Re: [css3-fonts] subscript/superscript variants

On May 10, 2012, at 4:41 AM, Christoph P=E4per wrote:

> It=92s usually the font developer=92s job to create complex =
substitution tables for that (although maybe for =91calt=92 or the like =
rather than =91subs=92 and =91sups=92), but I don=92t think it=92s =
totally unreasonable for CSS to prescribe some on the higher level =96 =
or introduce something akin =91@text-transform=92.

In the early days of our OpenType layout development, we did exactly =
this in features like 'sups'. We later decided that it was too complex =
and too subjective to try to build into a font. One factor was that we =
would have to know about all languages and their standards; and we would =
have to confront different user preferences; and emerging customs and =
standards; and so on.

While I won't say there is no good academic argument for trying to =
create complex substitution tables for all that, in practice (at least =
in our experience) it seemed to create more problems than it solved.

-Christopher=