Although the unit system used is unclear, it appears that the Roman design administration had a thoroughly worked-out system for the measurement of inscriptional letters, which allowed them to cut inscriptions in matching lettering styles and in consistent sizes throughout the extremely widespread area under Roman rule.

“It was part of a visual identity that shouted ‘Rome!’,” said the Institut’s vice-director, Robespierre Danton, waving his arms enthusiastically at the partially excavated site. “They projected their power and their brand through a coordinated system of graphics that was instantly recognizable anywhere in the Mediterranean world.” The manual’s threadbare pages, according to Danton, specify exactly how the visual system should be implemented, with hints (barely legible) of extreme penalties for misuse of the empire’s intellectual property.

Although the fragments are in a poor state of preservation, one intriguing supplementary find has excited the interest of Dr. Giambattista Farben, a color researcher with the Institut. “This broken tablet, made of baked and polished tufa,” he says, “was found in close proximity to the manual itself. The tablet shows traces of a pattern of varying colors in lead-based paint, and scratches that may be notations to identify the different colors.” Dr. Farben was cautious, but he said that one theory of the colored tablet was that it constituted a color chart for painters who would turn the Romans’ marble walls into a panoply of colors. “It could be the earliest Pantone matching system,” admitted Dr. Farben.

Scholars from the University of Northern California dispute the primacy of the Roman identity system. Professor Chien Su-ma of UNC says that he has spent more than twenty years cataloging a collection of inscribed tortoise shells found under a pile of Han-dynasty tax receipts at Dunhuang, on the edge of the Taklamakan Desert, in China’s Gansu province. “The Han Dynasty had a clearly defined visual identity,” claims Prof. Chien, “and I believe these fragments, which were preserved at a major entrepot and outpost of empire, are a key to the system in its earliest form. They certainly predate this Western find by at least a century.”

[Photo: Detail of the lettering at the base of Trajan's column, in Rome.]

Except in the execution of its typography. The choice of Bodoni for the type on this window text was clearly meant to emphasize the classic elegance of the brand. But the effect is spoiled by the typewriter apostrophes, which neither Giambattista Bodoni nor any type designer up until the advent of desktop publishing had ever conceived of. (It’s further spoiled by the fact that the second apostrophe doesn’t even belong there: the adjective is its, not it’s.)

Verdana was designed in the 1990s for Microsoft, developed specifically as a typeface for reading onscreen. The designer, Matthew Carter, has long experience of virtually every kind of typeface technology, and he brought that to bear on designing Verdana. Since text on a computer screen appears, of necessity, at pretty coarse resolution, the outlines of the letters have to be adapted somehow when rendering them at small sizes; there simply aren’t enough pixels available to reproduce the outline shapes perfectly. That’s where the art and craft of designing screen fonts comes in: making the most of those extreme limitations. In what was at the time a revolutionary turnabout, Carter first designed bitmapped letters for each of the target sizes, positioning pixels to get the most legible shapes he could; then he drew the outlines for the higher-resolution letters, based on the shapes of the lo-res bitmaps. Tom Rickner, a wizard of digital font technology, then created the “hints” that would tell the font software exactly how to distort the outlines at a particular size, when drawing a character on the screen, in order to achieve the ideal bitmap at that size.

One of the things that make Verdana legible onscreen, compared with a lot of other typefaces, is the generous space around the characters. There’s always a tendency among web designers to try to cram in as much material as possible in the space available, but that works against clarity and legibility. Without enough space between the letters, they all tend to run together. We’ve all seen this, much too frequently, on our computer screens. The clear, open shapes of Verdana’s letters can vary quite a bit from size to size at small text sizes onscreen, but one thing they have in common is that they’ve been given enough space to breathe.

Although Verdana was meant primarily for onscreen reading, it works surprisingly well on paper as well. It’s a simple, clean, unpretentious sans serif typeface, easy to read. I’ve used it for years as the typeface for manuscripts and drafts of anything I’m writing, because it’s easy to read both onscreen and on paper and it gets out of the way. I realized seven or eight years ago that Verdana had passed into general use, when I saw it on a billboard in San Francisco. (The same characteristics that make it legible onscreen may make it easy to read at a distance as you’re driving by.) I’ve never tried using Verdana in print, but I can imagine situations where I might want to.

It’s funny to see the choice of Verdana lambasted because it was designed for a different purpose. As Erik Spiekermann has pointed out, many of our most versatile typefaces were originally designed for one specific purpose, answering a particular set of constraints (Times New Roman, for instance, which was designed for the presses that printed The Times in 1931). Even Bell Centennial and Bell Gothic, both of which were designed for the listings in American telephone books, have been used successfully at huge display sizes by editorial designers with an eye for the unusual. Perhaps Verdana has unexpected uses as well.

I have no strong opinion about Ikea’s redesign. Certainly Verdana’s numerals are very clear and readable – even stylish, in a chunky, sturdy sort of way – and the numerals are what end up at the largest size on the pages of an Ikea catalog. And I alway felt that the Ikea version of Futura was a little too tightly spaced, though that’s not the fault of the typeface but of how it’s used.

One of the reasons Ikea chose Verdana is that it works across quite a lot of languages and scripts. The basic fonts include Greek and Cyrillic alongside the extended Latin alphabet; and Microsoft’s Japanese typeface Meiryo is based on Verdana, with the romaji (Latin letters) being essentially slightly revised and sharpened versions of Verdana’s designs. (As near as I can tell, from Ikea’s Japanese web pages, the Japanese catalog does use Meiryo, although with a different typeface for some text.)

Verdana may be about to become more versatile for both web and print use, since Ascender Corporation just announced that they are working with Matthew Carter and the Font Bureau to extend both the Verdana and the Georgia families with new weights and widths.

Whatever the merits of the case, what strikes me most forcefully in all of this is that a debate about which font to use could even be noticed, much less become a cause célèbre in the public consciousness. What typographic times we live in!

[Images: two details from Ikea's U.S. website (top and middle); sample of some of the forthcoming new members of the Verdana and Georgia families.]

Yesterday I dropped by a local website that I check periodically, Crosscut, a Seattle-based news and comment site with some smart thinking and good writing. As I browsed down the home page, I came across an ad from Amazon – and stopped cold. The four books being advertised were all design and printing books. All of them. How likely is that? Would the average reader have a yen for books about graphic design? No, but I would; in fact, my recent browsing habits on Amazon would probably show a lot of books about typography and related subjects. Was this ad tailored specifically to me?

The answer is yes. If you click on the ad’s link to “Privacy Information,” you find that, indeed, Amazon is using their records of what you’ve looked at on their site, and choosing which books to push in their ad based on that. Which means that each Amazon ad on Crosscut (or any other website that hosts such an ad) varies its content depending on who’s looking at it, or more precisely whose computer you’re using to look at it. The future is here.

If you refresh the web page, the Amazon ad changes its contents – but they’re still based on your own recent patterns at Amazon’s own site. The contents even seem to vary by web browser; I was using Safari when I discovered this, but logging on to Crosscut from Firefox brought up an Amazon ad with no obvious relationship to me. (Something to do with how my preferences are set in the two browsers? Probably.)

I’m never surprised by invasions of info-privacy; I’ve read enough science fiction to have been expecting this for a long time. But the potential for embarrassment, at the very least, is large. You log on from somebody else’s computer, and when you get to the Amazon ad, you see a suspicious number of books about…typography! “You, uh, look at a lot of, uh, ‘type,’ don’t you?” you ask your friend. His guilty secret is out, for all the world to see.

Update March 10:

Today’s New York Times has an article on this very subject – how web ads are being targeted directly to individuals – but it conspicuously fails to mention Amazon. A surprising omission.