This is the second of two excerpts in Salon 21st from Ellen Ullman's new
book, "Close to the Machine: Technophilia and its Discontents" (City
Lights Books, $21.95, 189 pages), an autobiographical exploration of the
lives and minds of software engineers.
It had to happen to me sometime: sooner or later I would have to lose sight of the cutting edge. That moment every technical person fears -- the fall into knowledge exhaustion, obsolescence, techno-fuddy-duddyism -- there was no reason to think I could escape it forever. Still, I didn't expect it so soon. And not there: not at the AIDS project I'd been developing, where I fancied myself the very deliverer of high technology to the masses.
It happened in the way of all true-life humiliations: when you think you're better than the people around you. I had decided to leave the project; I agreed to help find another consultant, train another team. There I was, finding my own replacement. I called a woman I thought was capable, experienced -- and my junior. I thought I was doing her a favor; I thought she should be grateful.
She arrived with an entourage of eight, a group she had described on the telephone as "Internet heavy-hitters from Palo Alto." They were all in their early 30s. The men had excellent briefcases, wore beautiful suits, and each breast pocket bulged ever so slightly with what was later revealed to be a tiny, exquisite cellular phone. One young man was so blonde, so pale-eyed, so perfectly white, he seemed to have stepped out of a propaganda film for National Socialism. Next to him was a woman with blonde frosted hair, chunky real-gold bracelets, red nails, and a short skirt, whom I took for a marketing type; she turned out to be in charge of "physical network configuration." This group strutted in with all the fresh-faced drive of techno-capitalism, took their seats beneath the AIDS prevention posters ("Warriors wear shields with men and women!" "I take this condom everywhere I bring my penis!"), and began their sales presentation.
They were pushing an intranet. This is a system using all the tools of the Internet -- Web browser, net server -- but on a private network. It is all the rage, it is cool, it is what everyone is talking about. It is the future and, as the woman leading the group made clear, what I have been doing is the past. "An old-style enterprise system" is what she called the application as I had built it, "a classic."
My client was immediately awed by their wealth, stunned silent by their self-assurance. The last interviewee had been a nervous man in an ill-fitting suit, shirt washed but not quite ironed, collar crumpled over shiny polyester tie. Now here came these smooth new visitors, with their "physical network configuration" specialist, their security expert, their application designer, and their "technology paradigm." And they came with an attitude -- the AIDS project would be lucky to have them.
It was not only their youth and high-IQ arrogance that bothered me. It wasn't just their unbelievable condescension ("For your edification, ma'am," said one slouch-suited young man by way of beginning an answer to one of my questions). No, this was common enough. I'd seen it all before, everywhere, and I'd see it again in the next software engineer I'd meet. What bothered me was just that: the ordinariness of it. From the hostile scowl of my own programmer to the hard-driving egos of these "Internet heavy-hitters": normal as pie. There they were on the cutting edge of our profession, and their arrogance was as natural as breathing. And in those slow moments while their vision of the future application was sketched across the white boards -- intranet, Internet, cool, hip, and happening -- I knew I had utterly and completely lost that arrogance in myself.
I missed it. Suddenly and inexplicably, I wanted my arrogance back. I wanted to go back to the time when I thought that, if I tinkered a bit, I could make anything work. That I could learn anything, in no time, and be good at it. The arrogance is a job requirement. It is the confidence-builder that lets you keep walking toward the thin cutting edge. It's what lets you forget that your knowledge will be old in a year, you've never seen this new technology before, you have only a dim understanding of what you're doing, but -- hey, this is fun -- and who cares since you'll figure it all out somehow.
But the voice that came out of me was not having fun.
"These intranet tools aren't proven," I found myself saying. "They're all release 1.0 -- if that. Most are in beta test. And how long have you been doing this? What -- under a year? Exactly how many intranets have you implemented successfully?"
My objections were real. The whole idea wasn't a year old. The tools weren't proven. New versions of everything were being released almost as we spoke. And these heavy-hitters had maybe done one complete intranet job before this -- maybe. But in the past none of this would have bothered me. I would have seen it as part of the usual engineering trade-offs, get something, give up something else. And the lure of the new would have been irresistible: the next cover to take off, the next black box to open.
But now, no. I didn't want to take off any covers. I didn't want to confront any more unknowns. I simply felt exhausted. I didn't want to learn the intranet, I wanted it to be a bad idea, and I wanted it all just to go away.
"And what about network traffic?" I asked. "Won't this generate a lot of network traffic? Aren't you optimizing for the wrong resource? I mean, memory and disk on the desktop are cheap, but the network bandwidth is still scarce and expensive."
More good objections, more justifications for exhaustion.
"And intranets are good when the content changes frequently -- catalogs, news, that kind of stuff. This is a stable application. The dataset won't change but once a year."
Oh, Ellen, I was thinking, What a great fake you are. I was thinking this because, even as I was raising such excellent issues, I knew it was all beside the point. What I was really thinking was: I have never written an intranet program in my life, I have never hacked on one, I have never even seen one. What I was really feeling was panic.
I'd seen other old programmers act like this, get obstructionist and hostile in the face of their new-found obsolescence, and there I was, practically growing an old guy's gut on the spot. But the role had a certain momentum, and once I'd stepped on the path of the old programmer, there seemed to be no way back. "And what happens after you leave?" I asked. "There just aren't that many intranet experts out there. And they're expensive. Do you really think this technology is appropriate for this client?"
"Well," answered the woman I'd invited, the one I'd thought of as my junior, the one I was doing a favor, "you know, there are the usual engineering trade-offs."
Engineering trade-offs. Right answer. Just what I would have said once.
"And besides," said the woman surrounded by her Internet heavy-hitters, "like it or not, this is what will be happening in the future."
The future. Right again. The new: irresistible, like it or not.
But I didn't like it. I was parting ways with it. And exactly at that moment, I had a glimpse of the great, elusive cutting edge of technology. I was surprised to see that it looked like a giant cosmic Frisbee. It was yellow, rotating at a great rate, and was slicing off into the universe, away from me.
I learned to program a computer in 1971; my first programming job came in 1978. Since then, I have taught myself six higher-level programming languages, three assemblers, two data-retrieval languages, eight job-processing languages, seventeen scripting languages, ten types of macros, two object-definition languages, sixty-eight programming-library interfaces, five varieties of networks, and eight operating environments -- fifteen, if you cross-multiply the distinct combinations of operating systems and networks. I don't think this makes me particularly unusual. Given the rate of change in computing, anyone who's been around for a while could probably make a list like this.
This process of remembering technologies is a little like trying to remember all your lovers: you have to root around in the past and wonder, Let's see. Have I missed anybody? In some ways, my personal life has made me uniquely suited to the technical life. I'm a dedicated serial monogamist -- long periods of intense engagement punctuated by times of great restlessness and searching. As hard as this may be on the emotions, it is a good profile for technology.
I've managed to stay in a perpetual state of learning only by maintaining what I think of as a posture of ignorant humility. This humility is as mandatory as arrogance. Knowing an IBM mainframe -- knowing it as you would a person, with all its good qualities and deficiencies, knowledge gained in years of slow anxious probing -- is no use at all when you sit down for the first time in front of a UNIX machine. It is sobering to be a senior programmer and not know how to log on.
There is only one way to deal with this humiliation: bow your head, let go of the idea that you know anything, and ask politely of this new machine, "How do you wish to be operated?" If you accept your ignorance, if you really admit to yourself that everything you know is now useless, the new machine will be good to you and tell you: here is how to operate me.
Once it tells you, your single days are over. You are involved again. Now you can be arrogant again. Now you must be arrogant: you must believe you can come to know this new place as well as the old -- no, better. You must now dedicate yourself to that deep slow probing, that patience and frustration, the anxious intimacy of a new technical relationship. You must give yourself over wholly to this: you must believe this is your last lover.
I have known programmers who managed to stay with one or two operating systems their entire careers -- solid married folks, if you will. But, sorry to say, our world has very little use for them. Learn it, do it, learn another: that's the best way. UNIX programmers used to scoff at COBOL drones, stuck year by year in the wasteland of corporate mainframes. Then, just last year, UNIX became old-fashioned, Windows NT is now the new environment, and it's time to move on again. Don't get comfortable, don't get too attached, don't get married. Fidelity in technology is not even desirable. Loyalty to one system is career-death. Is it any wonder that programmers make such good social libertarians?
Every Monday morning, three trade weeklies come sliding through my mail slot. I've come to dread Mondays, not for the return to work but for these fat loads of newness piled on the floor waiting for me. I cannot possibly read all those pages. But then again, I absolutely must know what's in them. Somewhere in that pile is what I must know and what I must forget. Somewhere, if I can only see it, is the outline of the future.
Once a year, I renew my subscription to the Microsoft Professional Developer Network. And so an inundation of CD-ROMs continues. Quarterly, seasonally, monthly, whenever -- with an odd and relentless periodicity -- UPS shows up at my door with a new stack of disks. New versions of operating systems, libraries, tools -- everything you need to know to keep pace with Microsoft. The disks are barely loaded before I turn around and UPS is back again: a new stack of disks, another load of newness.
Every month come the hardware and software catalogs: the Black Box networking book, five hundred pages of black-housed components turned around to show the back panel; PCs Compleat, with its luscious just-out laptops; and my favorite, the Programmer's Paradise, on the cover a cartoon guy in wild bathing trunks sitting under a palm tree. He is all alone on a tiny desert island but he is happy: he is surrounded by boxes of the latest programming tools.
Then there is the Microsoft Systems Journal, a monthly that evangelizes the Microsoft way while handing out free code samples. The Economist, to remind myself how my libertarian colleagues see the world. Upside, Wired, The Red Herring: the People magazines of technology. The daily Times and Wall Street Journal. And then, as if all this periodical literature were not enough, as if I weren't already drowning in information -- here comes the Web. Suddenly, monthly updates are unthinkable, weekly stories laughable, daily postings almost passé. "If you aren't updating three times a day, you're not realizing the potential of the medium," said one pundit, complaining about an on-line journal that was refreshing its content -- shocking! -- only once a day.
There was a time when all this newness was exhilarating. I would pore over the trade weeklies, tearing out pages, saving the clips in great messy piles. I ate my meals reading catalogs. I pestered nice young men taking orders on the other end of 800 phone lines; I learned their names and they mine. A manual for a new programming tool would call out to me like a fussy, rustling baby from inside its wrapping.
What has happened to me that I just feel tired? The weeklies come, and I barely flip the pages before throwing them on the recycle pile. The new catalogs come and I just put them on the shelf. The invoice for the Professional Developer Subscription just came from Microsoft: I'm thinking of doing the unthinkable and not renewing.
I'm watching the great, spinning, cutting edge slice away from me -- and I'm just watching. I'm almost fascinated by my own self-destructiveness. I know the longer I do nothing, the harder it will be to get back. Technologic time is accelerated, like the lives of very large dogs: six months of inattention might as well be years. Yet I'm doing nothing anyway. For the first time in nineteen years, the new has no hold on me. This terrifies me. It also makes me feel buoyant and light.