Fonts with variable metrics

blank's picture

Has anyone ever created or proposed a font format that allows metrics to vary with size? For example, a face might need more interletter spacing at small sizes and less at large sizes. So why not create a font that automatically increases or decreases interletter spacing by a set percentage as it changes from the default? Specific kerns could even be set to only activate above or below certain sizes.


blank's picture

Bill, would you please stop trying to derail threads by acting like Suetonius?

billtroop's picture

How can you figure out where you are now if you don't know how you got there and why?

blank's picture

Knowing how and why hardly requires us to know how well anyone got along with Robert Slimbach. So stuff the gossip or take it to the playground and share it with the other tween girls.

billtroop's picture

I don't think I can agree, James. Ultimately, gossip is what people talked about while they were doing things round which historical events, great and small, revolved. I think we can all agree that nobody in type has ever put more high quality work hours into multiple master design than Robert. Maybe one reason the technology isn't here has something to do with the extent to which people were willing to go to bat for him. Please understand: the technology was here - - we had it - - it was deeply embedded into the Mac and Windows way of life. To take the drastic step of ripping it out - - something that's almost unprecedented in technology - - ? I will never believe that personalities weren't involved. Same with GX and Win95. Personalities. All great events have small stories behind them.

Going forwards, there are other issues to be addressed. I'm a multiple master evangelist. I don't like to talk about the problems that existed with the fonts. MM problems don't get mindshare from me. But there were lots of people who did have problems and vowed never to use them. Adobe's general response was, they were using ancient PS drivers or ancient versions of ATM and needed to update their software. Was that all there was to it? I don't know. I bought into Adobe's explanation wholesale, as they say.

Were there similar problems with Variations fonts? Or did the GX model prevent that by simplifying the print stream that was ultimately seen by the printer? I guess we can never know, because there was probably never enough professional output. But I am guessing that, in this crucial architectural respect, Variations worked better than MM. Have I got this right? Is there something to learn from it?

charles ellertson's picture

From Bill Troop:

The user doesn’t even have to know it’s all happening. No fuss! No effort! Just superior type.

You seem to like outrageous language, so here's some for you. I think I'm a better typesetter than you. Keep your hands off my applications programs & what happens without my control. Give me good fonts & let me set the type.

One of the myriad of things I like about Matthew Carter is that he doesn't claim expertise in the use of type. Another person you seem to mention with a bit of affection is Sumner Stone -- I think he shares that view as well. I recommend it to you.

Charles Ellertson

billtroop's picture

Charles, I am willing to bet that you are a better typesetter than I am, even though I have never seen your typesetting and you have never seen mine. But Sumner is a good typesetter. Among other things, he composed the article I wrote about him in 2006 for MacDirectory ( and -- unfortunately the middle page, brilliantly showing off the different sizes, never got uploaded).

Re Suetonicity, my eye fell this morning on this week's Spectator in which Sam Leith reviews A.N. Wilson's concluding volume of British history. He heads the review with the sentence Wilson wrote to begin his chapter on Suez.

'Eden, the only male British prime minister known to have varnished his fingernails, was easily the best-looking individual, of either sex, to occupy that office in the 20th century.'

Leith admires the element of gossip and the 'irrelevantly judgmental'. I'll take my writing tips from Wilson, rather than a design undergrad (albeit a clever one), even though I was dreadfully upset by Wilson's slanderous treatment of Winifred Wagner. And I rejoice to say I do not know what a tween is, and I hope nobody will ever tell me.

charles ellertson's picture

Among other things, he composed the article I wrote about him in 2006 for MacDirectory

Hmm. Hard to tell with a PDF. Looks like Francesca Stauffacher's composition style; very similar to Type 1:1.

Typesetting (composition, if you want) is rarely about showing off type. Sometimes it is about showing off a product, sometimes communicating the written word. If you can't rewrite the copy, as is usually the case when the end is communicating another person's words, good composition is almost always about making good compromises. Whatever decisions you make, someone would always prefer you made others. For example, in the PDF you referenced, there are cases where, with ragged right composition, two letters are taken down. The graphic designer might approve; tough to find an editor who would let that pass.

But to stay on topic: With display type, be it an advertisement, a title page, or a chapter title, most anyone who makes any claim to typesetting skill is willing to fuss with the spacing of the letters -- the metrics. When you have 2,000 lines of foot- or endnotes, that isn't an agreeable option. The spacing of the smaller type better be good, because no comp I know is willing to adjust it in the setting part of the job. And as I said above, as important as the metrics are the character shape & weighting.

P.S. I like Sumner Stone, and have admired his both work and kindness to Jack Stauffacher. But we've never met face to face, & in any case, I wouldn't feel comfortable gossiping. I've also admired some of Chuck Bigelow's work. Is that allowed in your gossipy, gotta pick a side world?

billtroop's picture

Charles, I am way too ambivalent to pick sides because what interests me is less people than their work. With either, you can be on one side for one reason, and on the other for another reason. I regret to say I know very little about Bigelow but I have always respected him for having the eye to highlight Frutiger's incredible Icone and for his own and his partner's incredibly utilitarian (in the good sense) typeface design. If there's one person I regret not knowing more than another when I had the opportunity, it's Jack Stauffacher. I really respect you because you describe yourself as a compositor. Long may you multiply!

dezcom's picture

I have not seen Jack Stauffacher since I was a college kid. He ran the Laboratory Press at Carnegie Mellon (Carnegie Tech back then in the early '60s). It was a fun memory, watching him work! Thanks for reminding me, Charles!


billtroop's picture

Speaking of variable metrics and Jack Stauffacher, for several years he has only printed with two fonts: Metal Janson, and Stone's multi-size digital Cycles family. A lifetime of fine printing experience distills it down to those two fonts. Who really needs anything else?

billtroop's picture

Speaking of variable metrics and Jack Stauffacher, for several years he has only printed with two fonts: Metal Janson, and Stone's multi-size digital Cycles family. A lifetime of fine printing experience distills it down to those two fonts. Who really needs anything else?

charles ellertson's picture

I've always thought both the metal Janson's -- Linotype and Monotype -- were very nice fonts. I recut a version in Postscript/OT for use with lithographic DTP printing, for my (Tseng's) own use. We've used it a fair bit, including Brown / Retreat from Gettysburg.

(Small web sample, which never does any type justice. Only the interior, not the jacket, was set with this Janson.)

My problem with Janson is always what to use as display type.

As to the font generally, maybe some of the newer releases such as Storm have also solved the contrast problem of the letters, I don't know. In the original Linotype photocomp Janson fonts this was esp. egregious w/ offset printing. I do believe I like the metal Monotype cut over the Linotype, but it was one of the linecaster's better fonts. Printing with linecaster-set type was almost always worse than Monotype printed with any care.


billtroop's picture

Charles, I just get a blank from that link and I'm dying to see the type. Fred Brady when he was at Adobe was always talking about how he'd like to do a proper Kis revival, but of course nothing ever came of it. Sumner Stone made a valiant effort in his Autologic days which is best regarded as a practice run for later, more successful work. As late as the early 90s, one of the large printers was still using a very good Janson -- I'm thinking of one of the hardcover editions of 'Interview with a Vampire' -- and now I can't remember whether I thought it was metal or whether it was an amazing proprietary digitization that actually did the font justice. Which version does Stauffacher use?

charles ellertson's picture

Hmm. The link takes you, for some reason, to an "intermediate" place with the Amazon reader. Clicking on the little left-right arrows at the top of the gray turns a page & the text appears. If you want to go back to the ad for the book & get to the sample the long way, here's another link:

* * *

It is always best to look at type designed for high-resolution offset printing in a book. I'm not sure the pages on the link show anything worth judgment, be that favorable or unfavorable. Another book that I set using this font is Bach Perspectives Volume 1, designed by Dika Eckersley, wife of Richard Eckersly, and a fine designer in her own right. That title is probably more widely available in libraries. As I remember, I set vol. 1, vol. 2 was set at our shop, & the later volumes were set elsewhere.

I don't know which version of metal Janson Stauffacher uses. I would assume Monotype. I never knew him very well, our only contact was through Richard Eckersley, who was a very good friend. But now I'm gossiping; apologies to the thread.


Now that I think on it, that Janson was also used for the letters of Henry James (University of Nebraska Press), the last thing designed by Richard Eckersly. he died before the first volume was printed. It was a nice enough design, but will never make the book shows -- a lot of books Richard designed were for use, with nothing added excite show jurors (& detract from reading).

billtroop's picture

I've been mad about Eckersley's work ever since I discovered the philosopher Avital Ronell, with whom he worked for two decades though they never actually met. I'm now more than ever curious about your Janson. Can't you tell us a little more about it?

charles ellertson's picture

OK, with profound apologies to James Puckett, who's thread has now been hijacked beyond repair . . .

One of the problems us older photocomp comompsitor's face is when young designers come along & want to use classical fonts from the metal era; fonts they have read about, but all too often don't work well in photocomp. With the Linotron 202, Janson & even the later Janson Text were such fonts -- with a 10-point setting, the Linotron 202 Janson lower-case "o" looked like two tiny parentheses. Etc.

I got started using Fontographer somewhere around 1990, when the designer Mary Mendel was up to do the AAUP Book Show Catalog. Rather than design it & have it set (& printed) gratis -- the usual way for these catalogs to be produced -- she decided to take the then-new Macintosh/PageMaker tools & set it herself. She chose Ehrhardt & had at it, & then asked me for some help. First problem was the Ehrhardt, at that time a PostScript Type 3 font from Monotype. That led me to Fontographer.

Since I'd always liked Janson (metal), I started fooling around with it. All this is just background -- or maybe foreground -- to explain why I can't remember what's what. I don't remember just which version(s) of Janson I started with. On one corner of my table is PS Janson Text on a 5-1/2 inch floppy and the Ehrhardt in PS Type 3 format. There are some others of unknown origin, & specimens printed from Stempel Janson. Anyway, I didn't start from scratch, & strictly the fonts can only be used by us.

All this was going on while we were still setting type on the Linotron 202. The first person to use "my" re-worked Janson in a book was my wife, for one of her designs. As luck would have it, about the time the book was printed, George Mackie was over to the states & my wife got him to critique it. (He liked both the design and the type). I started using it for our own typesetting as we switched from the 202 to PostScript. All this preceded the PDF format, and everybody was still running out repro & pasting up pages on the light table. Printing plates were still burned from flats. I kept fooling around with the Janson according to how things looked when printed, when the amount of leading used by the design community changed, etc.

By the time PDF & DTP printing came along I was a bit tired of fiddling with it, but I think it holds up pretty well; several designers still specify it for interiors, and I still use it for some historical texts when Monticello seems inappropriate. Recently I converted it to OpenType, but didn't revisit either the characters or the kerning. It could obviously be improved, but best would be for one of today's talented young font designers to start over & create a new revival that could be commercially sold. Maybe there isn't enough interest though, revivals of classical designs don't seem to be of much interest to the new generation of type and graphic designers.

BTW, Richard's son, Sam Eckersley, is a designer living in your neighborhood.
Sam is a fine designer & shouldn't be though of as Richard's son any more than Richard should be though of as Tom's son. But maybe if you bought him a good dinner, he'd give you a copy of Remembering Richard, which was written, designed, and manufactured by Richard's friends. There is a nice piece by Ronell in the book, & she says "Of course, we would schedule meeting times in Berkeley or Lincoln or Paris--or later on in New York, to where his kids had migrated." So I'm not sure they never met.

billtroop's picture

Charles, it's a little confusing. In an interview somewhere on the web -- and I don't at all know when this dates from, Ronell said,

'And often I did argue with Richard Eckersley, the marvelous designer, because I felt that he was pulling away from the telephonic logic that I wanted his work to reflect and that he was becoming too autonomous–becoming a computer virtuoso. I didn't want the computer to overtake the telephonic markings that I felt needed to be continually asserted and reasserted. In a sense, we had a war of technologies–of course, over the telephone (I have never met Richard).'

Re the Janson, I think you should do something about it. A Janson that Eckersley used (in his last book no less) is a Janson we would all be interested in. I doubt that the original rights-holder would particularly care, but . . . I'll redigitize the font for you if you like -- no charge, no nothing -- just careful work!

Funnily enough, the first I heard of The Telephone Book was from Jeff Level, who thought it was the last great achievement of pre-Mac typesetting -- when in fact it was the first great piece of virtuoso typesetting on the Mac.

We are hijacking the thread, but in a sense it's relevant. Janson is one of the fonts that just doesn't react well to being designed as a single master. Variable metrics isn't enough. It can probably only work with a tiny size range. Janson is one of the toughest cookies! Another reason why I think you should find a way to make yours more available is that a young designer willing to put that much work into a font is more likely to spend the time on his or her own design.

charles ellertson's picture

Funnily enough, the first I heard of The Telephone Book was from Jeff Level, who thought it was the last great achievement of pre-Mac typesetting — when in fact it was the first great piece of virtuoso typesetting on the Mac.

No, Jeff Level is about right. If you're really interested in Richard Eckersley, you need to read Remembering Richard. There were only 300 copies printed, & it was privately circulated. I don't know who owns the rights, what permissions would be needed for more copies. Rich Hendel did most of the coordinative work, Paul Betz edited it, I set it, and Thomson-Shore printed it. After friends & family had their copies, the remainder were distributed at the AAUP Production Manager's Meeting. I believe I still have a PDF of the whole book; surely Rich Hendel does. I think he and Dika Eckersley would be the ones to ask about another distribution. There was a little bit of nonsense in it, but most was true to the man and his work.

I mention this because one of the pieces was by Michael Jenson, who, if anyone could be called the typesetter of The Telephone Book. I'll quote him:


Most all of The Telephone Book's base typography was not done on the Mac, but, rather, with "traditional" tools of phototypesetting, design creativity and visualization, creative use of darkroom equipment, and a lot of razorblade work. (p. 27)

. . .

We started by going over what was functionally possible using the Linotronic 202 and the Magna typesetting system. There was reasonably good WYSIWYG (What You See is What You Get) functionality on the Magna, but the L202 could also be pushed beyond what the Magma system could represent on the screen, which we also explored (p. 29)


A lot of us, including Richard, got tired of all the talk about The Telephone Book. Not the work itself, the talk. I remember providing Behren with some material. What came out in the article was "Hammer blows." Well, I said it, as a part of a sentence contrasting TTB to Eckersly's subtle humor, which was rather more prevalent in his designs. But that's not what got emphasized when the article was printed.

I suppose I'm guilty too. I said the design for The Letters of Henry James was Richard's last work. I remember having to make some last-minute decisions without him. But as anyone who's worked in a publishing house knows, a designer always has a number of projects going on, some nearing completion (final proof), some going to/just back as first proof, some still on the drawing board, some just sketched. There was no "last book." Annie Shahan, who more or less succeeded Richard as Senior Designer at Nebraska, would have the strongest vote as to the "last book."

As to the Janson, if you want to digitize it & have it available as an Open Source Font, fine with me.

billtroop's picture

Let's do it. I like the idea of open source.

I'm glad Jeff was right after all -- mostly - - and so will he be. But isn't it amazing how difficult it is to establish the facts? how complex the story is? Practically anything you say is just the tip of the iceberg!

Syndicate content Syndicate content