"Michael Everson" wrote at 2004-02-18 02:39: > Kenneth Whistler wrote: > > Alternatively, let's leave all subscripting to markup. > > It is too late for that, Ken. Sorry. Indo-Europeanists have > requirements just as real as Uralicists did. And having some > subscripts available but not the rest isn't acceptable. Why would it > be?
Why couldn't there exist a combining subscript modifier, added after a base character, to make it look like a subscript, and that could be given a maximum combining class 255? This would greatly limit the abuse of this character to write a whole sequence into subscripts when it should be coded more simply with markup in rich-text formats. So for example to encode the proposed "subscript lowercase x", we would code <LATIN SMALL LETTER X> then <COMBINING SUBSCRIPT MODIFIER> The combining modifier would normally create a conjunct or a ligature with its base grapheme cluster to modify it in order to make it look like a subscript; this is unlike the simple combining diacritics that MAY be simply overwriting the base character, but most often need to create a special ligature to handle a correct placement, including the already existing possibility to alter/resize the base character in order to place the diacritic above or under it. So that's something which can be handled in fonts. If we need a representative glyph for the lowerscript modifier, it could be a smaller dotted circle...
