Gy. Dobner asked: > But my original question was not how to encode a combining macron in one > more possible way but how to encode a length mark that would display as > something _visually_ _distinguishable_ _from_ _a_ _macron_ (because the > macron is functionally ambiguous and hence unsuitable for my purposes). > Is it e.g. possible (i.e. is it Unicode-compliant) to combine a macron with > some > non-displaying character for this purpose, and if so, with which > non-displaying > character? I understand that ZERO-WIDTH JOINER is not supposed to be used > in this way (or am I mistaken?).
But this is the wrong question. The Unicode Standard encodes characters for scripts (and writing systems). It doesn't provide a standard for the representation of syllabic structure or other phonological constructs per se. Even if you are using some phonetic transcription system like IPA, which is used as a technical system for representing sounds, the Unicode Standard's encoding of that is one step removed. It is the International Phonetic Association that defines how IPA characters, marks, and other conventions are used to specify linguistic sounds. What the Unicode Standard does, in turn, is encode those character and marks for digital representation on computers. So I think you have the cart before the horse here. What you (or the Classicist community in general) need to do is specify orthographical conventions for the representation of whatever length distinctions you are trying to systematically distinguish. That could be with a colon. It could be with the IPA length mark. It could be with a doubled-macron. It could be with some entirely different diacritic. It could be with some other visible convention. Once you know *what* you want to write for this, *then* you ask, how can this written text be represented in Unicode characters, so I can enter, transmit, print, and otherwise process it on computers. It isn't a matter of some hidden format code in Unicode that normatively denotes lengthiness. Rather, you decide what you want to write and print for the distinction you need to make. Hint: Pick some *other* diacritic that already exists in the Unicode Standard. That way, you won't need to spend two years hassling with the character encoding process to add some newly invented mark which isn't yet encoded. Hint #2: Pick some diacritic mark that is already widely supported in system fonts. That way you won't need to spend years hassling system vendors to add the glyphs you need, or scouring the web looking for custom fonts, in order to be able to easily display your research on the web and with easily available tools. --Ken
