Poring back over this voluminous thread to Stephan Stiller's original question:
> If one wants to indicate vowel length for the length-ambiguous vowels α, > ι, υ in Ancient Greek, one writes ᾱ, ῑ, ῡ. Is there a reason for why > there are no diacritic-precomposed characters? I guess it's because > macron usage is rare in orthographic practice, even though vowel length > here is not clearly less important than the other phonetic aspects > indicated by the various diacritics in use in polytonic orthography. > Thus I am wondering when and how the relevant decisions were made. The decision was made in 1992, in SC2/WG2, as part of the deal which made the merger of the Unicode Standard and ISO/IEC 10646 possible. Note that Unicode 1.0 did not contain any precomposed polytonic Greek. That was rather included in the early drafts of ISO/IEC 10646 at the behest of the Greek national body, which at the time was attempting to standardize polytonic Greek. Inclusion of the precomposed characters now seen in the U+1FXX block was part of the price of the merger. What was included was precisely the repertoire requested by Greece, and no attempt was made to further rationalize forms including macrons for Ancient Greek. --Ken