only to have that feeling taken away in the name of a reified Enlightenment.

In a Communist society such as Mao’s China, state needs and those of the people are assumed without much discussion to be a higher Enlightenment which incorporates and succeeds a 19th century engineering enlightenment, to which it seems Apple techs were wed. In a capitalist society, reified business needs perform the same ideological function.

But keep in mind that this dialectic (this dialog) wasn’t taking place at some bank, like Northern Trust, in Chicago’s loop, or in Albania. It was taking place in the poop of the liberatory pirate ship and at the cutting edge. Technicians think “logically” and as such are unfamiliar with the reversal logic (the “negative” logic) of embedded dialectics, where human phenomena thought buried return in the manner of the Spanish Inquisition in Monty Python: “nobody expects the Spanish Inquisition! Aaargh!”

Further material from Hertzfeld reveals a sort of ascending weirdness thought Dark Californian by many but which can be better recognized as The Way the World Works and by a reading of Marx, for Hertzfeld (after bursting into tears during the review in the cool, soft and sinister California night) went to Jobs…which had the predictable effect on Belleville, who lied about the entire incident.

Even at that early date, Apple’s lifeblood was investment capital, and investors in America want to see proper lines of authority and subordination…often for a damn good reason given what E. F. Hobsbawm calls “the anarchy of the lower middle class”, and, in Apple’s case, the extreme youth of its top developers (one of whom was hired at 14).

The Apple kids were already products of the 1960s, in some cases the kids of early Seekers of the beatnik and early Acid Trip era and whatever else you might say, my generation (and I do not exclude myself, being a Baby Boomer and having done a C minus job of fatherhood) was far too self absorbed to create true multigenerational communities with initiation and with that sort of fatherhood that makes a young man a good follower.

This combined with multi-generational traditions of real anger against the American game as played since the Civil War to create the American culture of programming…in which programming great code becomes a family romance, in which the disorganized long haired (or shaven skulled and tattoo’d) kid outcodes the conformists in their grey flannel suits or their Gap casual.

“Lieb Vaterland, magst ruhig sein”: at this stage of the romance, he shows himself loyal after all to patriarchal values while separating himself from the follower in hair length or hair lack.

However, this culture has consistently, from both the record as in the case of Apple, and in my own personal experience, collided with the Real World in a scripted, unchanging and Groundhog Day like way, covered up only by programmer shaming, in the next phase of the romance.

I experienced the same thing as Hertzfeld when working close by, at a company set up by Nortel for software and telecom R & D based on a “free” engineering culture. At this company, almost exactly at the same time as Andy, I was called in for a performance review after reengineering a compiler no-one else wanted to touch, and I was also told that (1) there wasn’t any problems at all with my tech work but that (2) I had an insubordinate attitude and also made strange noises of contentment, or dismay, while working, from “eep” to “fanorkle” to a triumphant “bugna” (these noises were all based on the Early Teletubby language I would speak to my kids before I left them).

I didn’t cry, but I did get drunk (at Dinah’s Shack, if memory serves). I lacked Andy’s courage. I also refused to even sign the review and filed a formal protest. Fortunately, the supervisor of my supervisor recognized the problem and its realities and reorganized the group, and I was later able to write a new compiler for an effective group and recover self-esteem.

However, the company ground on to become basically a failing cash cow for Nortel, on which was made increasingly insane demands for development and not research. Much later, Nortel, based on its experience, emphasised buying research results but this strategy was highly flawed, and in recent years Nortel and even its executives themselves have been scorned for nondelivery, and fudging results. Their original sin was first to regard the creativity of its core team of engineers up in Ottawa in an unimaginative dollars and cents way, and to foster a culture of secrecy and competitiveness.

In this culture, parallel to Apple, the engineering age of gold ended sharply during the first Reagan recovery of 1983. During that era, engineers were given schizophrenic messages typically in two-part harmony: (1) you are a great technician BUT (2) you have a bad attitude.

Of course, if you design boards or codes with no earthly application to real “business” needs, you AREN’T a great technician. As both Steve Wozniak, and Edsger Dijkstra knew (but many compsci profs don’t), there is NO PART of computer science that is pure theory in and of itself, in the way there is a pure physics, a pure math, or a reinen philosophy, a critique of pure reason.

And if you put in ninety hours a week, an objective member of the jury would have to concede that it sure sounds like you have a good attitude in a society where cab drivers can’t be bothered to learn their routes because that would cut into their drinking time.

The “script” of a performance review is cloven systematically by a false opposition between a technology characterised in its pure state as “ivory tower, not practical, overly insistent on its Tao or way” and a mystified set of Eleusinian “business needs”. I spent long and weary years in the 1980s trying to figure out what these needs exactly were, in some formalizable way, and then I gave up, and stopped drinking…at which point everything cleared up.

I came away with the realization that I was “powerless” over the way the world really works without in the least implying that I had to like it, or think it just.

A sort of Grandfather Spirit came to me those days, and driving home from work on 101, I would think of my mother’s father, for whom any idea that the world of the 1930s, the world of fleabag hotels for working men only, the world of the shape-up and the six month job search, the world of baseball and boxing, was at all set up as fair or meaningful was like, gimme a break. He’d struggled to keep his three daughters in shabby but respectable condition up in a dying farm in Massachusetts.

He of course voted for Democrats, any old Democrat, who knew how to sing at wakes, from Tom Foley to Ted Kennedy and did not live to see Reagan (being lucky). Maddened by false promises and soured by true miseries, his children and his children on my mothers side became Reagan democrats where they did not become academic success stories.

Strangely, he did not seem to condemn me for leaving the kids but acknowledged that just as his own father had come down from New Brunswick and he’d gone to New York, a man must leave his family, some times, to save it. Mostly, he just smoked, in my memory, and asked occasional questions.

I came away from the Valley with a very healthy skepticism about New Age, New Era Potato Chip promises that we’re (courtesy of capitalism’s evolution) now “free” to Be All We Can Be. The fact is, Tricky Dick Nixon (who also invented the food stamp) was right about domestic policy in that the old bastard knew that more and not less economic equality will in the long run really empower us to be everything we can be, unless we just get drunk on Saturday night.

You see, Nixon was very much a man of my father’s generation, and aside from sharing a sour name that implies negation (for both Nixon, and Nilges start with an Anglo Saxon and Germanic fragment of utter negation), they both were from an America in which the sons of well-to-do men (like myself, for my father was a neurosurgeon) were actually guilty about their economic privilege.

George Bush was a rarity, for he was foolishly proud that his Daddy had gotten his bonehead son into Yale. I was a more common type, who sang along with Mick Jagger’s Factory Girl and one of my early girlfriends was from a group of friends, from the declining Northwest side of Chicago, called the Dead End Kids.

Theodore Roszak would be puzzled how to reconcile Hertzfeld’s experience with politics in general and to Roszak, the idea that a device would liberate is to laugh.

The Apple is a great computer…that in many ways has destroyed opportunities for struggling sorts to draw images and use Zip a tone to snag commercial art jobs in Chicago’s loop, while creating millions of other opportunities.

The economist, Joseph Schumpeter, calls this “creative destruction”, and, since the whole is the untrue, we simply cannot know, for any one incident of creative destruction, whether on balance it’s a complete wash.

Now, as a programmer, I simply cannot insist (as I may have insisted in 1972 once or twice before sobering up) that the IBM 1401 architecture was ideal, and that we should just evolve with that paradigm, creating, year after year, greater and greater software for that crazy system. John Haanstra, an IBM executive, tried to insist on this course and was given the right-about by Watson Jr. and Vince Learson. In the 360, compatibility with the 1401 was an afterthought.

Likewise, Jobs’ genius and the way he avoided the messy deaths of firms like Osborne and Ohio Scientific was his clear insistence on his own self-interest. This included communicating to the people who “felt like” they created the Mac that thank you very much, but your product is now my property, and that of the investors in my firm, under the rules.

For it’s clear from Hertzfeld’s 2004 account that despite the fact that Andy felt he could appeal to Jobs after the Belleville auto da fe, and despite the fact that Jobs seemed to be quite on Andy’s side, ordering Belleville to rewrite the review, the game remained the same, and certainly no ordinary programmer is well-advised to appeal a bad review to higher management: in my own case, I simply adhered to the rules, and refused to sign, without appealing to management.

For Jobs’ final comment still leaves Mr. Hertzfeld puzzled. At a party in Andy’s honor, Jobs said “the thing I like about Andy is it’s so easy to make him cry”.

This was, I think, unacceptable, no matter how much money Steve Jobs has (and Andy is probably pretty well fixed). It's also par for the course in California, where childishness extends into late middle age and beyond.

In California, with kids to support along with my ex-wife, and with a spiritual mentor, I’d learned to think before speaking and not to cry except alone. Today, I am out of tears, like Mick Jagger.

But it is important for men to acknowledge the Fact of the Bully, and the Fact that much of technology and society in general is based on the surplus, and sado-masochistic, joy in simple one-upmanship. Gerald Weinberg protested this mechanism of control (for that is what it is) by showing how it damaged technical ends, but it's pretty clear that to the men at the top, technical ends take a back seat.

It’s important for programmers, who in many cases delight in work that the elite think clerkish and dull, to realize that like Hollywood screenwriters they won’t get the girl and will be subject to abuse that in some cases is off the charts…because of society’s and the elite’s fear of mere writing.

We “code” ourselves as males, associated with hard things, resistant things, and our family romance is to the tune of a Watch on Rhine: “dread Fatherland may’st peaceful be, thy faithful Sons will watch o’er thee”. And it’s true that (for example) the obvious way to parse and to execute 1-2+3 is in Dijkstra’s sense cruelly incorrect. And then, at the early stages of our career, society seems to confirm this by throwing money in our direction.

But when you’re bullied for having an “insubordinate” attitude, when your skills are pronounced obsolete, when you’re “not a team player”, or when you’re laid off “for a good reason, a bad reason, or no reason at all” despite the fact you owe back child support (that secret shame of a Puritan’s delight), think only of the great Andy Hertzfeld who made the Mac run and who seems to retain that naivety which Apple was so supposed to celebrate, to render above all safe.

The world should, in my opinion, be made safe for Parzival, the perfect or reinen fool who is raised by Herzeleide to think well of all he meets on the road. Such would be an indicator of Redemption. But in Wagner’s opera, Parzival merely becomes the leader of the knights in a false reconciliation by Wagner to his times, the falseness of which addicted the Germans to far more monstrous lies.

I’m deadly serious, here, for when you think of the waste of human potential, geometers dying in burning cities and my homeys at DeVry taking Associate degrees to become security guards, you ask an obvious question. Are we not in deep doo-doo world wide simply because so many people are sacrificed? So many people made to feel small, like Mr. Hertzfeld, as part of a surplus repression?

Ask why, if you are called “insubordinate”, you then work ninety hours to do what you’re told and ask if the insubordination is not actually just back-talk…without which the company is flying blind.

I was asked at a contract job a few years ago to stay on, but warned that (along with one of the Indian chaps) I had an excessive tendency to question management control. I THOUGHT, silly me, that I was presenting alternatives as a "consultant". But by this date I'd read Hegel and I realized that if social contradictions exist, one can find oneself embed in social contradictions, and should display grace under pressure.

The contradiction was that a mere technical specialist cannot avoid being a manager to the extent he makes expensive recommendation, but under United States law in the form of Taft Hartley, management retains control. Which means that in day to day praxis, you are forever crossing and recrossing, transgressively, a boundary.

Thus Andy protested in vain to Belleville that he was merely assisting his coworkers when insisting that a technical choice wouldn't work, displaying in fact teamwork, not "insubordination" (a term which I'd thought Weinberg had disposed of, once and for all, in The Psychology of Computer Programming).

The programmer being subject to a bad performance review CAN say something like “ok, boss, I can deal with the company being run as a company along Leninist lines, of democratic centralism where we implement when the time for discussion is over. But I raised objections to a fixed file 512 byte limit only in design meetings where this issue was on the table, and I resent like hell that my discussion is now called ‘foolish’, in a company so supposedly dedicated to listening to its employees”. The reference to Lenin will keep ‘em guessing.

“You ast me at the time for my opinion, and I gave it.”

Keep in mind that during the Hertzfeld situation under review in a meeting by Jobs, Belleville denied, according to Andy, that he even delivered the review in the first place. This Bush-league out and out lie is something you will encounter. As a lie, it deconstructs coherence itself and there is nothing you can do about it except to reflect that Kant said we’re honest and straight not in necessary hope for temporal reward, nor “the shadowy delights of posthumous fame”, nor for an eternal salvation, but because purity itself is a Good Thing.

Belleville, like so many MIS middle managers then and now, was the Invisible Man. Because he lied, he is deaf and dumb. But his contribution AS A MANAGER was key to the creation of the Mac, as key as Andy’s. I say this because I saw the icons of the Mac three years before its introduction in 1984, on a prototype Lisa loaner at Bell Northern Research, and long before that had read, in 1974, the Rolling Stone article about Xerox Parc. These contributions never saw commercial success because without middle managers, they never make it to the market.

In your performance review ask why you are not a “team player” when everybody comes to your cubicle for help and advice. Press the question: the company is, or is not, dedicated to the Enlightenment ideal of human betterment through technology of which the fortunes of the top men are only an expression to encourage the rest of us, and differences in income are just, but only insofar as they better the lot of the least well off (Macs in schools: Microsoft's very generous Bill and Melinda Gates foundation: the Woz's teaching).

And if you as I have, as Andy has, encounter the barbarian on the dark highway who lies, cheats, and steals, why then keep the faith, if for nothing else than to set an example for your kids of that sort of damaged, shop-soiled integrity to which they can relate.

Let us now praise famous men, and their children after them. The biggest mistake at Apple was that studied childishness which I preserved until my own forties, that fear of the dark man on the road who was myself, and as a result I denied my kids of half an upbringing. I did sit with my son in the light on my lap waiting for Happy Mac after he eagerly helped me install my very first Mac and nothing can take this away from me, or him.

At the same time, I accepted for far too long the infantilizing rules of the game. I acted like a child, like a student, to get a job at Princeton at half my former salary because my younger son got on the phone and said “hey, when you getting’ a job so that our Mom can get the munny she needs for us?” I arrived at Princeton with my Mac and the clothes on my back and only gradually realized that my destiny was what my old friend from my antiwar days, Linda, had said: not as Picasso, or Jobs, but as a mere teacher.

But there’s no point in being Catholic unless you can blarney your way out of sticky situations that occurred “through my fault, through my own fault, through my own grievous fault”. God’s a good listener, I find, and in the very blarney (commencing with writing things down in preference to arguing with my wife, always a desparate and doomed adventure) I have found some coherence and a way to take responsibility…while acknowledging that life’s a uniquely dirty business.

Are we not men? No, you don’t have to cry in a performance review: but no, I think no less of Andy for doing so. The fact is that the basic barbarism can only be faced squarely neither by becoming a sadist, or a masochist, but instead by finding your own values as did Andy and the Woz himself. These values are like beautiful code: good in themselves, and not as a means to another end.

Andy of course survived his performance review, and his book is very readable. As to Bob Belleville, I wonder, where are you tonight, Sweet Marie, with what I hope is compassion for those lost legions of middle managers who are unsung buffers between personalities like Steve Jobs and Larry Ellison, and us polecats.

I hope this long blog has been at all enlightening. It has been neither "oral history" like my 1999 article about Fortran in the IEEE journal for software history, nor disinterested academic history. The narrator in an oral history is patronized, and I have no wish to be patronized because after my own BNR performance review I reflected on the incident with what I hope was not only ascending weirdness but also ascending objectivity.

San Francisco, CA 2-4-2004 Edward G. Nilges, amdg


Source URL:
http://www.developerdotstar.com/community/community/node/102