/[One consequence of the string policy is that ]/we can no longer
encode new precomposed characters for grapheme clusters that are
already encoded in any existing standard form/[.]/
And you've truncated the end of my sentence
Well, I have not, unless you really want to count that little period.
While we're at it, I have a habit of trying to quote linguistic
constituents whenever it is possible to do so without too much
contortion, and the period has scope over the whole sentence in a way,
so if I omit the first part I want to omit the period as well, but
that's really just me ;-)
Okay. So I had truncated away /supplementary content/
yes technically they could still be encoded [...]
In other words, this duplicate encoding (without canonical
equivalence) is not needed and not desirable.
which is great to know, but still not part of the issue. Maybe it was a
mistake to precompose only {ᾱ,ῑ,ῡ}; or they were part of some
preexisting character set (and then Unicode policy prevented us from
adding more – a policy whose content I'm not exploring and whose sense
I'm not debating). It will also depend on the reason for precomposing
Greek in the first place. The question remains: If someone was thinking
far enough to be wanting to have these in some set, that person should
have known they're likely gonna be used in an environment where
additional diacritics may be added. Pitch accent depends on vowel
length, so it's unlikely someone would want these three characters only
in an otherwise diacritic-free string or only for say parenthetical
comment after a word or phrase. Here's another idea: perhaps a keyboard
layout didn't want to bother with a macron dead key but already provided
dead keys for the breathings and accent marks (but that'd waste 2 keys)
– and an encoding reflected that. But this is all speculation. It's a
little oddity, and I'm asking whether someone knows /something/ about
this topic.
S
