Misplaced code

Hilmar Schlegel Hilmar Schlegel" <schlegel@vossnet.de
Thu, 26 Nov 1998 10:38:33 -0500

Thierry Bouche wrote:
>  > (\char0 was usually OK).
>  I believe it is best when one can define the boundarychar by defining
>  which character is to be used for it, independent of its special
>  position in the encoding in question. Therefore no code position but a
>  character name.
> yes, i was implicitly T1, so \char0 is the grave accent--I always
> thought that lonely accents should be good candidates for
> boudarychars, but this is maybe short-minded (as some conventions put
> accents on capital letters not over but beside the capital...) ?

Yes and no. I have in mind that one could access accented characters via
ligatures and allow this way different (actually variable, independent
of the hyphenation patterns) encodings (and provide characters not
covered by Cork). It is necessary however that the hyphenation patterns
support the lonely accents and consequently also that lonely accents do
not represent a word boundary.

There are different suggestions which character never appears midst a
word and is not disturbed by the kerns with <space>. Optimal is
certainly something which is really never used.

> BTW, the idea of using kerning with the space charachters from AFMs
> that do declare them is a good one, be it implemented in a MTX or ETX
> file!

This has of course to go into latin.mtx for example ;-)
It works at least and compensates a bit for the missing mechanism to
have different weights for the strechability of spaces and for
hyphenation points in Tex...

Hilmar Schlegel