Jef Raskin on Technical Writing

Source: Interview with Jef Raskin, 13 April 2000.

Technical Writing and Development

Pang: One thing that was unusual about your background was that you had been working in publications and technical writing. Looking back, were there things about the way computers worked, or the way people worked with computers, or issues designers had to think about that you had discovered from having worked in publications, rather than having a more strictly technical background.

Raskin: Oh, but your premise is false. When I was even an undergraduate, I designed and physically built computers for the Biology department at-- then Oyster Bay, now Stony Brook. And when I was a kid, I was the only kid in high school with a Tektronics oscilloscope in his shop, when none of the other kids even knew what an oscilloscope was. So I don't come from that background; I come from a technical, even a hardware background, which is quite unusual now in the interface field. So that came long before I discovered the real key to making computers accessible-- I have an egalitarian streak for some reason-- was the interface.

So I came hardware to software, and my work in graduate school was the Quickdraw graphics system, which was all matrix operations and technical innards. It was only when working with graduate students and faculty from the humanities and arts that I began to realize how badly our computers were designed, whereas most of the other people who were in CS or were graduate assistants helping people using the computers said, "Oh those people just don't have the right technical background." I suddenly realized it was the bad design of the computers that was hindering them. What they were trying to do was very simple; we just had these arcane systems to do it. So that was moving from helping people use software, to recognizing that the problem was the design, not the people. That was 1966.

Pre-Apple Work

When I joined Apple, it was somewhat under false pretenses. I had sort of dropped out of computer science, because mainframes were really dull, and had spent the time being a musician in San Francisco. I was a conductor at the San Francisco Chamber Opera, teaching at the community music center in the city, being a professional musician. Then all of a sudden the microprocessor came along, and the MITS Altair became available, and I said, "Aha! This is what I've been waiting for." And I immediately bought one of those, and soldered it all together along with Doug Wyatt of PARC-- and this we did as a hobby together, it wasn't involved with PARC in any way. And then we built another, the IMSAI I think it was. We wrote some software and sold it to a stockbroker who used it, instead of using the GE timesharing service, on a teletype because our little program in BASIC would do everything his service would do, and cost a fraction. One or two months' timesharing charges and he could buy the whole computer.

We also discovered at that moment that these microcomputers could be very profitable [Pang laughs], because a computer we could buy for a thousand dollars we could sell for ten or twenty thousand dollars, if we bundled it with the I/O and everything working. I then started this company Bannister and Crun, on my own money, without outside capital. And also at the same time I was starting to write reviews-- I remember working on an earlier modem design by Lee Felsenstein, who I later got to know, and still know-- and started writing reviews.

Becoming a Writer

It's very strange, but in grade school, high school, college, I was a dreadful writer-- I was a D student in English. One of the things I did at Bannister and Crun was use one of these little microcomputers, a Poly-88, and wrote a word processor; and it may have been the first microprocessor word processor. I discovered my real problem was not lack of ability, but two things. One, it was so mechanically difficult to write, and make corrections, and do all that work, that it didn't seem worth the bother; and two, I didn't have anything to say. I remember going to college, and going to this class, and the first assignment the professor gave was, "Please write a five-page paper on the subject of... cucumbers." I had no idea how to write even more than one sentence on cucumbers: "Cucumbers are round, green vegetables that grow out of the ground, and I like to eat them." [Pang laughs]

I hadn't the foggiest idea of what writing was all about. Fortunately, Brian Howard and Doug Wyatt were both very good editors, and unsparing, and truthful; and while Brian worked for me at Bannister and Crun, he had no qualms about thoroughly savaging anything I did, and Doug also had great sensitivity and an ear for English. So the two of them unwittingly taught me how to write. That, and having a word processor, and wanting to say something, turned me into a writer. And soon I was writing for Byte, and Silicon Gulch Gazette, and Dr. Dobb's Journal. I imagine you've seen my c.v., and the number of articles-- I don't even have a complete list, but the ones I have listed amount to some three or four hundred.

Pang: Was it difficult breaking into that work, getting assignments?

Raskin: No, no! Nobody was doing it. That's one of the great secrets, to find something that no one's doing and do it, and then guess what? you're the leader. It's much easier to be in a field where there's nobody else. I've always been very good I think at choosing small ponds where I can be the biggest frog without any effort. Like when I created the Macintosh, there weren't any other computers at a popular price that had interface as a goal, so it was bound to make a big splash. Of course, choosing the right pond helps. But it's easier to be on the leading edge than the middle, that's what I always find. It's not very well populated; it's Daniel Boone and elbow room, and all that American stuff. So yes, it was easy to get published.

And then I had some good editors at the various magazines, who also attacked my writing, especially Jim Warren. In fact, my editors still attack my writing! Good editors make for good writers, I think. Some people have it naturally, but I certainly didn't. So that's how I got into writing.

So I didn't have the unusual background of being a writer moving into designing the Macintosh; I'd been designing computers in the 1960s, from the ground up-- using vacuum tubes, even-- so writing came later.

Documenting the Apple II

Pang: Here's why I asked the question. In one piece you wrote, you talked about the experience of working on the Apple II, and the difficulty of predicting what sorts of effects would come from people opening up the machine and installing various pieces of hardware, and how hard it was to account for that in documentation. So perhaps the question I should have asked was whether there were additional concerns of yours, or things you became interested in, as a result of your working on that way the Apple II--

Raskin: Yes, that's also true--

Pang: -- in addition to your hardware work.

Raskin: In fact, I think the particular paper you're referring to was not by me, but by Brian Howard, where he points out that we had thousands of different Apples. That was a brilliant paper that I thought should have been published, but Brian was never one to put himself out in the world that way. He wrote a little thing to point out that we now have thousands of different computers, and you can't possibly document them all.

Chris Espinosa talks in his interview about his work rewriting the Apple II manual.

Writing and Designing

But one of the things I think is good about manuals, and working with manual writers, is that I can use writers to find out where a design is flawed. We were just doing a manual, and it had some seventy pages just to install the modem. Our schtick here is that we're easy to install, and seventy pages is not an easy install. So, using all the information we had gathered as writers trying to document the existing system, we then redesigned it to simplify the process immensely, cutting out half the manual in the process. Now, any one manual has to cover a whole broad range of machines, but for any one machine it's a little thin manual: instead of having some sixty or seventy steps, it has five or six, and that's manageable.

But whether I was writing, or whether other people were writing, I learned to listen very carefully to writers, because they're forced to look at every detail minutely, and if they say something is complicated, or they don't understand how to write about it, well, what could we have designed that makes it so difficult to write about? If it's not easy to write about, it's not easy to use. So I use writing as a guide for design. This puts writers in a very unusual position with respect to development-- which people in development are somewhat surprised to learn about when it happens [Pang laughs]-- but they're a great resource. So is customer service, if you can get feedback-- if you keep records about what people ask and where they're having problems, it's extremely valuable. All those things go back into design.

Caroline Rose and Chris Espinosa in their interviews talk about the role of technical documentation in the development of the Macintosh.


Document created on 6 June 2000;