There is No Manual

This is how I start at Microsoft in 1996: of 10 people who interview me, the man who becomes my boss is the only one who doesn’t want to hire me. And he is the sweetest guy you could ever hope to work for. He’s a good and conscientious mentor, but I learn a hard lesson early on. In a dynamic company in fluid circumstances and a wildly oscillating market, there is no Manual. Within 10 years, the whole world will be like this, but I am still vaguely resentful. I’m willing to do all the work, to take home… whatever it is I could take home, if there were anything to take home, but there isn’t anything to take home. You just somehow have to absorb the job and ask a lot of embarrassing questions when you don’t know how to do something you should probably know how to do.

A Poorly Debugged Statement to the Press

Microsoft is a cornered animal in 1996. Wall Street’s opinion is pretty much unanimous. Microsoft is toast, thanks largely to an upstart company called Netscape, which is giving away its main product (the first popular web browser) by the millions. Wall Street doesn’t seem to notice that, ah, Netscape has not yet conjured up a plausible business model. Netscape’s tech founder is that mainstay of the business press, a charismatic twentysomething who says things like “Windows is nothing more than a poorly debugged set of device drivers.” These things sound good, and it’s time to take Bill Gates down a notch, right?

Right. Especially if you’ve never written a device driver. I have. Bill probably has. The Netscape kid: doubt it. Microsoft’s are pretty damn good. Technically, by the way, they aren’t even Microsoft’s responsibility. A device driver is the software that allows your computer to control hardware attached to it: for example, printing what you see on the screen, or taking advantage of a new video card. It should be written by the printer company, or the video card manufacturer, or whatever. They’re hard to write, and harder to write well. Microsoft uncomplainingly does the heavy lifting, and to a degree that as far as I can tell has seldom been described in the decade since I left. Time and again I find myself facing situations where my team has to write a workaround because big software companies like Adobe or Corel made improper use of system calls, or just did something plain wrong by mistake in a previous version of Windows, and now our team has to write a hack just to keep their old badly written programs tottering along because if they don’t, customers will blame us. Whenever we allocate time to this kind of remedial work we are consciously deciding to omit a cool feature that would be of great interest to our customers.

Device drivers? Way harder than a web browser, at least in 1996. By the way, debugging a device driver is a bitch. Though my wife and I understand how wrong the conventional opinion is, I’m a little scared. What if I’ve hitched my wagon to the wrong star? Middle age with one baby and another, endangered one on the way is not the time to be seriously wrong about your career. People I respect greatly are wrong, people a lot smarter than I am. I stick to my guns, but… man. Thank heavens my wife is rock-solid in her convictions and doesn’t give a damn what anyone else thinks. She paid attention when I made unconventional decisions before and trusts me more than I do.

Go Netscape

I’m a decade and a half older than the Netscape kid, read the industry histories, and followed the business press carefully, so I’ve seen what happened when Microsoft swaggered into the arena against its betters: Lotus, Ashton-Tate, WordPerfect. (Wait–you don’t remember Lotus, Ashton-Tate, or WordPerfect? Right.) Microsoft’s stock is at some kind of nadir when I join the company, and the first thing I do is announce my coworkers that I think its stock could easily double this year.

My coworkers listen politely, but they are busy buying Netscape. They’re twentysomethings too, three or four years out of college, and they think Microsoft is washed up. Microsoft is the only place this smart kids have ever worked, because Bill was smart enough to send emissaries to places like Harvard and Waterloo and MIT before the other microcomputer companies thought to bother.

I am, in fact, wrong. Microsoft’s stock triples that year.

By the time I leave, four years later, my stock options are worth 13 times what they were when I signed on. Netscape is history. (The Netscape kid leaves with a few hundred million and invests it wisely later.) I leave before all my options are vested because my wife is about a bear a child with serious health issues. I am worth seven figures for a split second. Uncle Sam takes a bite, and 40% of it is gone before I even get to see it in a bank statement somewhere. Still, that’s a decent retirement fund. I am half a millionaire and the last thing I graduated from was junior high school. This is no accident. All I did was understand history, apply a little common sense, and try not to wet my pants when I realize I am betting against some of the smartest people in the world.

Within six months of retirement the first dot com boom comes to a sickening end. It is at that point I buy my way in, and put the nest egg at serious risk.