New to Typophile? Accounts are free, and easy to set up.
Create an account
Typophile RSS | More Feeds
Not a Honda rebrand, rather UI fonts for the Palm Pre...
This story actually links to Typophile, so you can get that here:
Oops, sorry I didn't check the news page.
So why is my TV not working today? ;-)
Right the first comment at http://www.engadget.com/2009/06/12/font-bureau-takes-credit-for-palms-cu... -- "the uppercase I and the lowercase l still look the same. Why the hell do people do this with sans-serif fonts?"
Amazing how well propaganda works. Whom do I need to congratulate?
Why the hell, as best I know it:
1. Bell Gothic (Chauncey Griffith) is the first typeface to put serifs on sans cap I, 1937, to avoid confusion between I and 1/l, in low-quality printing of telephone directories.
2. DIN signage face has lower case "l" with tail, 1930s, to avoid confusion between I and l when both are rendered by a monowidth device, such as a technical pen or a router.
2. Orator and OCR-B are 1960s sans serif typewriter faces which include I and l with serifs, this time to "fill up" the empty letterspace in a monowidth system.
3. Meta has "l" with tail, 1985 (?), an homage to DIN?
3. Officina picks up the "office typewriter" look and repeats the Meta "l" tail (same designer, Erik Spiekermann), 1990.
4. Verdana and Trebuchet, 1994, but this time the serifs serve to differentiate I and 1/l when all are hinted to the same pixel stem width and height, and the visual cue of subtly different stem dimensions that exists in print is not operative.
In general, these faces were designed for "output devices" which lacked the precise resolution and sophisticated spacing nuances of good quality letterpress and offset printing.
So type for a hand-held device fits the scenario.
> “the uppercase I and the lowercase l still look the same.
> Why the hell do people do this with sans-serif fonts?”
david, do you have anything to say for yourself?
because this bugs the shit out of a lot of people.
National has "l" with a tail.
Is Johnston's Underground the first sans with a tailed l?
I come from Illinois, and the abbreviations Il or Ill. are often awkward in sans faces. But honestly aside from that, I don't recall ever having a problem when both the I and l are straight. Maybe in other languages... The straight sided also save space, and maybe that demand trumped disambiguation here.
Overshooting the cap height with the l ascender can help too. What is the fuss about?
...because this bugs the shit out of a lot of people.
Not me. I admire his reticence.
Maybe in other languages...
Types have to cater to all kinds of documents, and being used where letters and numbers are mixed, eg codes (postal and otherwise) is a consideration.
Nice time-line from Nick.
Here's his illustration to it, from the related thread How does the serif on a sans-serif i increase legibility/readability?
<-shutting off nick's excessive emphasis...
> Not me. I admire his reticence.
it's not "his reticence" that bugs people.
it's the too-highly-similar uppercase-i and lowercase-l.
(and no, it doesn't help if they're not _exactly_identical_.)
so i just wondered if david had anything to say about it.
Thank God someone sometime designed the U to differentiate from the V and (maybe the same guy/gal) the J to set that vowel apart from the I.
It could take a couple of centuries though to get this sans cap i vs lc l issue solved… : )
Or David could do it next week.
. . .
Bert Vanderveen BNO
It's really unfortunate that the blogger on Engadget used a pre-release mockup of the Pre interface for his post -- which actually does show Avenir, not Prelude! (cf. here)
I think DB is currently traveling, so as semi-official FB apologist, I'll offer some additional information in his place.
David actually did design an alternate I and 1 in Prelude. They're right there in the fonts:
(Defaults above, alternates below.)
I suppose the final usage may have been up to Palm.
Thanks Kent De Apologisto. Also:
>Font Bureau takes credit for Palm's custom 'Prelude' fonts; our hearts melt
I'm sorry the writers expressed it this way, or didn't read our release far enough?
Ascender Corporation was also integrally involved in the project though with was not mentioned in our release.
As for the details of style:
The uppercase I with serifs was presented to the client and remains as an alternate. The lowercase L with a tail was presented as well, and the client dismissed it for stylistic reasons on their own. I did not discourage them as the tailed l.c. single stroke and serif u.c. I character designs are nice until they gather in words.
While some people may believe firmly enough to only supply tailed and seriffed versions of these letters, regardless of the size range of use or distance of the user, I remain skeptical of such an extreme view, preferring apparently now unconventionally that words are more important to readers than letters at nearly any distance and size.
> Ascender Corporation was also integrally involved in the project though with was not mentioned in our release.
Yeah, that would be my fault. I tried to make up for it in some small way here.
And all this because I made a comment about a comment ...
[...] as the tailed l.c. single stroke and serif u.c. I character designs are nice until they gather in words.
Theory, based on precedent, could not have accounted for the success of The Mix.
And by that I mean not just popularity, but successful functioning, i.e. readability.
I think there's a lot more design space available for the judicious positioning of serifs, in both some-serif, semi-serif, and serifed types. Try it, and see what works.
Another thing you can do is loopy "g"s!
you're the expert.
nonetheless, i dissent.
William: But honestly aside from that, I don’t recall ever having a problem when both the I and l are straight.
Fontlab glyph names? Is that Iacute or lacute? In a sans font with a sans ui, I've grabbed the wrong glyph before.
>Fontlab glyph names?
Yes that's another case. Basically, when the letters have context there's generally no problem. Without context there is a problem.
bowerbird: dissent is great, parameterized dissent is best.
I think with the increased use of randomized-character passwords ambiguity has become more of a problem. (On the software registration window do I enter "plK0jl7IQ33" or "plKOj17lQ33")
Quite. I must say that whenever I have to email corrections to our typesetters, the moment any of the changes involves l/I/1, I select the entire email and change it to Georgia.
(If anyone knows how to make Outlook 2003 do rich-text emails in Georgia by default instead of Arial I'd be grateful to learn it!)
Ever since I chose to block pop-ups, my toaster's stopped working.
>If anyone knows how to make Outlook 2003 do rich-text emails in Georgia by default instead of Arial I’d be grateful to learn it!
Maybe.. tools / options / mail format / fonts
Duh. Thanks for that si; should've worked that out for myself. Sorry for drifting off-topic there folks
Stands for drifting topic widget? ;)
> bowerbird: dissent is great, parameterized dissent is best.
well david, i thought the parameter was clear here:
when two letters look highly similar, it's confusing.
Probably not relevant to regular usage, but I'll submit an example in Romanian when I and l (similarly designed) create problems in reading religious texts. It's customary, globally I believe, to capitalize all pronouns that refer to God in such documents. This can confuse the reader in that he/she cannot easily distinguish between I (a pronoun referring to God in this case) and l (a pronoun referring to a human).
My sample uses Myriad.
EDIT: Then again, I have to admit I dislike Verdana's I...
>when two letters look highly similar, it’s confusing.
Yes, I am aware of that effect. Yet... 90% of all sans share a lack of the features of which we speak, and... they've lasted in the market for sometimes a century. Is something else happening too?
...they’ve lasted in the market for sometimes a century. Is something else happening too
A few new things are happening. One is contracted text messaging, in which letters and also numerals are being used in novel ways, such that individual letter differentiation becomes more critical than in word recognition supported by familiarity and context. Another is the proliferation on non-linguistic informational text such as URLs, file names, etc. -- including critical information such as indicators on flight control monitors -- , requiring quick identification of individual letters rather than normal reading.
I don't think there is an a priori problem with styles of lettering that make less differentiation between forms than other styles of lettering. Just choose the style that is appropriate to the functional requirements of the typography.
> Is something else happening too?
yes. the people who design and run things are out-of-touch,
and fail to listen when the dissent bubbles up from the trenches,
so we groundlings must muddle through the mess they've made,
and make to this very day because "we've always done it that way".
for instance, here's what ed colligan, the head of palm, said in 2006,
when he was asked to comment on the upcoming debut of the iphone:
> “We’ve learned and struggled for a few years here figuring out
> how to make a decent phone. PC guys are not going to just
> figure this out. They’re not going to just walk in.”
that seems unbelieveably pigheaded and naive in retrospect, not?
oh well, at least unlike the financial shitheads, you font-makers
are not extracting massive tons of money out of the economy, or
-- once your house of cards has fallen apart -- insisting that you
need to continue receiving your obscenely-huge year-end bonuses.
so, as "experts" go, you're relatively benign... ;+)
p.s. of course, when global-warming chickens come home to roost,
that's gonna make the recent financial collapse look like peanuts...
Smartphone usage is generally one of rushed reading (boss’s top-posted E-mail), rarely longer reading (E-book), very often precise text duplication (passwords, a living hell on my iTouch).
Hence an intelligent system might use straight or tailled els in different circumstances, or at different point sizes.
Eventually somebody is going to mention that, like Segoe by Microsoft, Prelude is merely a clone of Avenir. I mean “mention” as in “level an accusation.”
I was prepared to turn this job down when the VP of engineering sent me an email. He responded that this was not going to be like Windows or the Mac. I was skeptical, so I ticked off a list of capabilities that I felt were required in any model modern device, desktops included, to make me (users) happy. (I ticked off exactly the same list here sometime ago to dead silence). This client though, he said "We can do that... all of it!" The Pre is typographically for real: it uses TT, it scales according to code, it's got a sweet rendering, OT is supported and feature acceptance expanding, 180 dpi!, integer metrics, font fidelity and open.
I hear what's being asked of the some-serif, and will endeavor to please, thank you.
Read on! Hug a Tree!
@Joe: Today’s iPhone OS 3.0 software release solves at least one of your beefs: copy/(cut)/paste has been implemented at last.
>He responded that this was not going to be like Windows or the Mac.
That's exactly what struck me when I first saw the interface (before knowing David and Tom were involved) when looking at videos and static shots like these was that with good interface design, and a high-res display, the need for font tricks like cap I, and ClearType is diminished. Not sure the Pre is there yet, but seems to be headed in the right direction.
PS. What surprised me was lack of sub-pixel positioning in the browser. But that could be a webkit / perf trade-off.
OT is supported and feature acceptance expanding
Could you say more about this? (Or perhaps a link to further information?)
Eventually somebody is going to mention that, like Segoe by Microsoft, Prelude is merely a clone of Avenir.
And then someone else will mention that Avenir is merely a clone of Nobel.
David, thanks for trailblazing the market (again).
How long before Pre users can install any font they want?
But Nobel wasn’t what they were cloning. When your daughter dresses up like Jessica Simpson, she isn’t cloning Eve.
That Starling looks awful familiar, but I can't quite place it. Give me enough times and I'm sure it will come to me. ;-)
Actually the Starling 'ultra', now the featured face, has old style stress, like the regular, whereas Times Bold has 'modern' vertical stress. Of course Mike Parker says its the original, and Times the copy. I wonder what all the differences are...
...like Segoe by Microsoft, Prelude is merely a clone of Avenir.
Let's see if Adrian Frutiger returns his SOTA award.
If he does then David can send him his.
Segoe was found to be too much like Frutiger--not like Avenir--by the EU.
Do Avenir and Frutiger look alike to you, Joe? :)
>Do Avenir and Frutiger look alike to you, Joe? :)
I've seen the secret Linotype roadmap - next year they ship Frutenir - then year after next it's Frutenir Next.
>I’ve seen the secret Linotype roadmap - next year they ship Frutenir - then year after next it’s Frutenir Next.
Is that an announcement? Will Microsoft ship it with their device, the Erp Mlap this year?