> LICR objects are also known as "font encoding specific commands", ie stuff
> that is declared by the font encoding files, eg
> \" \ss \texteuro ...
> since 1996 or so when inputenc was first introduced it was an open question
> whether input chars mapped by inputenc should be "text", "math", "text+math"
Given the mathematical origins of TeX I think that some kind of
math LICR should be included. While it's true that a few years ago
most of fonts contained just a few math symbols, currently with
the advent of Unicode things are changed dramatically, and the
combination of true symbols (instead of macros like \alpha or
\nabla), perhaps combined with a package like nath, can make
formulas much clearer.
I would separate math and text, but it happens that Unicode
does not, iiar, so it seems that the only possibility seems
that you have written. Definitely, kerning must not be killed.
> should such an implementation be incorporated (somehow not this one),
> - as a package,
> - as a standard,
> - not at all ...
I would vote for a package (making it standard in LaTeX3 if