I practice human-centered design. But what does that mean? It actually can mean a variety of things. We teach a method. We say, "First of all, let's go out and watch
people." Observe. Do field work. See what they really do in their lives. Whether it's education, or entertainment,
or their jobs. Find out what happens because... don't ask
them. People often don't realize what they need. Those people are used to what they're doing,
and if you ask them what their problems are... Well, they'll tell you something about what they're trying
to accomplish today, and what's in the way. Whereas if we watched and observed, we might
think of entirely different solutions to their problems. So we start off by observing. Then we actually try to figure out what is
the real problem they're facing, and what are the possible solutions. We think of a variety of solutions. And then, we try to see which one might be
most appropriate and we test. We do quick, rapid prototypes. We believe in our prototype... I can build something in a half hour that
I can test with you. You know, a lot of engineers say, "Oh, gee,
it's going to take me six months to build this. How could we test until we have it built?" No, you fake it! We use methods from theatre to make it make-believe. We take something out of our pocket, we take
a piece of paper and we say, "Make believe this is your new display. How would you use it?" We make things out of chalk. We make things out of cardboard. We make things out of foam core. We can cut things and make them into the right
shapes. We can see how people would use them and experiment
with them. You can even mock up an automobile with just
a bunch of chairs in the right ordering. And then, as we learn what people need and
what we're going to build, we make better and better and better models. And pretty soon a pretty final product. But we're always testing. Always iterating, changing what we're doing
because we're learning. Even though I've had pretty near 50 years of experience
watching, observing, and understanding people, a PhD in psychology, I'm always wrong. I'm always surprised. Because people are so variable. They do things that are so unexpected. You've always got to go out and watch and
test. Now, that's the essence of human-centered
design, but you know something? It usually doesn't work in a company. The reason it doesn't work is because... well,
Norman's Law. The day the product team is put together,
it's over its budget and behind schedule. So, that means that all these wonderful, early
explorations and tests and iterations. Oh, the person in charge of the product is
apt to say, "Yeah, what you want to do is the correct way of doing it, but we don't
have time. So we'll have to skip it. Or maybe short-circuit it. Instead of spending a week, we'll give you
a few hours. But next time, we do it better." There is never a next time, because the same
story happens next time. So... we have to change our ways. We have to adapt what we're doing. This means we should always be studying people,
even when there's no product in mind; on the side; so that when the product team is assembled,
we're ready, we can say, "Oh yeah, we've been studying that." But it also means we sometimes short-circuit. We say, "Okay, let's quickly do something. Do a few cycles in a day so we can learn something." But it's really important that we keep up with
the product cycle, because the normal methods... well, they don't really fit what we're trying
to do. And the new Agile programming methods are
very good for some things, but they're not so good for us; because the programmers want
to start programming on Day 1, even before we know what we want to build. So, we can't say, "Oh no, you must do it our
way." We have to figure out how to work with them
as partners. That's human-centered design.
