Pang: Let's turn to the Apple mouse.
Pang: I've talked to Jim Yurchenco about the design of the ribcage, Jim Sachs about the electrical work, and I want to understand what you did on it.
Kelley: I actually did very little on it. The Apple story starts pretty clearly with McKim-- when I tell the story, I'm amazed at how big an influence McKim is.
Anyway, when we formed the company, McKim said, "There's this guy who graduated a few years before you guys, who's working with this company Apple, and you should go meet him. His name is Jerry Manock." So I wandered down one day to his office.
Manock was on University, on the other side of High Street, in a big building with an atrium in the center-- I don't know if that building still exists-- and he was just by himself, just sitting in there drawing. I went in and introduced myself, and since we're this small little program, anybody else from Product Design you're endeared to immediately. Manock was working away: he was a consultant, he wasn't an employee at Apple, and he'd done the Apple II case. We introduced ourselves, let him know we were existed and were getting started, and he right away said he had more work than he could do with Apple. So we got started on that.
The first thing we did was-- I'm trying to remember, we've done about 50 projects for Apple now, and the mouse was not the first things we did-- the first thing may have been the Apple III. I did all the industrial design personally, because we didn't have any industrial designers. Manock wasn't an industrial designer, and Dresselhaus wasn't; they were both better than me, but they were really engineers. So I think-- and someone with a better memory can tell you this-- we were working on Lisa-- Apple IV-- first, and I remember saying we wanted to work on our own project, because Apple IV was really mixed up. Douglas Dayton was working on Lisa with Dresselhaus, and he worked on the keyboard and the first Lisa mouse.
[Pulls out box of mice prototypes, molds, and first Lisa mouse; we sort through it for a moment.]
This is the one where we really got going [gets second prototype], where we learned that the ball had to float. In this one [holds up first prototype from May 1980], when you pushed down hard on it, it would force the ball against the table, and it would skip, because the ball didn't float. This was the second prototype, and this the one we really got going on. The ball now floats on the table, we have now the two commutators, and the thing works great.
Pang: Does this also have the spring-loaded roller?
Kelley: No. By the time you get the ribcage you get that. Anyway, my Apple projects basically were the Apple III, a joystick, a bunch of other stuff, so I wasn't really on the team: I helped the brainstorm, but I wasn't very deeply involved.
I was working on what the mouse should look like, about the time this [the second prototype] was done. I went around and asked, "What feels really good in the hand?" And I found one of these sanding blocks. You ever see these sanding blocks, where it's a big hunk of rubber and you put the sandpaper in it? I took one of those and chopped it up, and then put little divots in it like a golf ball, and painted two little eyes like a mouse, and that was the first mouse case. Apple rejected it completely.
So it wasn't really my project. In those days there were only five or six of us in the office, so I can remember trying to figure out how to get the ball to not make noise on the table, and keep the eraser junk from getting in it, but I wouldn't put myself down as one of the designers of the actual mouse. But it was great fun, because you got to make all these discoveries. I don't know if anyone told you, but it took like no time whatsoever: we had nine months, and we were hardly paid anything, but we were happy as clams.
Pang: What were some of the other shapes that you experimented with?
Kelley: Well, after the one that I did that had way too much personality, shall we say, we started making wooden blocks. Douglas did a lot of that, and I did some of them [gets box of blocks] and we started doing all kinds of shapes. The path I went down for a long time was, there were trackballs-- in fact, our mechanism was similar to a trackball-- and I thought you should be able to use it as a mouse or a trackball. We went down that path for a long time, where you could use it as a trackball or a mouse. It didn't fly, and didn't need to: people caught on to the mouse much faster than we expected them to.
Pang: You've got all these different models. What was it that you thought would define a really good mouse shape?
Kelley: To me, the really important thing was, what's the scenario of use? People kind of react to new things according to what they're used to holding. That's why I went to the sanding block. So how are you going to use it? Are people going to hold it like a bar of soap? Are you going to drag your hand across the table, where your wrist is up, and it's something you'd do more precisely, or would you rest your hand on the table? Watching people play with them, it looked like users were going to rest their hand on the table, so it was going to be like the sanding block; it was not something that needed to be precise, because you were holding its weight, or because you were going to get your larger muscle groups involved. When you watched people, they were more relaxed when they held it.
So now it was a question of, How big should it be? We could make the guts of the mouse small, so the mechanical design wasn't much of a limiting issue. You'd normally think that smaller was better, but it needed to be to human scale. I learned that at Boeing: when I worked at Boeing, you'd calculate what size screw you needed to hold two pieces of metal together, and the answer would come out 0.080 screw. That was structurally enough, but you needed something big enough for people to handle manually, so you'd use a number 6, because that was the first size humans could see and hold. It was the same thing here: what's the size, what's the shape that works on a human scale.
There are all kinds of issues. Was precision important? A big thing came when we realized that the human brain's in the loop: I'm looking at the screen, and I'm moving the mouse, and it doesn't have to be absolutely precise, because my brain will see that it's off and will correct for minor errors automatically. Was right-handed and left-handed a big deal? We were obsessed with putting the button under the index finger, but once you do that you make the mouse handed. Today, I'd bring in hand surgeons, to make sure that no muscle groups were used unnecessarily, and have tests with typical users; back then, we were just blasting it out, it wasn't that way at all. You know, you'd use your intuition, and show it to whoever you could find-- "What do you think about this one? Okay, now what about this one?"-- but not in any systematic way. We were just trying to get done so Steve wouldn't beat us up. [Pang laughs] That's still true today.
Pang: Some of these the hand slopes downward, whereas on the final mouse the hand curves upward.
Kelley: I think these were basically early models, and we were a little naive. The sanding block model makes sense to slope down because your hand's supported up higher; but we saw that most people would rest the palm of their hand on the table.
Pang: There was a story that Jim Sachs told about you cutting off the knob of the stick shift of your BMW during your search for various kinds of shapes that worked well in the hand.
Kelley: We certainly took stuff off my BMW. Cut it off? I don't remember that. I remember the same day we went out and bought the sanding block we bought a butter dish, which was the first cover of the mouse, I can certainly remember having gear shift knobs on the table, that was really early.
And again, this is not Macintosh, which was a lot later, this was Lisa. But let's put it this way: we were not above going to Macy's and buying things and later returning them-- we didn't have a lot of money, right? But going to the hardware store, we just bought all kinds of stuff there. The gear shift knob also could have come from Dean's BMW. We bought twin BMWs, for grins.
But the butter dish was the most memorable, because we went to Walgreens, and we needed something plastic to put over the guts of the mouse, and that was it.
[Holds up P3 mouse] This one is one we did before we found out that eraser stuff was going to come up off the desk and jam up the mouse.
Pang: It sounds like Jobs was involved in the early stage, working with Dean, and then you'd bring him something at the end of the process--
Kelley: That's kind of how I remember it, but having done 50 projects with him, I'm sure he was more involved along the way. On the other hand, he was really busy. In looking back, once we got the initial design, and he saw the thing move around on the screen with our electronics, he saw that we were going to be okay, and he focused on other things. When we got into this fight about whether we were going to manufacture the mouse was when he was next heavily involved.
Another really interesting story is about the two-button or one-button mouse. Apple couldn't decide which one to do, and decided on the one-button mouse for ease of use: you didn't have to explain so much if you just had one button.
Pang: I know there were people within Apple who were arguing about that, and it sounds like Hovey-Kelley is working fairly separately--
Kelley: Yeah, it was a client relationship--
Pang: But were you involved in Apple's debate over how many buttons there should be? It sounds like Hovey-Kelley comes to the same conclusion in favor of one button independently.
Kelley: Involved first-hand? No. My remembrance of the thing is that our guys came back and said, "There's a fight about the buttons, and they decided in favor of ease of use." The woman who was writing the user's manual was heavily involved, and they thought that it would be easier to explain how to use it if it had one button.
Pang: Would that have been Joanna Hoffman?
Kelley: Joanna Hoffman would be my guess, but I wasn't there.
Pang: I also found a mention in some correspondence of an interest in manufacturing the mice as well.
Kelley: Oh yeah. It was more than interest. We really had the expectation that-- See, Apple was really focusing on all the stuff they had to focus on, and the mouse, they really gave it to us. We owned it. There was little involvement from them, other than getting the protocol right so Sachs could talk to the computer.
So we went to them-- Dean, being the business guy, went to them-- and said, "We should be more involved, we should be making these. Then we could make better ones, and different ones," and we could be a mouse company, basically. That happened later, with a guy named Kirsch--
Pang: Steve Kirsch?
Kelley: Yeah, he had a mouse company.
Pang: Was he connected with Hovey-Kelley?
Kelley: No, he wasn't connected to us in any way; he just proves that we could have had a mouse company. And so we were counting on it: we were starting to think about it, and then some guy from Apple came in and said, "Nope," and just shut it down. We should have been more adamant about it and pushed harder, but we were just kids, so we didn't know what to do.
So it was our expectation-- not at the very beginning, but somewhere in the middle of the project-- that Jobs and Dean had had a verbal agreement that we were going to manufacture the mice. Remember, Dean is moving towards that anyway: he's thinking, "This isn't a very good business, this consulting business. We need to find a way to make a product," which he eventually did.
Pang: There was a second company that Apple contracted with, who was working parallel with you guys--
Kelley: I don't think I ever knew that--
Pang: I've found a couple references in memos. Apple felt it was such an important project, they wanted a second source.
Kelley: That makes perfect sense. I can remember them doing that on other things, so I'm not surprised, but I'd never heard that.
Pang: My next question was whether you knew anything about that other company, but obviously not.
Kelley: It doesn't surprise me, and wouldn't have been a problem; it probably would have made us more aggressive about doing a good job, which was part of why they would do such a thing. But I don't remember it. Did you ask Dean? He'd be more likely to know.
Pang: I haven't yet, though I got the memo from him. Bill Lapson remembered very vaguely another company, and that the design wasn't worth much.