[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: FT2 design question

From: Just van Rossum
Subject: Re: FT2 design question
Date: Sat, 26 Feb 2000 17:45:38 +0100

(Getting more & more off-topic here... Never thought I'd have a typographic
argument on this list ;-)

At 2:50 PM +0000 26-02-2000, Werner LEMBERG wrote:
>> I think (but I'm not 100% sure) that this notion of "correct"
>> ligature replacement only exists in german. I quite positive that in
>> english it's quite alright to set 's' 'h' 'e' 'l' 'ff' 'u' 'l'.
>Ha, this is the very example Don Knuth gives in his TeXbook for not
>using a ligature :-)

Hmm... I have Knuth's "Digital Typography" (1999) here, which contains
TEXDR.AFT, and the only reference I can find is this:

[...] We get ligatures by replacing
      f f    by    <ff>
      f i    by    <fi>
      f l    by    <fl>
     <ff> i  by    <ffi>
     <ff> l  by    <ffl>
Such ligature formation and glue adjustment takes place when the character
boxes are being appended to the current line, before hyphenation etc.; this
means that we lose the chance to break "shuffling" into "shuff-ling", but
so what.

It appears to say quite the opposite in an idirect way... (Although the
"shelfful" example is better.) Do you have an exact quote? Would be the
first time I hear someone outside of Germany claim this is "correct".

(To be honest, I have a lot of respect for Knuth, but more in the field on
CS and *digital* typography, but not so much in typography per se.
Brilliant scientist, lousy type designer, so-so typographer.)

>I don't agree.  It increases readability in German -- since we use
>sooo long words, even subtle details are helpful.

We'll have to agree to disagree... I think it's backwards. Ligatures were
invented to improve shapes that would otherwise collide (the "flag" of the
f against the dot of the i), I fail to see why this "improvement" should be
turned off at syllable boundaries.

>Well, the new German writing rules make it `ba-cken' instead of
>`bak-ken', and I agree that hyphenation after the syllable makes
>sense.  But bac-ken is *never* possible...

I don't know about bac-ken, but c-k *used* to be possible... Glad it isn't
anymore, though.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]