"Angelo Dalli" <[EMAIL PROTECTED]> wrote:
....
> Note that representing these two digraphs is fraught with problems,
> especially due to the context sensitive capitalisation rules. ‘gh’ at the
> start of a word is capitalised as ‘Gh’ while for an all-capitals word it is
> written as ‘GH’. Similarly ‘ie’ is capitalised as ‘Ie’ at the start of a
> word and as ‘IE’ for an all-capitals word.
...
On the big assumption that you are able to convince the powers tht be that
these are unique characters, I don't think you need the two kinds of context
sensitive capitals - this sort of thing is now more easily handled by having
a single capital letter character that may take on two glyph forms dependant
on context. Which form is actually displayed can be handled easily by AAT or
OpenType fonts dependant on context. Other scripts, especially Arabic and
Indic, have many single characters which have multiple context sensitive glyph
forms - and modern font rendering systems such as those in Mac OS/9 and
Windows 2000, originally developed with exactly this sort of thing
in mind, can take care it with ease.
In fact I should think that proposing encoding two forms of capitals could
weaken the case you need to make in order to prove that these are discrete
characters - and encoding two forms surely complicates things like sorting
and searching.
- Chris
