At 21.57 +0200 2001-05-14, Javier Bezos wrote:
>> For the base fonts I agree this makes sense, but for top level fonts (i.e.,
>> the ones which TeX/Omega actually works with) it is mainly a nuisance. It
>What do you mean with "base fonts" and "top level fonts"?
I explained what a top level font is. A base font (a.k.a. a raw font) is a
concept used in discussions of virtual fonts; it means a font which is not
a virtual font. In the present context (and in particular for modern fonts
which contain much more than 256 glyphs) one could also view a reencoding
of the font as a top level font in contrast to the base form the font had
before it was reencoded. The main reason for making this distinction is
that foundries traditionally package their glyphs into fonts more based on
how common they are than on what makes sense for TeX, so you have to create
some intermediate layer to sort things out.
>> also goes against much of the philosophy in LaTeX as you end up with for
>> each character separately specifying what it should look like rather than
>> what it logically is.
>I think we are speaking about different things. When you write a document
>you use the logical value (eg, "fi") and TeX converts it to the
>corresponding glyph (ie, the fi ligature).
>The same should apply to, say, Greek. If I write "barbaros" [well,
>imagine it written in Greek] using the same beta, sometimes I would like to
>see the first one using a differenf glyph from the second one (a medial
>beta, not used currently).
And then I suggest that you do this by selecting a (top level) font in
which the beta has the medial form, not by using a special \medialbeta
command or by requesting that the LICR should incorporate something
equivalent to this.
>That can be done with ocp's -- the file ell.fd
>defines a language property named beta with two possible values (oneform
>and twoform) which specifies how beta will be rendered. That can be done
>with tfm ligatures too, I think, but I cannot change how its rendered from
>my document, except if a create new vf/tfm files for every font.
Provided that both glyphs have slots in the same font, yes, but I cannot
see any reason why they should.
>> Now I don't understand. Are you saying that there is an OCP, but that it
>> never changes anything?
>It changes things, but TeX was built in such a way that only the font
>encoding (ie, the encoding used in the tfm file; ie, the final form after
>every single ocp has been processed) is used when hyphenating
>words. (And this problem has not been solved by Omega.)
This implies that there aren't any OCPs acting on typeset material at
\shipout time, so why didn't you say so in the first place?!