Typefaces made for legibility...

svenni's picture

I'm currently working on designing for a safety product at school, and one of the challenges we have is to make sure the instruction text on the product is as legible as possible.
So i've set out to find a typeface that was designed with super legibility in mind, or typefaces that have excelled in legibility testing...
google searches and casual asking around school have gotten me nowhere, so i was wondering if you guys could offer some pointers?

thanks...
Svenni

J.Montalbano's picture

There is always ClearviewOne for text and display

http://www.terminaldesign.com/retail_specimen.php?ID=1001&viewFam=1

and ClearviewHwy for roadsigns:

http://www.terminaldesign.com/retail_specimen.php?ID=1129&viewFam=1

ANd although not yet posted I have a set of 20 ClearviewADA fonts that meet all requirements of ADA (Americans with Disabilities Act) for interior signage.

James

-I guess those banner ads aren't working

svenni's picture

"-I guess those banner ads aren't working"

well, they are, actually. your typeface was the first thing i considered, but 200$ is a bit more than i can spend on a school project :/

It's a beautiful typeface though!

J.Montalbano's picture

What is the school project. You say a safety product, could you provide more details?

Nick Shinn's picture

I don't suspect that in your circumstances legibility would have too much to do with choice of typeface.

Legibility only becomes an issue in extremely adverse reading conditions, such as highway signage, very small packaging information, or bad printing conditions such as newsprint.

Otherwise, it's like saying which plays a clearer tune, the piano or the guitar?

hrant's picture

> Legibility only becomes an issue in extremely adverse reading conditions

This does include imperfect eyesight as well, though - not a very uncommon issue.

And if you factor in reading duration (probably not an issue here however), the smallest typographic features become highly relevant.

hhp

Nick Shinn's picture

Hey, stop blaming the font!
What would be of more use for legibility would be a "shizzolator" that rewrote text to avoid visually ambiguous/difficult words.

For instance, Clearview has the commonplace "modern/modem" problem with the "r-n" combination. A "legibilizer" would replace words containing "r-n" with alternates, eg "contemporary" or "up-to-date" for "modern". It would also vary the length of adjacent words, and prefer words with extenders over those that don't. That would be bouma-bastic, Hrant, eh?





hrant's picture

Hey, stop believing the WRBWWRM hogwash!

But the divergence mechanism you describe would indeed be very effective, although computationally very heavy: you wouldn't just replace "rn" combos with something that looks less like an "m"* (you could do that just with OpenType), you'd need some serious code: first build a "conflict network" for the entire lexicon of a language (using "confusion values" assigned to different letter sequences), and then pairwise-diverge all boumas beyond a certain threshold.

* Like maybe use an original-Cheltenham-style ascending "r".

hhp

meredithalix's picture

WRBWWRM?

Nick Shinn's picture

>* Like maybe use an original-Cheltenham-style ascending "r".

Not the first choice.

1) a more pronounced, longer ear that curls down
2) a wider "r", so that the space between the three vertical stems in "rn" is uneven

beejay's picture

>> WRBWWRM?

We read best what we read most....

from Rudy and Zuzana

Dan Weaver's picture

In Manhattan here they replaced the crossing signs that use to say walk, don't walk with a big red hand and a symbol of a person walking. Maybe the safety instructions could be communicated easier using symbols. I think they made the sign change here because of the increasing amount of non english speaking who live here.

J.Montalbano's picture

Not just Manhattan Daniel. Those signs have made it all the way over to Brooklyn, as well.

hrant's picture

> a more pronounced, longer ear that curls down

The danger there of course is that you're moving closer to an "n".
The x-height region of the Latin alphabet is already over-used and closed in on itself.

Plus you're making the "r" wider (see #1 below).

> a wider "r", so that the space between the three vertical stems in "rn" is uneven

And there you have three issues:
1) Word width is a key determinant of bouma resolution, and having divergence in letter widths (as in narrow "r" versus wide "n") helps a lot.
2) You could just use a kerning pair to achieve a similar (if lower quality) effect.
3) The gap at the lower-right of the "r" (especially in a sans design) might get too close to a word space.

The great thing about an ascending "r" is that the space between the x-line and the ascender height is very poorly used in the Latin alphabet, and having a modestly rising beak on the "r" would diverge boumas wonderfully. In the same way, it would be great of the "e" had a modest descender.

hhp

hugocristo's picture

> We read best what we read most....
> from Rudy and Zuzana

Eric Gill said that almost 60 years before them.
And he wasn't talking about desconstruction or
post-structuralism at all :-)

Dan Weaver's picture

James its good to Brooklyn, maybe its a city wide project. I think those signs are super clear,bright and no words.

J.Montalbano's picture

As opposed to an ascending r, there are quite a lot of examples of a decending r in Caroline scripts (as well as decending x). It actually works quite well, but getting the highway engineers to go for it would have taken another 500 years.

Nick Shinn's picture

modem

The big eared, wider "r" doesn't get confused with "n" -- see top examples.

However, it does introduce a big negative space in combinations such as r-t and r-v. So if you put it into a typeface, then other characters also have to be given extra negative space, so it doesn't stand out too much. That's what I did in Preface, with the wide, open-sided "a" and "t," and "i" with serif.

A consequence of more open-sided characters is that the space between words becomes less distinct, which may or may not be a problem.

hrant's picture

> The big eared, wider "r" doesn't get confused with "n"

When you have many seconds to deliberate, of course not.
But in actual reading, a difference of half of a stem (not to mention the less relevant lower half) gets very easily lost, especially when it's in the x-height region.

If you confuse deliberation with immersion, then notCaslon could be a superb book face!

hhp

sham's picture

Svenni, I'd also humbly suggest you consider legibilty through the entire process. The legibilty of the font has to be balanced with other things like the quality of the paper and dot gain (if we're talking about printed instructions). The most legible font could be foiled by dropping the ball later in the process. I had some very smart safety instructions for a pleasurecraft that were laminated since it's likely you'd need to refer to them while on the water.

John Hudson's picture

But in actual reading, a difference of half of a stem (not to mention the less relevant lower half) gets very easily lost, especially when it's in the x-height region.

Prove it.

It seems to me that determining an optimal norm for the shape of a lowercase r is one of those rare things that could actually be relibaly informed by empirical study. Most reading studies try to determine too much, but if one focused solely on problems of word recognition involving different forms of the letter r, one might actually be able to draw a useful conclusion.

hrant's picture

> Prove it.

Pay me to. :-)

> determining an optimal norm for the shape of a lowercase r is one of those
> rare things that could actually be relibaly informed by empirical study.

Indeed. Although the danger is that the more focused a study the more it relies on assumptions in order to extrapolate to the big picture. Field research that's broad is harder to extract practical usefulness from, but it gives bigger clues. I guess it's a matter of gradually improving focus, in terms of what level of specificity we really need most at the moment.

Anyway, more importantly:
Lacking empirical proof, we shouldn't just wring our hands and pray to dark gods. We should use what we do know, for example that the lower right half of an "r" seen in the middle of a word in the parafovea is nearly irrelevant, come on.

hhp

rs_donsata's picture

This is turning interesting, James why do you
consider Clearview as very readable?

I have retarded the writing of my thesis project
because of some academic issues. I am planning to
do it about Spanish Linguistics for Type Design
Improvement, mainly on the readability area for
long texts.

I have been doing some light reading on related
areas and would like to confirm my understanding
of the topic (raised mainly in suppositions) and
ask for some guidance.

Ok, as far as I understand and parting from the
Javal fndings* reading efficiency is related
mainly to the hability of the reader to predict
the sense of the text as this determines where
are the eyes going to be fixated on the next
saccades for the sake of revealing information
for comprehension. This implies a knowledge of
the language, of it

rs_donsata's picture

Hugo I copied the quote from these forums:

"Legibility, in practice, amounts simply to what one is accustomed to." - Eric Gill - Essay on Typography, p.44

Not that I fully agree with this, but it

hrant's picture

I for one think you're on the right track.

> the way a type design can improve readability is aiding the eye navigation
> trough the text by facilitating the discrimination of the shapes (bouma?) of
> the key elements for its decypheration based on the syntax and morphology
> of the language.

Yes, but at the type design level (as opposed to writing custom software to dynamically optimize live text or something), you still have to address the individual letterforms. What you can do is map the entire lexicon of Spanish and find the sequences of letters that tend to get confused most, and diverge their constituent letters in key ways. This is not an easy task. For one thing, it requires grammatical sensitivity to work really well.

BTW, don't focus on Javal's specifics too much: he was a stellar pioneer, but for that reason it was too early for him to be very exact about the details of readability.

--

BTW, one can't doubt that Clearview works very well for the intended purpose: highway signage. But to me "readability" refers most to long passages of text (like a book), and just the fact that Clearview is a sans would make it not extremely readable, even though I think it's a highly legible, handsome design.

hhp

gerald_giampa's picture

Readability is greatly overestimated.

Such trivia is so easily dismissed simply by viewing said text through tumblers of wine. If you still can't read it, pour another glass and make something up. You have two distinct choices, drink from them, or wear them. Glasses I mean. That's what happened to me.

What a great reminder of why I always circumambulate squandering moments with tattooed Manhattoes and bearded typographers wearing skirts during my Pythagorean transmigration of the universe.

Friends, think harmonics and you think like the gods.

One reads best what one wants to read most. For me "ugly" would not one of them. But be my guest.

If you are so concerned about the r m confusion buy this. . .

http://www.lanstontype.com/Bodoni26.html

You will never be confused again. But If you are, you just know who to ask.

Hugo,

What book?



Dan Weaver's picture

If
I
set
any
type
like
this.
Ordidtheoppositeandsettypelikethis.
Would it matter what the letterforms looked like. The use of typography for instructions is less important than the writing of the instructions.

"DANGER" Taking a shower with you're hair dryer on and in the tub will make you feel really, really bad.

timd's picture

Keep the measure short, increase the leading
and many typefaces would be acceptable,
it is not just the font which determines legibility.

hrant's picture

> The use of typography for instructions is less important than the writing of the instructions.

But it's a matter of context: We don't write the content, we make the fonts, and use them - so our concern is that "micro" level.

Feeding the hungry is more important than entertaining the rich, does that mean we should stop doing graphic design? Well, maybe.

hhp

aquatoad's picture

One reads best what one *wants* to read most.
Instruction manuals get read when the thing won't work. (Of course, it helps if it isn

svenni's picture

it seems like i kind of missed my own party :|...

we are designing and marketing a new kind of fire extinguisher.

the legibility is needed because people need to be able to read the user guide on the device as quick as possible, in badly lit, possibly smoky conditions.

sham's picture

The instructions should be very simply written, possibly with informative illustatrations for use. I hope a fire extinguisher has a small set of instructions.

I mean this only half-jokingly but you could add a small LED to light up the instructions when you remove the safety pin.

aquatoad's picture

Hi Sveinbj

John Hudson's picture

I should think focusing your attention on the instructive illustrations is priority number one.

I'd be burned to death trying to interpret the illustrations. Diagrams and illustrations are necessary because not everyone can read, or read the language in question, but I really can't believe that getting information from illustrations is faster than from clearly worded text. I think the world's best fire extinguisher would have brightly coloured parts and instructions that read:

Remove green pin.
Point yellow nozzle at fire.
Squeeze blue handle.

hrant's picture

Or just "Remove green pin to use".
I think only Patrick (Spongebob's friend) would have trouble with that.
Otherwise we'd need "Do not stab self" emblazoned on kitchen knives.

hhp

Nick Shinn's picture

OK, I'll get back to the original topic.

A monospaced typeface would be better for a short instructional phrase.

That would remove the ambiguity that occurs in combinations like "rn=m" and "cl=d," especially in faces like Myriad, which is not a very "legible" typeface.
dose

aquatoad's picture

Hmm. I always thought myriad pretty legible. In this test, it

hrant's picture

> why is most highway signage sanserif

Part of it is cultural, but there's also the difference between legibility and readability, in terms of the reader having time to decipher a short word in the fovea (by compiling the letters, which have to be highly decipherable in adverse conditions of weather and lighting) versus immersive reading. Sans forms are more decipherable during deliberation. And note for example that highway signage is loosely spaced, while that's anathema in a book.

BTW, one practical consideration for setting text on a fire extinguisher is that it's on a narrow cylinder, so you want a narrow font (unless you set the text along the length, which might however momentality confuse people), since you don't please having to twist the thing back-and-forth to read lines.

hhp

J.Montalbano's picture

> why is most highway signage sanserif

The simpler and more realistic answer to this question is that, at least in the US, highway signage comes down from both sign painting and an engineering template lettering. Sans Serif glyphs are easier to paint, and to render via a template.. In the case of Standard Highway Alphabet Series B,C,D,E, Emod, these were derived from what is essentially Leroy lettering, with some of the round terminals squared off.

ClearviewHwy is a sans because it would have been impossible to get highway engineers to consider a serif, hell, it was almost impossible to get them to consider ClearviewHwy at all.

Nick Shinn's picture

What we really need is type that knows when it's being read, and by whom, and can respond accordingly. If the lights are out, it will even talk to you.

We're a little bit along the way with the preferences users apply to web pages (as well as Text Zoom).

LettError's Twin font, with the selection of glyph alternates being database-driven, is another step forward.

Even without a dynamic interface, there are options, and could be more. Shania's latest album was recorded in Country, Pop, and Dance (europe only) mixes, color-coded. Many books come in large-print options. Harry Potter came with alternate adult and child covers, but the guts were the same. On-demand publishing should be able to offer readers a choice of settings. "Would you like that in Bembo or Times, or, for the really adventurous, perhaps a sans serif?"







Dan Weaver's picture

Hrant, most of the fire extingushers I've seen here have a tag attached with the instructions on the tag. As for symbols, English is a second language for most people in Manhattan. I can't travel a block and not hear someone talking in something I don't know. Its just as important for non-english speaking people to understand how to operate the extingusher. If you didn't use graphics you'd have to have the instructions in at least 5 languages. You should see MTA posters that try to explain stuff in 5 languages, its a visual nightmare.

hrant's picture

You have a point about including a visual explanation.
But I don't get how a tag is not a Very Bad Idea...

BTW, Lego manuals have a standard warning text... in 29 languages! In six scripts!
Great source for setting comparative sample text in languages you don't know.

hhp

Dan Weaver's picture

Hrant, its like the tag on a bed its illegal to remove. There is another reason, the tag has the date of when the extingusher was last recharged. It would be bad to have the thing and not have it work because someone didn't get it recharged. I guess the chemicals that create the foam only last so long.

Wow 29 languages

I could see the fire extingusher now "Where is the Sanscript instructions"

hrant's picture

> its illegal to remove.

Oh, OK - that totally makes it tamper-proof! :-)
BTW, I wonder: since mattress tags can be legally removed by the consumer, does that mean fire extinguisher tags can be legally removed by somebody being burned alive? ;-)

Seriously:
Have a tag for the "freshness", sure, but the usage instructions shouldn't be so easily detachable.

hhp

meredithalix's picture

I'd think the best "instructions" on a fire extinguisher would be a big, fluorescent numeral "1" on the pin and "2" on the trigger. Granted, this might be a problem for people whose languages don't use Arabic numerals, but most of the world at least recognizes them.

meredithalix's picture

Hrant, I didn't have you pegged as a Spongebob fan.

hrant's picture

Well, not a really huge one.
Beavis & Butthead though will probably never be matched in my book.

hhp

John Hudson's picture

He said bouma, huhuhuhuh

gerald_giampa's picture

Remove green pin.
Point yellow nozzle at fire.
Squeeze blue handle.


"Are you suggesting we eliminate colour blindness from the gene pool of survivors?"

John: Too much Darwin for you! :-)

hugocristo's picture

Gerald: An Essay on Typography

Syndicate content Syndicate content