0:00

[SOUND] Anybody who's ever addressed an open-ended problem,

it could be a really complex one like designing a building or

a piece of software or a device or a really simple one like deciding what

restaurant that you going to go to for dinner, has run into the same

problem whose never seem to quite get it right on the first try.

It always takes a couple of tries until you figure out what it is

that you're going to decide upon at the end of the day.

So when we say that design involves iteration, what we really mean, what

we're referring to is the idea that we're doing it over and over and over again.

But it's not iteration if you do it over and over again without a change.

That's repetition.

Iteration means that you do a new version.

It gets a little bit better or a lot better with each iteration that we try.

And this helps us to refine our design to figure out

what we're really working on and sometimes for new ideas to emerge.

So an example of iteration is how the FedEx logo came about.

You may be familiar with this logo because it has a little arrow that's hidden

between the E and the X, and the negative space between the letters.

This is a very famous solution that won numerous design awards.

Designed in 1994 by Lindon Leader at Landor Associates.

And the way that he tells the story about coming up with

this design is that they had decided upon FedEx as the logo partially because

people kept using the term FedEx even as a verb to describe mailing things.

And so, it's natural that FedEx itself would become the major part of the logo.

And the question was in the typographic treatment.

So, Lindon was interested in two typefaces at the time,

Universe 67 and Futura bold, and he was trying out variations of them.

And, as he put those letters together, all of the sudden, he saw the beginnings of

an arrow start to emerge, and all of the sudden, that's when he knew he had it.

So, he says that they went through over 200 different designs in order to

get to that FedEx logo.

But it was by that one moment of finding the arrow within the design that he

knew that he had it.

And this system works really well.

It's a very simple design, and by changing the color and

the variations of the words below the X is able to apply it across an entire system.

The point is that iteration here, not only allows the solution at the end, but it

also was the process by which the actual design, the final design was discovered.

2:36

Another way to think about iteration is to think about classes of things.

So a whole set of design solutions around a similar idea.

So think about a vacuum cleaner.

There are lots of different kinds of vacuum cleaners.

There's upright ones, there's Shop-Vac's, the big tubs,

there's the Roomba Robots that move around your house.

There's whatever that one is called, that's the big box you drag around on

the floor, now it gets caught on every doorway.

And there's the Dyson rethink of how the vacuum cleaner moves around your house by

putting a big ball in the middle.

So these are all variations on the same common problem of how to clean your house.

And there are iterations within a set.

3:15

You think about, each iteration gets you

slightly closer to an idea that you think has a right set of characteristics and

trade offs that, are what you're seeking for an optimal solution.

So the solution that you can finally settle upon.

You think about that as being dots on a page.

So when you put one dot down, another one, and

you get a little closer, and a little bit closer, and then finally,

they start to cluster around probably what the end result is going to be.

Big question is, how and why does iteration work in design?

Well, in 1970, John Chris Jones who wrote a book called Design Methods,

proposed an idea that the design process could be divided into three phases.

Divergence which is lateral thinking, creative exploration of ideas.

The ah-ha moment, he calls the transformation phase.

And then convergence is when we take that ah-ha moment, that insight, and

we start to make it real.

That's in the design process how we get from problem to solution.

In that early phase, divergence, the lateral thinking, free wheeling thinking,

lots of different ideas come together.

This is where iterations particularly important.

So, an example of that is Phillipe Starck's juicer which

he designed in 1990 for Alessi.

Now, there's a lot of debate about whether or

not this juicer actually works as well as it is intended, but

most people realize that somehow that thing gets at the essence of juicerness.

It's got three legs to keep it stable.

It's sort of shape like a juicer and

it should drip down right into your glass that the whole device straddles.

Now, it's not necessarily true that it works like that.

But that project started as a sketch on

a pizzeria placemat that Starck did one evening.

And we're really lucky that Starck managed to save that placemat.

It looks a little bit like this.

If you look, you could see the pizzeria placemat.

And within it, you can see all the variations then he started to explore.

You can see how the ridging might guide the juice down.

And an idea about the ridging falling around in the hand motion that we

typically do when we juice something or

whether it should drip straight down into the glass.

And you can see Starck exploring lots of different ideas.

5:55

That kind of early looseness really worries the people who

are in charge of the design process, who aren't the designers.

Typically the managers or

the clients, they're not sure that design should work like that.

They tend to describe design more as a waterfall model.

So you can see here an early model from 1970s about how design might

proceed through stages in a fairly rational, linear manner.

But, everybody knows who's ever designed anything that

this don't work necessarily in those stages.

And it's very difficult to go through a design process with one idea in mind that

come up with a product at the end.

So design theorists then revised that model into something called

the spiral model.

When you look at the spiral model,

you can see that the phases of the design process still exist at the poles.

But if you look really closely,

you'll see that there are series of prototypes that occur.

So we move from the outside of that spiral to the inside of

the spiral through a series of four prototypes.

Winding up, hopefully, focusing graphically on the final product.

Now, might be worthwhile to think about what would happen if you unspool

that spiral.

You wind up with a line that basically describes a design process for

four iterations, those four iterations or as many as you need in that process,

each of which is getting you a little closer to our final stopping point.

And so, if we take those two ideas, the spiral of getting us in closer and

the idea of those prototypes or those iterations as dots that

are converging on our final answer or our final design product.

We get, we put those two ideas together, and

we get to what John Zeisel described as the Design Development Spiral.

This is a little bit more of a complicated diagram.

But you can imagine the long white tube there in the center is following time,

and it holds, it's the parameters of acceptability.

So you may,

in the beginning of your design process, be circling around that but

still have an idea or a set of iterations that don't quite work into what you need.

And as time goes by, you spiral into the center and

as soon as you're within those parameters, then you're getting very close in

the convergence phase and we wind up at the end with our final design project.

8:15

So, where does this leave us?

Do we really have to do 200 variations on every

project like Linden Leader says he did on his?

But I, I'm not so sure.

I think the answer might be a little bit less than 200.

So, a few years ago, I saw a presentation by an architecture professor who asked

the students and the audience a really simple question, why is it

that we always ask for three variations whenever we are talking to a designer?

So what would happen if we asked for ten variations?

So, he did this with his students.

He asked them to come up with ten variations of a house design and

to keep track of the order in which they did them.

9:12

The next three answers tended to be variations on the first three.

So the first three ideas done slightly differently, iterations themselves.

And then, the next two tended to be solutions that

focused on one idea in particular.

So we may have had three radically different house designs,

three variations on those house designs.

And then, answers seven and eight tended to be a house design about one

particular thing, how the light comes in,

how the wind works the orientation of the building, or whatever it might be.

So, they went really deep into one idea.

And then, finally, 9, 10, 11, 12, however many more they were,

those tended to be answers that created a synthesis of the problems before.

So, we had three variations,

variations of those variations depth into particular ideas.

And then after number nine,

the idea started to bind them together into synthetic,

holistic, whole satisfied more aspects of the project than the first ones did.