My life in typefaces
-
0:01 - 0:02Type is something we consume
-
0:02 - 0:04in enormous quantities.
-
0:04 - 0:05In much of the world,
-
0:05 - 0:07it's completely inescapable.
-
0:07 - 0:10But few consumers are concerned to know
-
0:10 - 0:12where a particular typeface came from
-
0:12 - 0:15or when or who designed it,
-
0:15 - 0:19if, indeed, there was any human agency involved
-
0:19 - 0:21in its creation, if it didn't just sort of materialize
-
0:21 - 0:25out of the software ether.
-
0:25 - 0:29But I do have to be concerned with those things.
-
0:29 - 0:30It's my job.
-
0:30 - 0:32I'm one of the tiny handful of people
-
0:32 - 0:34who gets badly bent out of shape
-
0:34 - 0:37by the bad spacing of the T and the E
-
0:37 - 0:39that you see there.
-
0:39 - 0:40I've got to take that slide off.
-
0:40 - 0:42I can't stand it. Nor can Chris.
-
0:42 - 0:44There. Good.
-
0:44 - 0:46So my talk is about the connection
-
0:46 - 0:49between technology and design of type.
-
0:49 - 0:52The technology has changed
-
0:52 - 0:55a number of times since I started work:
-
0:55 - 0:59photo, digital, desktop, screen, web.
-
0:59 - 1:01I've had to survive those changes and try
-
1:01 - 1:04to understand their implications for what I do
-
1:04 - 1:05for design.
-
1:05 - 1:10This slide is about the effect of tools on form.
-
1:10 - 1:13The two letters, the two K's,
-
1:13 - 1:17the one on your left, my right, is modern,
-
1:17 - 1:18made on a computer.
-
1:18 - 1:20All straight lines are dead straight.
-
1:20 - 1:23The curves have that kind of
mathematical smoothness -
1:23 - 1:27that the Bézier formula imposes.
-
1:27 - 1:29On the right, ancient Gothic,
-
1:29 - 1:33cut in the resistant material of steel by hand.
-
1:33 - 1:35None of the straight lines are actually straight.
-
1:35 - 1:38The curves are kind of subtle.
-
1:38 - 1:42It has that spark of life from the human hand
-
1:42 - 1:44that the machine or the program
-
1:44 - 1:46can never capture.
-
1:46 - 1:48What a contrast.
-
1:48 - 1:51Well, I tell a lie.
-
1:51 - 1:54A lie at TED. I'm really sorry.
-
1:54 - 1:56Both of these were made on a computer,
-
1:56 - 1:58same software, same Bézier curves,
-
1:58 - 1:59same font format.
-
1:59 - 2:02The one on your left
-
2:02 - 2:04was made by Zuzana Licko at Emigre,
-
2:04 - 2:06and I did the other one.
-
2:06 - 2:09The tool is the same, yet the letters are different.
-
2:09 - 2:11The letters are different
-
2:11 - 2:12because the designers are different.
-
2:12 - 2:15That's all. Zuzana wanted hers to look like that.
-
2:15 - 2:18I wanted mine to look like that. End of story.
-
2:18 - 2:20Type is very adaptable.
-
2:20 - 2:24Unlike a fine art, such as sculpture or architecture,
-
2:24 - 2:27type hides its methods.
-
2:27 - 2:30I think of myself as an industrial designer.
-
2:30 - 2:31The thing I design is manufactured,
-
2:31 - 2:33and it has a function:
-
2:33 - 2:35to be read, to convey meaning.
-
2:35 - 2:37But there is a bit more to it than that.
-
2:37 - 2:39There's the sort of aesthetic element.
-
2:39 - 2:41What makes these two letters different
-
2:41 - 2:44from different interpretations by different designers?
-
2:44 - 2:46What gives the work of some designers
-
2:46 - 2:49sort of characteristic personal style,
-
2:49 - 2:52as you might find in the work of a fashion designer,
-
2:52 - 2:55an automobile designer, whatever?
-
2:55 - 2:57There have been some cases, I admit,
-
2:57 - 2:58where I as a designer
-
2:58 - 3:01did feel the influence of technology.
-
3:01 - 3:04This is from the mid-'60s,
-
3:04 - 3:06the change from metal type to photo,
-
3:06 - 3:08hot to cold.
-
3:08 - 3:09This brought some benefits
-
3:09 - 3:12but also one particular drawback:
-
3:12 - 3:15a spacing system that only provided
-
3:15 - 3:1918 discrete units for letters
-
3:19 - 3:22to be accommodated on.
-
3:22 - 3:24I was asked at this time to design
-
3:24 - 3:26a series of condensed sans serif types
-
3:26 - 3:29with as many different variants as possible
-
3:29 - 3:33within this 18-unit box.
-
3:33 - 3:35Quickly looking at the arithmetic,
-
3:35 - 3:38I realized I could only actually make three
-
3:38 - 3:42of related design. Here you see them.
-
3:42 - 3:44In Helvetica Compressed, Extra Compressed,
-
3:44 - 3:48and Ultra Compressed, this rigid 18-unit system
-
3:48 - 3:50really boxed me in.
-
3:50 - 3:51It kind of determined the proportions
-
3:51 - 3:54of the design.
-
3:54 - 3:58Here are the typefaces, at least the lower cases.
-
3:58 - 4:00So do you look at these and say,
-
4:00 - 4:04"Poor Matthew, he had to submit to a problem,
-
4:04 - 4:07and by God it shows in the results."
-
4:07 - 4:09I hope not.
-
4:09 - 4:11If I were doing this same job today,
-
4:11 - 4:14instead of having 18 spacing units,
-
4:14 - 4:17I would have 1,000.
-
4:17 - 4:19Clearly I could make more variants,
-
4:19 - 4:24but would these three members
of the family be better? -
4:24 - 4:26It's hard to say without actually doing it,
-
4:26 - 4:28but they would not be better in the proportion
-
4:28 - 4:31of 1,000 to 18, I can tell you that.
-
4:31 - 4:33My instinct tells you that any improvement
-
4:33 - 4:36would be rather slight, because they were designed
-
4:36 - 4:39as functions of the system they were designed to fit,
-
4:39 - 4:41and as I said, type is very adaptable.
-
4:41 - 4:44It does hide its methods.
-
4:44 - 4:46All industrial designers work within constraints.
-
4:46 - 4:49This is not fine art.
-
4:49 - 4:51The question is, does a constraint
-
4:51 - 4:53force a compromise?
-
4:53 - 4:55By accepting a constraint,
-
4:55 - 4:57are you working to a lower standard?
-
4:57 - 4:59I don't believe so, and I've always been encouraged
-
4:59 - 5:02by something that Charles Eames said.
-
5:02 - 5:03He said he was conscious of working
-
5:03 - 5:04within constraints,
-
5:04 - 5:07but not of making compromises.
-
5:07 - 5:10The distinction between a constraint
-
5:10 - 5:12and a compromise is obviously very subtle,
-
5:12 - 5:18but it's very central to my attitude to work.
-
5:18 - 5:21Remember this reading experience?
-
5:21 - 5:22The phone book. I'll hold the slide
-
5:22 - 5:27so you can enjoy the nostalgia.
-
5:27 - 5:30This is from the mid-'70s early trials
-
5:30 - 5:32of Bell Centennial typeface I designed
-
5:32 - 5:34for the U.S. phone books,
-
5:34 - 5:37and it was my first experience of digital type,
-
5:37 - 5:41and quite a baptism.
-
5:41 - 5:43Designed for the phone books, as I said,
-
5:43 - 5:46to be printed at tiny size on newsprint
-
5:46 - 5:49on very high-speed rotary presses
-
5:49 - 5:51with ink that was kerosene and lampblack.
-
5:51 - 5:55This is not a hospitable environment
-
5:55 - 5:59for a typographic designer.
-
5:59 - 6:00So the challenge for me was to design type
-
6:00 - 6:02that performed as well as possible
-
6:02 - 6:07in these very adverse production conditions.
-
6:07 - 6:10As I say, we were in the infancy of digital type.
-
6:10 - 6:12I had to draw every character by hand
-
6:12 - 6:14on quadrille graph paper --
-
6:14 - 6:16there were four weights of Bell Centennial —
-
6:16 - 6:19pixel by pixel, then encode
them raster line by raster line -
6:19 - 6:20for the keyboard.
-
6:20 - 6:25It took two years, but I learned a lot.
-
6:25 - 6:26These letters look as though they've been chewed
-
6:26 - 6:28by the dog or something or other,
-
6:28 - 6:30but the missing pixels at the intersections
-
6:30 - 6:31of strokes or in the crotches
-
6:31 - 6:35are the result of my studying the effects
-
6:35 - 6:38of ink spread on cheap paper
-
6:38 - 6:41and reacting, revising the font accordingly.
-
6:41 - 6:44These strange artifacts are designed to compensate
-
6:44 - 6:47for the undesirable effects of scale
-
6:47 - 6:50and production process.
-
6:50 - 6:52At the outset, AT&T had wanted
-
6:52 - 6:56to set the phone books in Helvetica,
-
6:56 - 6:57but as my friend Erik Spiekermann said
-
6:57 - 7:00in the Helvetica movie, if you've seen that,
-
7:00 - 7:02the letters in Helvetica were designed to be
-
7:02 - 7:05as similar to one another as possible.
-
7:05 - 7:08This is not the recipe for legibility at small size.
-
7:08 - 7:10It looks very elegant up on a slide.
-
7:10 - 7:13I had to disambiguate these forms
-
7:13 - 7:16of the figures as much as possible in Bell Centennial
-
7:16 - 7:18by sort of opening the shapes up, as you can see
-
7:18 - 7:21in the bottom part of that slide.
-
7:21 - 7:23So now we're on to the mid-'80s,
-
7:23 - 7:26the early days of digital outline fonts,
-
7:26 - 7:28vector technology.
-
7:28 - 7:30There was an issue at that time
-
7:30 - 7:32with the size of the fonts,
-
7:32 - 7:35the amount of data that was required to find
-
7:35 - 7:40and store a font in computer memory.
-
7:40 - 7:42It limited the number of fonts you could get
-
7:42 - 7:45on your typesetting system at any one time.
-
7:45 - 7:49I did an analysis of the data,
-
7:49 - 7:51and found that a typical serif face
-
7:51 - 7:53you see on the left
-
7:53 - 7:55needed nearly twice as much data
-
7:55 - 7:57as a sans serif in the middle
-
7:57 - 8:00because of all the points required
-
8:00 - 8:04to define the elegantly curved serif brackets.
-
8:04 - 8:07The numbers at the bottom of the slide, by the way,
-
8:07 - 8:09they represent the amount of data
-
8:09 - 8:13needed to store each of the fonts.
-
8:13 - 8:15So the sans serif, in the middle,
-
8:15 - 8:18sans the serifs, was much more economical,
-
8:18 - 8:2081 to 151.
-
8:20 - 8:24"Aha," I thought. "The engineers have a problem.
-
8:24 - 8:26Designer to the rescue."
-
8:26 - 8:29I made a serif type, you can see it on the right,
-
8:29 - 8:30without curved serifs.
-
8:30 - 8:33I made them polygonal, out
of straight line segments, -
8:33 - 8:35chamfered brackets.
-
8:35 - 8:39And look, as economical in data as a sans serif.
-
8:39 - 8:42We call it Charter, on the right.
-
8:42 - 8:44So I went to the head of engineering
-
8:44 - 8:46with my numbers, and I said proudly,
-
8:46 - 8:48"I have solved your problem."
-
8:48 - 8:52"Oh," he said. "What problem?"
-
8:52 - 8:53And I said, "Well, you know, the problem
-
8:53 - 8:57of the huge data you require
for serif fonts and so on." -
8:57 - 9:00"Oh," he said. "We solved that problem last week.
-
9:00 - 9:03We wrote a compaction routine that reduces
-
9:03 - 9:05the size of all fonts by an order of magnitude.
-
9:05 - 9:07You can have as many fonts on your system
-
9:07 - 9:09as you like."
-
9:09 - 9:11"Well, thank you for letting me know," I said.
-
9:11 - 9:13Foiled again.
-
9:13 - 9:15I was left with a design solution
-
9:15 - 9:19for a nonexistent technical problem.
-
9:19 - 9:22But here is where the story sort
of gets interesting for me. -
9:22 - 9:25I didn't just throw my design away
-
9:25 - 9:26in a fit of pique.
-
9:26 - 9:28I persevered.
-
9:28 - 9:30What had started as a technical exercise
-
9:30 - 9:33became an aesthetic exercise, really.
-
9:33 - 9:36In other words, I had come to like this typeface.
-
9:36 - 9:38Forget its origins. Screw that.
-
9:38 - 9:41I liked the design for its own sake.
-
9:41 - 9:43The simplified forms of Charter
-
9:43 - 9:45gave it a sort of plain-spoken quality
-
9:45 - 9:47and unfussy spareness
-
9:47 - 9:49that sort of pleased me.
-
9:49 - 9:52You know, at times of technical innovation,
-
9:52 - 9:54designers want to be influenced
-
9:54 - 9:55by what's in the air.
-
9:55 - 9:58We want to respond. We want to be pushed
-
9:58 - 10:01into exploring something new.
-
10:01 - 10:04So Charter is a sort of parable for me, really.
-
10:04 - 10:08In the end, there was no hard and fast causal link
-
10:08 - 10:11between the technology and the design of Charter.
-
10:11 - 10:15I had really misunderstood the technology.
-
10:15 - 10:18The technology did suggest something to me,
-
10:18 - 10:20but it did not force my hand,
-
10:20 - 10:23and I think this happens very often.
-
10:23 - 10:25You know, engineers are very smart,
-
10:25 - 10:27and despite occasional frustrations
-
10:27 - 10:28because I'm less smart,
-
10:28 - 10:30I've always enjoyed working with them
-
10:30 - 10:32and learning from them.
-
10:32 - 10:35Apropos, in the mid-'90s,
-
10:35 - 10:37I started talking to Microsoft
-
10:37 - 10:40about screen fonts.
-
10:40 - 10:42Up to that point, all the fonts on screen
-
10:42 - 10:45had been adapted from previously existing
-
10:45 - 10:47printing fonts, of course.
-
10:47 - 10:50But Microsoft foresaw correctly
-
10:50 - 10:52the movement, the stampede
-
10:52 - 10:55towards electronic communication,
-
10:55 - 10:57to reading and writing onscreen
-
10:57 - 11:00with the printed output as being sort of secondary
-
11:00 - 11:02in importance.
-
11:02 - 11:06So the priorities were just tipping at that point.
-
11:06 - 11:08They wanted a small core set of fonts
-
11:08 - 11:11that were not adapted but designed for the screen
-
11:11 - 11:14to face up to the problems of screen,
-
11:14 - 11:18which were their course resolution displays.
-
11:18 - 11:21I said to Microsoft, a typeface designed
-
11:21 - 11:23for a particular technology
-
11:23 - 11:26is a self-obsoleting typeface.
-
11:26 - 11:28I've designed too many faces in the past
-
11:28 - 11:32that were intended to mitigate technical problems.
-
11:32 - 11:35Thanks to the engineers, the
technical problems went away. -
11:35 - 11:37So did my typeface.
-
11:37 - 11:40It was only a stopgap.
-
11:40 - 11:42Microsoft came back to say that
-
11:42 - 11:43affordable computer monitors
-
11:43 - 11:45with better resolutions
-
11:45 - 11:47were at least a decade away.
-
11:47 - 11:50So I thought, well, a decade, that's not bad,
-
11:50 - 11:52that's more than a stopgap.
-
11:52 - 11:54So I was persuaded, I was convinced,
-
11:54 - 11:57and we went to work on what became Verdana
-
11:57 - 11:58and Georgia,
-
11:58 - 12:01for the first time working not on paper
-
12:01 - 12:04but directly onto the screen from the pixel up.
-
12:04 - 12:08At that time, screens were binary.
-
12:08 - 12:11The pixel was either on or it was off.
-
12:11 - 12:14Here you see the outline of a letter,
-
12:14 - 12:16the cap H,
-
12:16 - 12:18which is the thin black line, the contour,
-
12:18 - 12:21which is how it is stored in memory,
-
12:21 - 12:23superimposed on the bitmap,
-
12:23 - 12:25which is the grey area,
-
12:25 - 12:27which is how it's displayed on the screen.
-
12:27 - 12:30The bitmap is rasterized from the outline.
-
12:30 - 12:32Here in a cap H, which is all straight lines,
-
12:32 - 12:34the two are in almost perfect sync
-
12:34 - 12:39on the Cartesian grid.
-
12:39 - 12:42Not so with an O.
-
12:42 - 12:45This looks more like bricklaying
than type design, -
12:45 - 12:48but believe me, this is a good bitmap O,
-
12:48 - 12:50for the simple reason that it's symmetrical
-
12:50 - 12:52in both x and y axes.
-
12:52 - 12:55In a binary bitmap, you actually can't ask
-
12:55 - 12:57for more than that.
-
12:57 - 12:59I would sometimes make, I don't know,
-
12:59 - 13:01three or four different versions of a difficult letter
-
13:01 - 13:03like a lowercase A,
-
13:03 - 13:06and then stand back to choose which was the best.
-
13:06 - 13:09Well, there was no best,
-
13:09 - 13:11so the designer's judgment comes in
-
13:11 - 13:12in trying to decide
-
13:12 - 13:15which is the least bad.
-
13:15 - 13:18Is that a compromise?
-
13:18 - 13:19Not to me, if you are working
-
13:19 - 13:23at the highest standard the technology will allow,
-
13:23 - 13:25although that standard may be
-
13:25 - 13:27well short of the ideal.
-
13:27 - 13:29You may be able to see on this slide
-
13:29 - 13:31two different bitmap fonts there.
-
13:31 - 13:33The "a" in the upper one, I think,
-
13:33 - 13:35is better than the "a" in the lower one,
-
13:35 - 13:37but it still ain't great.
-
13:37 - 13:39You can maybe see the effect better
-
13:39 - 13:42if it's reduced. Well, maybe not.
-
13:42 - 13:45So I'm a pragmatist, not an idealist,
-
13:45 - 13:46out of necessity.
-
13:46 - 13:48For a certain kind of temperament,
-
13:48 - 13:50there is a certain kind of satisfaction
-
13:50 - 13:54in doing something that cannot be perfect
-
13:54 - 13:57but can still be done to the best of your ability.
-
13:57 - 14:02Here's the lowercase H from Georgia Italic.
-
14:02 - 14:05The bitmap looks jagged and rough.
-
14:05 - 14:06It is jagged and rough.
-
14:06 - 14:08But I discovered, by experiment,
-
14:08 - 14:12that there is an optimum slant
-
14:12 - 14:14for an italic on a screen
-
14:14 - 14:16so the strokes break well
-
14:16 - 14:18at the pixel boundaries.
-
14:18 - 14:21Look in this example how, rough as it is,
-
14:21 - 14:23how the left and right legs
-
14:23 - 14:25actually break at the same level.
-
14:25 - 14:29That's a victory. That's good, right there.
-
14:29 - 14:32And of course, at the lower depths,
-
14:32 - 14:34you don't get much choice.
-
14:34 - 14:39This is an S, in case you were wondering.
-
14:39 - 14:41Well, it's been 18 years now
-
14:41 - 14:44since Verdana and Georgia were released.
-
14:44 - 14:46Microsoft were absolutely right,
-
14:46 - 14:48it took a good 10 years,
-
14:48 - 14:50but screen displays now do have
-
14:50 - 14:53improved spatial resolution,
-
14:53 - 14:56and very much improved photometric resolution
-
14:56 - 15:00thanks to anti-aliasing and so on.
-
15:00 - 15:03So now that their mission is accomplished,
-
15:03 - 15:05has that meant the demise
-
15:05 - 15:07of the screen fonts that I designed
-
15:07 - 15:10for courser displays back then?
-
15:10 - 15:13Will they outlive the now-obsolete screens
-
15:13 - 15:15and the flood of new web fonts
-
15:15 - 15:17coming on to the market?
-
15:17 - 15:18Or have they established their own
-
15:18 - 15:21sort of evolutionary niche
-
15:21 - 15:24that is independent of technology?
-
15:24 - 15:26In other words, have they been absorbed
-
15:26 - 15:29into the typographic mainstream?
-
15:29 - 15:33I'm not sure, but they've had a good run so far.
-
15:33 - 15:36Hey, 18 is a good age for anything
-
15:36 - 15:38with present-day rates of attrition,
-
15:38 - 15:40so I'm not complaining.
-
15:40 - 15:42Thank you.
-
15:42 - 15:45(Applause)
- Title:
- My life in typefaces
- Speaker:
- Matthew Carter
- Description:
-
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.
- Video Language:
- English
- Team:
- closed TED
- Project:
- TEDTalks
- Duration:
- 16:01
Morton Bast edited English subtitles for My life in typefaces | ||
Morton Bast edited English subtitles for My life in typefaces | ||
Morton Bast approved English subtitles for My life in typefaces | ||
Madeleine Aronson accepted English subtitles for My life in typefaces | ||
Morton Bast edited English subtitles for My life in typefaces | ||
Morton Bast edited English subtitles for My life in typefaces | ||
Morton Bast edited English subtitles for My life in typefaces | ||
Morton Bast edited English subtitles for My life in typefaces |