From: Mark E. Shoulson (mark@kli.org)
Date: Wed Jun 01 2005 - 21:34:57 CDT
Hans Aberg wrote:
> At 20:14 +0300 2005/06/01, Jukka K. Korpela wrote:
>
>> The use of optional ligatures cannot be reasonably
>> decided on algorithmic grounds alone. Whether you use a ligature for
>> "fi"
>> or for "st" is a stylistic choice.
>
>
> I think that historically, the choice has in part been dictated so
> certain common combinations within each language, as well by the fact
> that keeping a large number of glyphs is expensive, and possibly
> cumbersome in printing.
>
Historically, the choice of ligating "fi" has been dictated by the fact
that otherwise the overhang of the "f" forces the "i" sort to be too far
away. The main reason for the f- ligatures is actually the printing
problem of the shape of the "f" (and the long s, for that matter).
Other ligatures were more handwriting-dependent.
Donald Knuth and TeX did many things, but solidifying the fi ffi etc.
ligatures isn't among them. Typefounders already had made those the
foremost ligatures in ordinary Humanist-based typesetting centuries earlier.
>> If you say that ligatures would not be
>> needed at character level at all, you are saying that typographic
>> styling
>> must be handled elsewhere.
>
>
> Right. It could be handled by a separate rendering character font,
> which could very large, just as the Unicode set. Or, as I pointed out,
> they could be added within the character set, possibly as a special
> type of Unicode characters. One can also think of developing smart
> fonts, that can compute ligatures automatically.
>
Ideally one would pick the option among these choices that doesn't
actually break the basic models of Unicode. A "separate rendering
character font" is already available, by the name of Code2000, thanks
for James Kass. It isn't the only option out there, of course. (I
should add more characters to my Marin font...)
>> While that's surely a possible view, and shared
>> by many, it is far from self-evident.
>
>
> There are several possibilities, and one needs to think carefully
> before choosing one model over another.
>
See above.
>> It can well be argued that in English, the letter combination
>> "ae" (in words of Greek or Latin origin) can be written as a ligature
>> with no change in meaning, as a purely stylistic matter. On the other
>> hand, in some languages, such a "ligature" is definitely a character on
>> its own.
>
>
> One should note that every ligature can be given a semantic use,
> namely by quoting it directly, like in the sentence 'An example of a
> ligature is "Þ" [ligature fi]'. Perhaps textbooks in Arabic want to
> name those ligatures and different letter representations explicitly.
> Many glyphs can thus made into semantic objects, by simply
> objectifying them.
>
That isn't a use, that's a mention. And it doesn't count. Yes, it's
important to be able to handle such abuses of orthography for the few
cases in which they occur, hence tricks with joiners and non-joiners and
whatnot, but it doesn't elevate these symbols to the status of
"character." Such semantics attach due to regular widespread use;
mentions like this are red herrings.
The-Artist-Formerly-Known-As-Prince attached a semantic use to his O(+->
symbol, much much more convincingly than your example, and it isn't a
character either. I can attach meanings to any random squiggles you'd
care to envision with your sentence; that doesn't make them characters.
~mark
This archive was generated by hypermail 2.1.5 : Wed Jun 01 2005 - 21:36:21 CDT