Re: Public Review Issue: UAX #24 Proposed Update

From: Peter Kirk (
Date: Thu Sep 09 2004 - 04:31:04 CDT

  • Next message: John Cowan: "Re: Public Review Issue: UAX #24 Proposed Update"

    On 09/09/2004 04:47, Jony Rosenne wrote:

    >FB1D, HEBREW LETTER YOD WITH HIRIQ, should be assigned to the unknown group.
    >It is not a Hebrew character, notwithstanding the misleading name.
    Nevertheless, it is canonically equivalent to the sequence <05D9, 05B4>,
    and this sequence is used in Hebrew, and that implies in some sense that
    FB1D is used in Hebrew because it is the same as <05D9, 05B4>. I don't
    know if it is technically required that script naming be stable under
    normalisation etc, but it would certainly seem to make sense to make
    this true unless there are very good reasons to the contrary.

    FB1D is redundant because it is a precomposed character, like most of
    the rest of the presentation forms and indeed most of the extended Latin
    blocks. And from a Hebrew viewpoint it is anomalous because it is just
    one of hundreds of consonant-vowel combinations, and one of only a few
    (FB2E, FB2F and FB35 are others) for which there is a precomposed
    character. But fortunately these presentation forms, which are
    composition exceptions, can be more-or-less ignored.

    On another issue, I was surprised to see that the most used Arabic
    combining marks are "Inherited", although the Hebrew combining marks are
    "Hebrew". I would expect these to be listed as "Arabic" for the same
    reasons. I know some of them are occasionally used with other scripts.
    But we know that there are cases in which scripts are mixed, e.g. the
    Latin letters in Cyrillic Kurdish, and so we can't expect to avoid all
    such inconsistencies.

    Peter Kirk (personal) (work)

    This archive was generated by hypermail 2.1.5 : Thu Sep 09 2004 - 04:32:22 CDT