My life in typefaces

Pick up a book, magazine or screen, and more than likely you'll come across some typography designed by Matthew Carter. In this charming talk, the man behind typefaces such as Verdana, Georgia and Bell Centennial (designed just for phone books -- remember them?), takes us on a spin through a career focused on the very last pixel of each letter of a font.

Type is something we consume

in enormous quantities.

In much of the world,

it's completely inescapable.

But few consumers are concerned to know

where a particular typeface came from

or when or who designed it,

if, indeed, there was any human agency involved

in its creation, if it didn't just sort of materialize

out of the software ether.

But I do have to be concerned with those things.

It's my job.

I'm one of the tiny handful of people

who gets badly bent out of shape

by the bad spacing of the T and the E

that you see there.

I've got to take that slide off.

I can't stand it. Nor can Chris.

There. Good.

So my talk is about the connection

between technology and design of type.

The technology has changed

a number of times since I started work:

photo, digital, desktop, screen, web.

I've had to survive those changes and try

to understand their implications for what I do

for design.

This slide is about the effect of tools on form.

The two letters, the two K's,

the one on your left, my right, is modern,

made on a computer.

All straight lines are dead straight.

The curves have that kind of
mathematical smoothness

that the Bézier formula imposes.

On the right, ancient Gothic,

cut in the resistant material of steel by hand.

None of the straight lines are actually straight.

The curves are kind of subtle.

It has that spark of life from the human hand

that the machine or the program

can never capture.

What a contrast.

Well, I tell a lie.

A lie at TED. I'm really sorry.

Both of these were made on a computer,

same software, same Bézier curves,

same font format.

The one on your left

was made by Zuzana Licko at Emigre,

and I did the other one.

The tool is the same, yet the letters are different.

The letters are different

because the designers are different.

That's all. Zuzana wanted hers to look like that.

I wanted mine to look like that. End of story.

Type is very adaptable.

Unlike a fine art, such as sculpture or architecture,

type hides its methods.

I think of myself as an industrial designer.

The thing I design is manufactured,

and it has a function:

to be read, to convey meaning.

But there is a bit more to it than that.

There's the sort of aesthetic element.

What makes these two letters different

from different interpretations by different designers?

What gives the work of some designers

sort of characteristic personal style,

as you might find in the work of a fashion designer,

an automobile designer, whatever?

There have been some cases, I admit,

where I as a designer

did feel the influence of technology.

This is from the mid-'60s,

the change from metal type to photo,

hot to cold.

This brought some benefits

but also one particular drawback:

a spacing system that only provided

18 discrete units for letters

to be accommodated on.

I was asked at this time to design

a series of condensed sans serif types

with as many different variants as possible

within this 18-unit box.

Quickly looking at the arithmetic,

I realized I could only actually make three

of related design. Here you see them.

In Helvetica Compressed, Extra Compressed,

and Ultra Compressed, this rigid 18-unit system

really boxed me in.

It kind of determined the proportions

of the design.

Here are the typefaces, at least the lower cases.

So do you look at these and say,

"Poor Matthew, he had to submit to a problem,

and by God it shows in the results."

I hope not.

If I were doing this same job today,

instead of having 18 spacing units,

I would have 1,000.

Clearly I could make more variants,

but would these three members
of the family be better?

It's hard to say without actually doing it,

but they would not be better in the proportion

of 1,000 to 18, I can tell you that.

My instinct tells you that any improvement

would be rather slight, because they were designed

as functions of the system they were designed to fit,

and as I said, type is very adaptable.

It does hide its methods.

All industrial designers work within constraints.

This is not fine art.

The question is, does a constraint

force a compromise?

By accepting a constraint,

are you working to a lower standard?

I don't believe so, and I've always been encouraged

by something that Charles Eames said.

He said he was conscious of working

within constraints,

but not of making compromises.

The distinction between a constraint

and a compromise is obviously very subtle,

but it's very central to my attitude to work.

Remember this reading experience?

The phone book. I'll hold the slide

so you can enjoy the nostalgia.

This is from the mid-'70s early trials

of Bell Centennial typeface I designed

for the U.S. phone books,

and it was my first experience of digital type,

and quite a baptism.

Designed for the phone books, as I said,

to be printed at tiny size on newsprint

on very high-speed rotary presses

with ink that was kerosene and lampblack.

This is not a hospitable environment

for a typographic designer.

So the challenge for me was to design type

that performed as well as possible

in these very adverse production conditions.

As I say, we were in the infancy of digital type.

I had to draw every character by hand

on quadrille graph paper --

there were four weights of Bell Centennial —

pixel by pixel, then encode
them raster line by raster line

for the keyboard.

It took two years, but I learned a lot.

These letters look as though they've been chewed

by the dog or something or other,

but the missing pixels at the intersections

of strokes or in the crotches

are the result of my studying the effects

of ink spread on cheap paper

and reacting, revising the font accordingly.

These strange artifacts are designed to compensate

for the undesirable effects of scale

and production process.

At the outset, AT&T had wanted

to set the phone books in Helvetica,

but as my friend Erik Spiekermann said

in the Helvetica movie, if you've seen that,

the letters in Helvetica were designed to be

as similar to one another as possible.

This is not the recipe for legibility at small size.

It looks very elegant up on a slide.

I had to disambiguate these forms

of the figures as much as possible in Bell Centennial

by sort of opening the shapes up, as you can see

in the bottom part of that slide.

So now we're on to the mid-'80s,

the early days of digital outline fonts,

vector technology.

There was an issue at that time

with the size of the fonts,

the amount of data that was required to find

and store a font in computer memory.

It limited the number of fonts you could get

on your typesetting system at any one time.

I did an analysis of the data,

and found that a typical serif face

you see on the left

needed nearly twice as much data

as a sans serif in the middle

because of all the points required

to define the elegantly curved serif brackets.

The numbers at the bottom of the slide, by the way,

they represent the amount of data

needed to store each of the fonts.

So the sans serif, in the middle,

sans the serifs, was much more economical,

81 to 151.

"Aha," I thought. "The engineers have a problem.

Designer to the rescue."

I made a serif type, you can see it on the right,

without curved serifs.

I made them polygonal, out
of straight line segments,

chamfered brackets.

And look, as economical in data as a sans serif.

We call it Charter, on the right.

So I went to the head of engineering

with my numbers, and I said proudly,

"I have solved your problem."

"Oh," he said. "What problem?"

And I said, "Well, you know, the problem

of the huge data you require
for serif fonts and so on."

"Oh," he said. "We solved that problem last week.

We wrote a compaction routine that reduces

the size of all fonts by an order of magnitude.

You can have as many fonts on your system

as you like."

"Well, thank you for letting me know," I said.

Foiled again.

I was left with a design solution

for a nonexistent technical problem.

But here is where the story sort
of gets interesting for me.

I didn't just throw my design away

in a fit of pique.

I persevered.

What had started as a technical exercise

became an aesthetic exercise, really.

In other words, I had come to like this typeface.

Forget its origins. Screw that.

I liked the design for its own sake.

The simplified forms of Charter

gave it a sort of plain-spoken quality

and unfussy spareness

that sort of pleased me.

You know, at times of technical innovation,

designers want to be influenced

by what's in the air.

We want to respond. We want to be pushed

into exploring something new.

So Charter is a sort of parable for me, really.

In the end, there was no hard and fast causal link

between the technology and the design of Charter.

I had really misunderstood the technology.

The technology did suggest something to me,

but it did not force my hand,

and I think this happens very often.

You know, engineers are very smart,

and despite occasional frustrations

because I'm less smart,

I've always enjoyed working with them

and learning from them.

Apropos, in the mid-'90s,

I started talking to Microsoft

about screen fonts.

Up to that point, all the fonts on screen

had been adapted from previously existing

printing fonts, of course.

But Microsoft foresaw correctly

the movement, the stampede

towards electronic communication,

to reading and writing onscreen

with the printed output as being sort of secondary

in importance.

So the priorities were just tipping at that point.

They wanted a small core set of fonts

that were not adapted but designed for the screen

to face up to the problems of screen,

which were their coarse resolution displays.

I said to Microsoft, a typeface designed

for a particular technology

is a self-obsoleting typeface.

I've designed too many faces in the past

that were intended to mitigate technical problems.

Thanks to the engineers, the
technical problems went away.

So did my typeface.

It was only a stopgap.

Microsoft came back to say that

affordable computer monitors

with better resolutions

were at least a decade away.

So I thought, well, a decade, that's not bad,

that's more than a stopgap.

So I was persuaded, I was convinced,

and we went to work on what became Verdana

and Georgia,

for the first time working not on paper

but directly onto the screen from the pixel up.

At that time, screens were binary.

The pixel was either on or it was off.

Here you see the outline of a letter,

the cap H,

which is the thin black line, the contour,

which is how it is stored in memory,

superimposed on the bitmap,

which is the grey area,

which is how it's displayed on the screen.

The bitmap is rasterized from the outline.

Here in a cap H, which is all straight lines,

the two are in almost perfect sync

on the Cartesian grid.

Not so with an O.

This looks more like bricklaying
than type design,

but believe me, this is a good bitmap O,

for the simple reason that it's symmetrical

in both x and y axes.

In a binary bitmap, you actually can't ask

for more than that.

I would sometimes make, I don't know,

three or four different versions of a difficult letter

like a lowercase A,

and then stand back to choose which was the best.

Well, there was no best,

so the designer's judgment comes in

in trying to decide

which is the least bad.

Is that a compromise?

Not to me, if you are working

at the highest standard the technology will allow,

although that standard may be

well short of the ideal.

You may be able to see on this slide

two different bitmap fonts there.

The "a" in the upper one, I think,

is better than the "a" in the lower one,

but it still ain't great.

You can maybe see the effect better

if it's reduced. Well, maybe not.

So I'm a pragmatist, not an idealist,

out of necessity.

For a certain kind of temperament,

there is a certain kind of satisfaction

in doing something that cannot be perfect

but can still be done to the best of your ability.

Here's the lowercase H from Georgia Italic.

The bitmap looks jagged and rough.

It is jagged and rough.

But I discovered, by experiment,

that there is an optimum slant

for an italic on a screen

so the strokes break well

at the pixel boundaries.

Look in this example how, rough as it is,

how the left and right legs

actually break at the same level.

That's a victory. That's good, right there.

And of course, at the lower depths,

you don't get much choice.

This is an S, in case you were wondering.

Well, it's been 18 years now

since Verdana and Georgia were released.

Microsoft were absolutely right,

it took a good 10 years,

but screen displays now do have

improved spatial resolution,

and very much improved photometric resolution

thanks to anti-aliasing and so on.

So now that their mission is accomplished,

has that meant the demise

of the screen fonts that I designed

for coarser displays back then?

Will they outlive the now-obsolete screens

and the flood of new web fonts

coming on to the market?

Or have they established their own

sort of evolutionary niche

that is independent of technology?

In other words, have they been absorbed

into the typographic mainstream?

I'm not sure, but they've had a good run so far.

Hey, 18 is a good age for anything

with present-day rates of attrition,

so I'm not complaining.

Thank you.

(Applause)