Re: Pre-1923 characters?

From: Philippe Verdy (
Date: Sat Jan 03 2004 - 16:37:37 EST

  • Next message: Michael Everson: "Re: Latin letter GHA or Latin letter IO ? (was: Pre-1923 characters?)"

    From: "Michael (michka) Kaplan" <>
    > From: "Michael Everson" <>
    > > At 09:03 -0800 2004-01-03, Peter Kirk wrote:
    > > >But in the light of naming errors like this one implementers should
    > > >be advised not to use character names, because they are not reliably
    > > >helpful.
    > >
    > > I wouldn't say that. It would better to advise them, as we do, that
    > > they cannot rely on the names being perfect. That's different from
    > > not using them at all.
    > It makes me wish we had a CouldaWouldaShoulda_CharacterName property that
    > contains what the name ought to be, and we document this as one that
    > change any time there is a mistake made in the original character name. We
    > just make a nice informative property and go through all of our known
    > mistakes and the maintenance after the initial pass should be minimal....

    Note that a fundamental property of character identity is its most common
    classification as a vowel, consonnant, or semi-vowel. Instead normative
    names just use the term "letter" which does not help identifying letters
    that were incorrectly named like "OI" which is better represented by the
    informative name "gha" that reveals its consonnantal identity...

    In that case, this "OI" letter gha is inappropriate to represent the "I with
    lower hook" letter exhibited by Peter, which is really shown as being a
    vowel (if used as a varyant of "i") or semi-vowel (if used as a variant of
    "y" or "j" or iota). And then we have to look for another candidate
    character to represent the missing letter: could it be a variant of the
    already encoded Latin letter iota? Or a variant of "j" (which was already a
    variant of "i" with a lower hook in the reverse direction). Considering the
    exhibited alphabet that sorts this new letter between I and J, and
    considering that the dotless-i will not represent it well, I think there's a
    better place to see this new character as a mirrored J...

    We can't say from the exhibited uppercase alphabet that this should be a
    mirrored dotless j or a mirrored soft-dotted j if it is converted to
    lowercase. So Peter, where did you find this image of an alphabet?

    This archive was generated by hypermail 2.1.5 : Sat Jan 03 2004 - 17:04:08 EST