The Serif is Dying

The pendulum will swing back once more. As soon as all k3wl designers notice that Tom, Dick and Harry are using sans-serifs, they'll opt for serifs.

I look into the crystal, and I see ... the popularity of sans-serifs will lean more and more towards cross-over fonts (grostesque with more detail or just slab serifs), until the trend will transform into clean serif fonts.

I think the "clean serif" trend will happen somewhat in parallel (but a bit later) to the Grotesque (AKA Vernacular) trend that's now gathering steam.

hhp

The reason is simple, the sans has come of age. Soon the readability studies will all show the sans is preferred.

> Soon the readability studies will all show the sans is preferred.

*Legibility*, sure. In fact it's not a matter of "soon": many studies have shown that already, even though (because?) they don't distinguish between that and *readability*.

Readability depends on letterforms coming together to form words, and serifs help them do that. It sounds paradoxical, but serif letterforms make for more readable fonts partly *because* they are less legible as individual letters! They are like worker ants that ensure overall functionality, as opposed to human actors on a stage.

As for Familiarity, yes it's fundamental (not least in the familiarity with word shapes, even more than letters), but it's not the only thing. And even the true role of familiarity has been ignored by its worshipers, probably because they're too busy being "artiste"s to bother *thinking* for a change.

There is a reason FF Extra can never be as readable as FF Celeste, and it's essentially a *physical* reason, inexorably tied to our flesh and blood. Javal discovered how we really read over 100 years ago - please, let's stop these relativist delusions already.

hhp

Might it simply be that the top ten queries at Identifont are sans designs merely because they're harder to tall apart than serifs? There are so many new post-Meta designs from Europe, all good, but harder to identify without having a specimen handy. Fedra, Info, Scala Sans, Quadraat Sans, Prolinea, the list just goes on and on.

I still can't even always tell serif book faces apart in the roman. I have to find a phrase in italics before I know exactly what it is.

Serifs aren't going to go away.

Sans and display faces may outsell serif faces, but that doesn't support the conclusion that serifs are being used less overall (it may be true in display work, for the reasons following). A single likeable (or even boring and despised) serif face may serve countless books, newspapers, etc. It's versatile in the sense of the well-dressed man or woman, neither distractingly shabby nor distractingly flashy, just the right amount of style to communicate without drawing attention to the typeface instead of the message. By contrast, flamboyant display and subtly stylish sans faces "burn out" quickly, serving either a single short-lived project or being `hot' one month and "yesterday's news" the next. So you need to buy more disposable fonts, but most people already have a dependable collection of stand-bys. Sales figures do communicate a marketing message for typefounders and distributors -- but not necessarily a message about usage, especially in heavy text publishers' realm.

i suppose it makes sense to claim that serifed fonts are declining in use, since their birth was in pen-and-ink, where serifs actually made sense. most of today's fonts are constructed in postscript. serifs become just another stylistic maneuver with that medium.

i actually use serifed fonts quite regularly, but the evidence is never shown online...primarily because i'm working in OS X (codename "Vaseline Vision" in our house) which makes serif faces even more difficult to read onscreen that they are to begin with. i'm sure the dearth of online work over the past few years has lots to do with the decline.

I don't see any evidence that serif faces are declining in use. A thoughtful graphic designer (Brian Morgan, who is designing the promo material for next year's ATypI conference in Vancouver) explains the situation very succinctly: Every graphic design company needs a handful of really good serif faces and *lots* of different sans serif faces, because the latter are used to help build visual identities for clients.

Sans serif faces are not replacing serif faces: they're doing a different job that requires greater variety and, at least for the client, the illusion of uniqueness.

> that doesn't support the conclusion that serifs are being used less overall

> I don't see any evidence that serif faces are declining in use.

Did anybody say that? Maybe.

In terms of usage per font chosen, I think they're about equal. In terms of usage per typeset letter, serifs win by a landslide. It's only if you factor in point size (ie visibility) that sans have been gaining.

I thought the talk was more about sales, and the interesting possibility that one day serif faces will become the flamboyant cash-cows.

hhp

BTW, would it help or hinder if serif fonts (or least those for text) were priced higher?

Since they have more usefulness/longevity...

hhp

Hrant: 'Did anybody say that?'

Yes, PK in the post immediately before mine wrote: 'i suppose it makes sense to claim that serifed fonts are declining in use'.

It was also implied, at least, by some of the comments on the ATypI members list, although I think most of these comments shjould be read as confined to particular contexts of graphic design. As Gerard Unger pointed out, our books, magazines and newspapers don't seem to be abandonind serif type in any hurry. Thankfully.

i think it's also notable to mention that those comments were made on a website, not in a book or newspaper. i'd think that many people posting there may well be thinking in a specifically digital context. i wonder how much of the work produced by ATypI is web-bound. that would definitely bias the discussion.

Following Hrant's advice on attempting logic before art, I stumbled upon this conclusion; technology is not built for serifs. Think about it - cell phones and such...even highway signs - why do you think they're sans-serifs? It's more legible, not readable(in the context of the comments above). Although this is the information age, our life is 'hustle-bustle' and sans-serifs logically fit.

Could you see your cell phone screen set in Mrs. Eaves?

... :X

While I agree with several of the above points, I also have to wonder how much of this is tied to marketing. Many people (even type nerds) aren't always aware of the many amazing typefaces currently on the market.

Offshoot discussion: Personally, if I owned a cell phone, it would be nice if it was designed by Erik van Blokland, or Frantisek Storm, or Jonathan Hoefler. If I knew how to use a 3D graphics program myself I'd design a spec cell phone that looked like it was an antique from the Victorian era, complete with brass, wood, (faux) ivory and good, solid, serifed numbers.

> technology is not built for serifs.

In that Technology is Simplification's boyfriend, yes. More significantly, current mainstream text-display technology (even on a full-fledged PC) is not conducive to *immersive* reading, which does make sans fonts a better choice, overall. On the other hand, some carefully-set Georgia displayed in moderately high resolution is much more readable than any sans. And we're still waiting for the next level: hand-crafted anti-aliasing (where for example serifs could be set in a medium gray).

hhp

PK, your comment about technology not being kind to serifs is wrong headed. The issue is the type of technology. Serifs came into being in the first place because of the technology in use at the time.

Clearly low-res screens aren't much good for serifs, but high-res is neutral.

Either way, the natural form for type is the sans, the serif was just a temporary lapse of taste.

Brian Morgan, quoted by John Hudson, has it completely wrong - it's a couple of good sans and several dozen serif faces you need to get anywhere.

> the natural form for type is the sans

If it's worth talking about what's "natural", then the natural form of type is -of course- what's easiest to read (in quantity), and that -of course- is the serif form.

> it's a couple of good sans and several dozen serif faces you need to get anywhere.

It sounds like the place you're going is not a popular destination... Nothing wrong with that, as long as you don't want to drag everybody else there.

hhp

and that -of course- is the serif form.

oh, of course. are we going to trot out this tired old legibility debate again?

Do you mean legibility, or readability? Are you lumping them together to better get rid of them? Ignore them (and their difference) at your own discrection, but I personally grow tired by lilly-livered "artiste"s thinking they can create a text face through divine inspiration.

hhp

Are you lumping them together to better get rid of them? Ignore them (and their difference) at your own discrection, but I personally grow tired by lilly-livered "artiste"s thinking they can create a text face through divine inspiration.

hrant, what are you talking about? nobody's lumping together or getting rid of anything, and that generalization at the end makes no sense. we were talking about declining sales and possibly usage...not legibility or any artistry required to create said legibility (or readability, or whatever you're getting at).

If it's worth talking about what's "natural", then the natural form of type is -of course- what's easiest to read (in quantity), and that -of course- is the serif form.

for you, not necessarily everyone. this argument will go nowhere; everyone reads different faces with differing levels of ease.

Hrant: Would you mind explaining to the clueless and lilly-livered why you insist on arguing from the perspective of text faces? I see no such scope limitation in the original posting. Yes, the natural form for a face is whatever is easiest to read in quantity. For a text/body face. For a face meant purely for display, it's not nearly as important, and possibly an afterthought if the design is intended to be evocative rather than obvious.

You're coming across like fundamentalist foaming at the mouth and quoting scripture out of context in response to arguments nobody made in the first place

> we were talking about declining sales and possibly usage...not legibility or any artistry

They're related. Everything is.

But if I've branched too far, to the point of harming our focus, then I apologize.

> everyone reads different faces with differing levels of ease.

Only to a limited extent. Neither readability nor legibility are dependent only on familiarity. That stance is a pretense, serving to excuse people from the necessity of analytical thought.

The human immersive reading system prefers serifs. While what might be called the "contemplative" system prefers sans. This does *not* mean you *have* to choose a serif for text - that depends on many factors. But ignoring the relevance of these distinctions makes you less of a designers, and much less of a typogapher.

> why you insist on arguing from the perspective of text faces?

Because it is the higher form of type. Display design (which is admittedly more profitable, and certainly more celebratory) is a shadow of type's true being. Amen. :-)

> You're coming across like fundamentalist foaming at the mouth

Yes, I have that problem.

Angry, I am. Because I think people are doing damage to the craft by promoting "Art" too blindly.

Fundamentalist, only in my moderation. Example: the famous "We Read Best What we Read Most" mantra. I'm not saying it should be thrown out. But I *am* saying that:

1. It is generally used (and probably was carefully devised) to relieve people of the necessity to consider the true nature of legibility/readbility.

2. It needs to be: "We Read Better What We Read More Often". Huge difference. Unfortunately, such a revised version is too soft -too subtle, too introspective- to sell fonts to college kids, and to make oneself feel better about making illegible consumables, which is why the original version is preferred, by the *real* fundamentalists.

Fundamentalists like things to be simple: "Legibility doesn't matter, just ignore it." Puritan hogwash. Humans are extremely complex, and *thinking* is not a waste of time.

hhp

> we were talking about declining sales and possibly usage...not legibility or any artistry

They're related. Everything is.

in which case, let's talk about avocados.

But if I've branched too far, to the point of harming our focus, then I apologize.

actually, hrant, you've removed any hope of focus. now we're simply talking about what you think with blatant disregard to any form of structured debate.

> everyone reads different faces with differing levels of ease.

That stance is a pretense, serving to excuse people from the necessity of analytical thought.

and so is the following silly piece of pedantry:

Because it is the higher form of type.

...which comes from absolutely nowhere and is an utterly unsupportable statement. i'm not going to argue your opinion, hrant. i don't care enough. but i deeply resent you blowing up a somewhat productive thread to focus it on your own opinion with no supporting evidence of any sort to your points. see you in the next thread.

i'm sorry to see PK & SU tear down the comments by Hrant which were both factual correct and to the point

with what seems to belong in a preschool courtyard

And to comment on the actual post now:

I think the decline in sales of serif typefaces has to do with several factors

1: Sans to Serif count in a designers collection stems from what type of work h is involved in - most graphic designers these days are lap dogs of advertising

(generaly using more display types than setting long books)

2: "Stupidity" - or "boom of the webdesigner" - a large mass of uneducated designers who have little knowledge of any typography teachings - and much less knowledge on serif faces

3: Enter the fashion plague - design as spectacle (related to point 2) but it's also a larger more general (and unfortunate) sweep of things getting ever more style focussed - people today often think graphic design is about getting attention - which is only a tiny part of what graphic design really can be - and getting attention is not the serif way

I will 'Amen' Peekay's argument. Hrant likes to rant. However, where would the 'fun' of this site be without it? ;)

Speaking of avocados, I couldn't use the one I bought today - it was rotten. Needless to say, my pita was unhappy.

There are some valid points in almost every post here - but none of them strike me as really getting to the answer of why sans is currently in vogue. I don't think there's anything inherently wrong with the trend as a similar sort of bias might well have been happening in the late sixties/early seventies, for example - with decorative font designs proliferating as a result of technological advancements - but serifed type survived that. Hrant is right - notwithstanding his overly defensive stance - that serifs aid legibility and readability - in general. And it strikes me as bemusing and a little ironic that Xander's site, for one example, uses fairly unreadable small sans type on his site, while he seems to defend Hrant's position on the sans fad.

> now we're simply talking about what you think with blatant disregard to any form of structured debate.

The goal -and merit- of structured debate in a *blog* can be argued...

But in any case I myself will naturally talk about what *I* think. If you think that's bad, but you continue to focus on what's wrong with what I think, then you're perpetuating the [supposed] problem! :-/ I'm not being fascetious in saying: feel free to ignore me.

----

> none of them strike me as really getting to the answer of why sans is currently in vogue.

Isn't it probably a result of wanting Control - an ideal that's been central to Western ideology for as long as anybody can remember? And Simplification is a tool of Control.

> I don't think there's anything inherently wrong with the trend

I personally think there's everything *right* with the trend, in the realm of *display* type.

But due to ignorance of the true nature of readablity (an ignorance stemming from deficient education - both autodidactic and formal), the sans "fashion" is intruding into text design, hence functionality - and that sucks.

hhp

::Attempting to wade through all of the ad hominem attacks and counter-attacks::

A lot of excellent points have been made on all of the previous posts, but I think that perhaps the biggest reason for the sans-serif trend hasn't yet been touched.

More than ever, we live in a word where anyone with a computer seems to consider him- or herself a designer (or "desktop publisher"). This means that as soon as all of these pseudo-designers discover that there are font choices beyond the Ariel / Charcoal / ComicSans / Courier / Gadget / Georgia / Impact / Palatino / Techno / Textile / Times / Trebuchet / Verdana / God-help-us menu, many of them rush to snatch them up. Of course, they like best what they notice most, which tends to be display faces rather than text faces. ("What? You mean all serif fonts DON'T look alike?" they might ask).

I suspect that this probably explains the sales popularity of sans (and script) faces better than any theoretical discussion about the future of serifs. (I am reminded of the "End of Print" prophecy.)

For those interested in good, substantive design, the serif will remain the text style of choice. For the toothless masses that like to set entire magazines in Arial, they'll continue clamoring for Optima and the like.

(Perhaps we should be grateful that they're looking for alternatives to Arial at all...? ;-) )

David Thometz

> but I think that perhaps the biggest reason for the sans-serif trend hasn't yet been touched.

Whoops -- apologies. Of course, the last few posts DID touch on it. So I touched it some more.... ;-)

David Thometz

hehe funny seeing my name with an x ..

martin the "illegible sans" is frutiger...

if you had real trouble reading the few words - i recommend ... glasses

there are under 70 words on the page .. it survives .. be it legible or not - legibility is not prime in every piece of graphic design - or trained monkeys would earn less than they do now doing it.

Hrant:

Where is this study that shows that serif faces are undoubtedly more readable than sans?

Most studies I've seen (and admittedly, there are few) the concensus is that people find what they read the most the most 'readable'. So, if you are used to reading old german blackletter, you would say that THAT would be the most readable type style.

I've also heard that while the argument that serifs help create a visual 'line', proper letter and line spacing can do infinitely more to make type more readable than deciding on serifs vs. sans.

As for Serif selling less than sans, I would agree that most people, including trained graphic designers, don't see a broad range of styles in serif faces. I use maybe 3-4 serifed faces at most compared to 20 or so sans faces (though I can gladly get by with only a full weight set of Univers anyday!)

And as to the comments that technology isn't serif-friendly, it's a much simpler correlation than that. Low-res displays aren't serif friendly. Whether that be your monitor, the chalk menu board at the deli, or the LED alarm clock on your nightstand.

> Where is this study that shows that serif faces are undoubtedly more readable than sans?

Don't look for one concluding paragraph in one study: it doesn't exist, and probably never will, certainly not in this materialistic age.

What you need to do, if you feel inclined, and/or if you don't trust me enough, is to read all the studies you can get your hands on, analyze them, consider the thoughts of smart people, converging in your mind the overall reading model, discarding the "conclusions" that seems errant, and finally arriving at something you can live with (if not die by). I've been doing just all this for four years, and when I started I had no idea what I would find. Now, I agree with Ovink's blunt statement: "No one can seriously dispute the important function of the serifs in the constitution of the word-gestalt." But of course I've arrived at some of my own conclusions too.

The problem with taking a single (or even a collection) of studies too literally is that every single one of them contains a flaw that makes it imperfect. Some of them assume that we read in a smooth flow (*100* years after Javal proved that we don't), some of the try to compare vastly different fonts like Times and Helvetica, and hope to extract a naive rule like "Serifs are better" from their numbers. You have to respect their work, but do your own thinking, between the lines, so to speak.

> Most studies I've seen (and admittedly, there are few) the concensus is that people find what they read the most the most 'readable'.

There are many many studies. But I have encoutered not a single one that measures the effect of familiarity.

The only remoteley related thing is a small aspect of Tinker & Patterson's landmark work, where they cursorily used a comparison between the familiarity effect of two fonts (a roman and a blackletter), but merely as a check to ensure the validity of the main test. In fact, the implications (although hazy) of that one small test were very interesting: familiarity *seemed* to come within a matter of minutes! Which of course implies that familiarity with a *style* might not be very important - what might in fact count is familiarity with the general boumas (word shapes) "library".

Maybe the champions of familiarity should actually organize some testing to see how it really works? But no! That would mean something besides their great artistic talents has importance! Why would they want to explore a possibility that might reduce your worth?...

> if you are used to reading old german blackletter, you would say that THAT would be the most readable type style.

No. Fraktur is inherently more readable than Roman:

http://www.themicrofoundry.com/ss_fraktur1.html

> proper letter and line spacing can do infinitely more to make type more readable than deciding on serifs vs. sans.

Everything works together. Sure, if you choose a too-narrow linegap, your font choice becomes irrelevant. Readability is a very delicate thing - everything has to be just right. But why would you choose a lousy linegap? Assuming a good linegap, the font choice *does* become relevant.

hhp

I was struck my Paul Stiff's comments on 'evidence based typography' at the recent conference in Thessaloniki. After many years of involvement in legibility and reading studies, he seemed to be saying that there is little that can be positively concluded, but much that can be negatively inferred in the sense of 'Don't do this, because the evidence suggests that people find it harder to read'. In other words, the value of empirical studies is in the evidence, not in the conclusions, because the evidence can be grounds for deciding how not to do typography.

Paul and I had an interesting talk about reading studies -- i.e. not readability of typefaces, but studying the way people read documents. Paul suggested that one interesting area for study is the moment at which a reader decides whether to refer to some extra-textual element indicated in the text, e.g. a graphc, figure, foot- or endnote, or continute reading. Can typography do anything to help the reader make the 'right' decision? Fascinating subject, and a reminder that reading studies are about a lot more than typeface readability. But anyone conducting such studies has to be aware that the most important factor in the decision may be the most difficult to empirically describe: the reader's previous reading experience and knowledge. I always consult notes unless I'm fairly sure that I know what the notes will contain.

At the same conference, someone (it may have been Paul, or perhaps his colleague Mary Dyson) made referece to the 'readability' of samizdat literature in Central and Eastern Europe during the communist period. I don't doubt that any empirical study of readability of 4th or 5th generation gestetnered 'books' would determine that they are extremely difficult to read, but readers in those countries read them, lent them to friends, made new, even worse copies and treasured them. From this one might derive the new aphorism: We read best what we most need to read.

It's the content, stupid.

> reading studies are about a lot more than typeface readability

Generally, yes. But it's entirely possible to isolate design elements (like "serifity") and test them. It just hasn't been properly done. Is it worth doing? Financially, probably no. Functionally, definitely yes, even though:

Of course I agree that the *author* has the most power/responsability. But shouldn't we look at this by discipline/area_of _control?

1. The typographer needs to know how people read pages (irrespective of content) to create a functional end-result.

2. The type designer needs to know how people read "raw" text (irrespective of the overall typography) in order to create a functional tool (for the typographer).

Sometimes the various disciplines are united in one person (or highly coordinated among different people), and then you can do some interesting things, but that's the exception.

----

BTW, I missed Dyson's talk (I could only make some of the talks on the day that I myself was on), and I *really* wanted to catch it. If you have the time/inclination, could you please provide some elaboration?

hhp

I also missed Mary's talk (I was busy trying to get my own presentation ready), but I spent some time chatting with her, Paul and Ole Lund about readability stuff, and I can't remember who made the comments about samizdat. It might, indeed, have been someone else.

I have to question your notion of 'raw' text. Is it really possible to consider text 'irrespective of overall typography'? It seems to me that as soon as you display text in a typeface, you have typography: you have decisions about how widely spaced words are, how much interline space there is, what size the text is.

If a typeface is to be a 'functional tool' for the typographer, I don't think he can expect it to be universally useful: useful irrespective of the use to which he puts it. There are plenty of typefaces out there that are only useful -- only readable, in fact -- in a very narrow range of application. I don't consider this to be to their detriment.

If we think about fitness of form to function, we need to recognise that function is not legibility or readability, but legibility and readability in the particular circumstances of reading. So the function is various, and so must the form be. One might go as far as to say that the value of readability studies is to measure the fitness to function of typefaces in the particular circumstances of, um, readability studies. I don't imagine that the results would be the same for the particular circumstances of reading a book while standing on a London Tube train during rush hour.

Of course, I do believe that there are general findings that might be applied generally and, as Paul Stiff suggested, might give clues as to what not to do. I suspect such findings are of more use to typographers than they are to type designers. I think it is healthy, though, to recognise the limited relevance of the general to what is, after all, a very particular business.

I agree with everything you wrote, and I feel that my central point is reinforced: legibility/readability are complex, but this does not relieve designers from trying to understand it/them as best they can.

Choosing one thing:

> I don't imagine that the results would be the same for the particular circumstances of reading a book while standing on a London Tube train during rush hour.

Great angle. Why does a news face need a larger x-height than a book face? What is the optimal letterspacing for a given setting? These are important questions, ones Familiarity cannot answer.

hhp

I don't think I know any designers who do deliberately avoid understanding what it means to make a typeface legible or readable. I think many of them are not very interested in attempts at empirical study because a) the studies are so limited in what they can tell us, b) none of the studies to date seem to contradict what most designers would consider to be common sense, and c) legibility and readability are of variable importance in the works of individual designers. There are type designers for whom readability is clearly a primary concern in all their work (I think of Gerard Unger, for example), but there are other designers for whom it is a secondary concern, if much of a concern at all, who are interested in exploring other ideas in their work. Obviously the results of such design have a different intent, function, and market. The Emigre dictum which you seem to find so offensive is a statement about the role of readability in a particular approach to type design: an acknowledgement that we can read almost anything with even a minimum amount of word-gestalt, and that this remarkable human faculty for grammatological adaptation allows writing and typography to be an expressive medium as well as a communicative one. Personally, the thing I find most remarkable and inspiring about the whole topic of readability is not the typically minute difference in reading speed and comprehension between typefaces, but the enormous range of forms that we are capable of recognising as letters and words and of reading with so little difficulty. Bloody amazing!

> I don't think I know any designers who do deliberately avoid understanding what it means to make a typeface legible or readable.

You'd be surprised - even some of the top names in the field don't bother looking beyond the surface, relying instead on the mimicry of precedent to ensure the good performance of their results. That's not a human understanding what he's doing - that's a monkey dancing for coins. You can't make progress with such an attitude - all you can do is make money to buy junk.

To be fair (?), maybe few of the people I'm thinking of actually practice what they preach, but they certainly claim loudly enough that "legibility doesn't matter, it's what you read most" - and that sounds pretty clear to me! Are they bullshiting to make their lives easier, and sell more illegible fonts? Maybe...

John, you're giving them too much credit - there is no "context" for very many of these people. When you say "this font would be more readable with less letterspacing" and they come back with crap like "the pre-WWII Germans read blackletter just fine, so get off my case" then there *is* something wrong in their heads.

You can't make a real text face while worshiping at the altar of Familiarity. You have to understand the human reading "hardware" - which is indeed spectacular in its heuristics.

hhp

I think there are quite a lot of 'real text faces' that have been designed by people who necessarily had a limited understanding of human reading 'hardware'. Attempts at empirical studies of reading are a relatively recent phenomenon, and I'm afraid that I'm not ready to condemn the types of Jenson, Garamond, Kis, Caslon, etc. as 'not real text faces'.

Nobody should expect type designers to conduct their own field research into readability - you use what's available. Before Javal there were no (known) empirical studies about type (although there might have been clues about boumas from other fields concerning the preference for decoding silhouettes, for example). And it's very possible that certain people had noticed that we don't read in a smooth flow (simply by looking at somebody's eyes while he's reading!), and this is a central issue.

Javal's findings had a direct effect on type design at that time (about a hundred years ago), for example throught the opportune shortening of the descenders with respect to ascenders. This was because type designers had the humility and insight to see their role as *craftsmen*. There was a resurgence of scientific study of typography in the 1970s and 80s, but by that time the Artists had taken over, and they wanted to hear none of it.

I have little doubt that many of the greats you mention would have loved to know how we really read, but there was nothing there. They did the best with what they had, with marvelous results. You can tell they were serious; they knew how to make smaller sizes larger in body, wider, and looser. Compare their work (and intent) to the hooliganism of the 90s, where all the findings from emprical studies for a hundred years was outright ignored, and the conclusions of typographic scholars such as Ovink sidelined. And it still is! In the realm of Display this actually has advantages, but for Text work (which I maintain is the higher form) it's wasteful, and irresponsible.

hhp

HRANT/John:

Great comments. Lot's of good points there. I especially like the 'monkeys dancing for coins' comment!

Here's something to add to the fire:

From Paul Mijksenaar:

Hierachy of graphic resources

(what the readers think)

1. Order (chronololgy); Time, 3-dimensional

2. Position (page); 2-dimensional

3. Type size

4. Contrast (e.g. bold versus light)

5. Column width

6. Line spacing

7. Alignment

8. Typeface

Which says that, yes, typeface selection is important, but not necessarily as important as most graphic designers think.

But, of course, that's just one study. ;o)

That's a very interesting -and useful- hierarchy. But it only deals with consciousness. Unlike when you ask somebody what he think of a poster for example, immersive reading is essentially subconscious, and a reader simply cannot make explicit judgments. This doesn't mean type choice is important - we just don't know enough (yet) to say.

A great example of the difference between the two "modes" of reading is this: it has been shown (empirically) that readers claim to prefer a larger size of type than they can actually read fastest.

hhp

The above heirarchy, if my memory is correct, was actually determined via observation, not concious questioning. That said, it wasn't necessarily targeted at immersive reading environments (such as a book).

Hrant: Text work (which I maintain is the higher form)

There was a time when I would probably have said something similar, but I've progressed (at least, I hope it's a progression) to the view that this kind of fustian value judgement isn't very helpful. Text typeface design is one thing and display type design is another; they have a fairly small amount of things in common and quite a lot of differences. I don't think we gain anything from saying that one is a 'higher form' than the other.

I was a guest presenter in a graphics class; I showed them a menu I'd designed, and for headings, I'd chosen JSL Ancient (a slightly-antiqued serif for an older feel).

The entire class, as one, went, "Why did you pick a font that's so . . . default?"

Now, JSL Ancient is anything but "default"; it's pretty flamboyant, especially its italic. But they couldn't tell it apart from TNR.

-.-

I was kind of curious about what they'd been taught: apparently nothing.