[OS X TeX] Zapfino install progress, ``Peace on Earth'' card

Will Robertson will at mecheng.adelaide.edu.au
Wed Mar 24 01:03:31 CET 2004

William F. Adams wrote:

> On Tuesday, March 23, 2004, at 12:09  AM, Will Robertson wrote:
>> I'm not sure if you make sense---are you saying it's easier in Mac OS 
>> X, or easier "than"? Madly clicking through the character palette 
>> (with glyphs overlapping sometimes) is not my idea of easy!
> Using my setup is easier than Mac OS X 'cause there's no mousing involved.

Good! I agree.

>>> * snip stuff over my head. I'll ask questions later
>>> ...
>>> \usepackage{omegazapfino}
>>> ...
>>> \ZapfinoText{This would be typeset in Zapfino w/ contextual ligatures.}
>>> \ZapfinoRebus{This would have ornaments / pictures for words like 
>>> (-(duck)-) and (-(pen)-)}
>> If you're going to be calling \ZapfinoRebus anyway, why not 
>> automatically replace "duck" and "pen" with their pictures. Let the 
>> writer escape words that are required as text! What do you think?
> I don't think that'd go over very well, and it wouldn't suit my usage, 
> nor anything approaching what I'd call normal.

If you say so. You'd know better than me! I can't actually see many 
situations in which I'd want to many of the ornaments anyway...

>>> For swashes I was thinking something along the lines of:
>>> ----a (that'd be the first level)
>>> +----a (second)
>>> ++----a (third)
>>> +++----a (fourth)
>>> \----a (a swash from the upward direction)
>>> /----a (a swash from the downward)
>>> and reversing everything for forms appropriate to the ends of words 
>>> --- does that sound okay?
>> I'm not 100% sure I follow. If I was writing "america" and I wanted 
>> swashes on both "a"s (ignore the plausability for now), I'd write 
>> ----america---- with appropriate +,\,/ at the very begining and end of 
>> the word and hyphens like +----america----/?
> Yes, that's correct.
>> Four hyphens seems excessive. Aren't there some unused ascii symbols 
>> that might be more space efficient? ~ ^ * come to possible mind.
> Those are already used for other things, and I do want it to stand out 
> somewhat.

Combinations of them aren't however (AFAIK). You're using four hyphens; 
would using two tildes or two asterisks or whatever not be as easy to 
detect? I can just imagine prefering to type ~~america~~ than 

>> How about a method which doesn't break up the word?
>> ...
>> \swash[g,2,g,4]{engage} would make the first g have the second swash, 
>> and the second (and subsequents) have the fourth.
> That's beyond my understanding of Omega Translation Processes --- I 
> suspect it could be done though, and if I do reach an understanding like 
> that, I'll think about it.

I agree it's trickier to implement, and it's better to get the whole 
thing working before refining the interface!

> Doesn't sound that user friendly though --- gotta count letters if 
> there're dupes, get the count right, and transcribe it correctly and you 
> don't know for sure 'til you process the file.... (I'm lousy at math --- 
> why I became a graphic designer

I was responding to your question of having to have to count up to eight 
hyphens before AND after a letter to apply the swash to it. If you find


harder to work out than


then I'm not sure if you understand my method correctly. The two 
previous examples would result in the same output. You don't have to 
count letters, the swash command should do that for you. The number 
refers to the swash choice.

It certainly makes it a hell of a lot clearer to read in running text. 
(Note I'm arguing about the _hypothetical_ benefits and am not fussed 
about your initial implementation.)

>> But I like not having to break the word. This method would also work 
>> whereever the letter is in the word, so you get the same command in 
>> all cases.
> That is an advantage.

I thought so :)

>> You could even leave out the arguments and randomly swash everything 
>> (\swash{cogent}). It'd be nice to be able to define swashes in words 
>> in the preamble just like custom hyphenations, too.
> I've considered that, but I think you need a certain measure of 
> repeatability to preclude user frustration (::aah!:: it was perfect, but 
> I fixed a typo and now it's not!)

That's hilarious. Maybe when it's all done you could implement a 
\randomswash command on the assumption it's NOT repeatable.

I'm not down with the technical details of program TeX/LaTeX/Omega, or 
I'd write the \swash[]{} command for you. (More's the pity.) Perhaps if 
the syntax was changed to \swash{g2,g4}{engage} it would make more 
sense. I would follow some sort of algorithm like:

read first argument (g2)
=> make all subsequent g letters use 2nd swash
read first letter (e)
=> since e is not g, no change (e remains e)
read second letter (n)
=> since n is not g, no change (n remains n)
read third letter (g)
=> argument matches letter! use 2nd swash, so replace g by (--(g)--)
read the next argument (g4)
=> make all subsequent g letters use 4th swash
read fourth letter (a)
=> since a is not g, no change (a remains a)
read fifth letter (g)
=> argument matches letter! use 4th swash, so replace g by (----(g)----)
read sixth letter (e)
=> e not g, no change
no more letters => finish

So maybe the best thing to do is implement it as you suggest and later 
on add the \swash command to make things easier for everyone. It's 
certainly something that can be tacked on the end!

Please see <http://www.esm.psu.edu/mac-tex/> for list
guidelines, information, and LaTeX/TeX resources.

More information about the macostex-archives mailing list