Re: Case mapping of dotless lowercase letters

From: jcowan@reutershealth.com
Date: Mon Dec 15 2003 - 12:24:43 EST

  • Next message: Philippe Verdy: "RE: [OT] Euro-English (was: Corea? (Re: Swastika to be banned by Microsoft?)"

    Arcane Jill scripsit:

    > I sometimes wonder whether or not it was a wise choice to regard "LATIN
    > SMALL LETTER I" and "LATIN SMALL LETTER DOTLESS I" as distinct. Too late
    > to change it now, of course, but (with the benefit of hindsight) it
    > occurs to me that if U+0069 had been regarded as dotless, all these
    > problems would never have arisen. Western fonts could still have
    > rendered it with a dot, Turkish fonts could have rendered it without a
    > dot, and everyone would have been happy.

    Unicode didn't get any choice about it. The 8859-3 and 8859-9 equivalents
    separated dotted-i from dotless-i (Turkish needs both) and identified the
    former, not the latter, with the ASCII i.

    > As an analogy, albeit a rather silly one, if (in mathematics) I put a
    > dot over a (single-letter) variable name to indicate (say) first
    > derivative or something, I would have to put an /extra/ dot over i,
    > would I not? Does that not make it "conceptually" dotless, even though
    > it's rendered with a dot?

    In a sense that's correct: an i or j with an accent loses its dot, and if
    there is a dot present regardless, it is a dot-above diacritic and not
    the native dot.

    -- 
    John Cowan  <jcowan@reutershealth.com>  http://www.ccil.org/~cowan
            Raffiniert ist der Herrgott, aber boshaft ist er nicht.
                    --Albert Einstein
    


    This archive was generated by hypermail 2.1.5 : Mon Dec 15 2003 - 13:01:27 EST