Re: String name and Character Name

From: Patrick Andries (
Date: Wed Apr 20 2005 - 07:08:44 CST

  • Next message: Frank Yung-Fong Tang: "Re: Unicode lexer"
    Andrew C. West a écrit :
    On Wed, 20 Apr 2005 10:40:20 +0300, Erkki Kolehmainen wrote:
    I specifically oppose the proposal, too.
    It never occurred to me even as a remote possibility that no reaction to 
    this proposal could be interpreted as support by anyone, including the 
    Claiming the silence of the majority as evidence of support for one's position
    seems to be an unfortunately common ploy from some members of this list.
    Not only on this list but also sometimes in more august technical assemblies.

    Perhaps like others on this list I didn't specifically oppose Peter's proposal
    as I did not want to hurt his feelings by calling it an infantile response to a
    non-existent problem (I do not doubt that some characters are misnamed, but
    misnamed or not all Unicode character names are adequate for their intended
    The question is obviously what is their intended purpose.

     Surely it cannot be an accurate description that could help find the character (without looking at the glyph which may vary or be wrong) since :
    1) you have to know the English transcription or translitteration system used (and know English)
    2) there are errors that actually given the internal transcription system would refer to other characters (saw another series of Ethiopic ones yesterday <>).

    Now, the purpose may be to serve as an official unique identifier (but there are also unique official French ISO 10646 names), but really "U+[xx]xxxx" is a much shorter real unique identifier (and language-independent). In fact, a living memory of the early days of the ISO 10646 tells me that it was decided in 1995 that the only identifiers would have the following form : Uxxxx[xxxx] (8 x for ISO 10646 in those days).

    P. A.

    This archive was generated by hypermail 2.1.5 : Wed Apr 20 2005 - 07:09:33 CST