RE: Character proposal: SUBSCRIPT TEN

From: Philippe Verdy (
Date: Wed Jan 16 2008 - 04:27:58 CST

  • Next message: Asmus Freytag: "Re: Latin J capital letter with caron"

    > De : [] De la
    > part de Leo Broukhis
    > Envoyé : mercredi 16 janvier 2008 05:04
    > À :
    > Objet : Character proposal: SUBSCRIPT TEN
    > The SUBSCRIPT TEN (== decimal exponent base) character remains the
    > only character that is present in the GOST-10859 standard
    > ( and in the character set of
    > the Soviet ACPU-128 drum printer but is absent is Unicode.
    > Also see the the German character set ALCOR
    > (
    > It cannot be replaced by SUBSCRIPT ONE + SUBSCRIPT ZERO, because it
    > has to occupy one character position for the sake of text aligned for
    > a fixed-width font.

    Font constraints (subscripts, half-width/narrow) are generally perceived as
    compatibility characters. In your case, the character would really represent
    the standard digits, except that it would exhibit a size restriction for
    specific monospace fonts in a grid layout.

    It differs here from the one and zero subscripts only for the halfwidth
    presentation possibly with narrow glyphs to make them fit in some monospaced
    font styles.

    But for all other cases, it would still remain normal subscripts, and should
    be treated as compatibility.

    Those environments where you really need monospaced fonts are those needed
    for editing very precise data, and where no layout feature is encodable, and
    string length is restricted. For normal plain text documents, not
    constrained in size they are not needed, and there's no problem even if you
    edit them using monospaced fonts in some text editor, even if they are "too
    much spaced".

    The monospaced font is just there to show which characters are present, and
    it is justified most like the "visible controls" mode. In normal text, they
    are not different, and even plain text documents should use the existing
    zero and one subscripts.

    I wonder if it was really required to count this character as occupying only
    one character position in restricted input forms; I still think that even
    the printed newspaper ads would not consider it as one character but still
    two, given that it will display the same as normal subscript when finally
    printed using variable fonts.

    Can you find for example a paper form with input rulers where you would have
    used the same position for writing the whole subscript? Or some database
    where it was used as a distinct identifier, or sorted specially?

    > Is there a chance for this character (or a way to request halfwidth
    > subscript/superscript characters) to appear in Unicode?

    But yes, it could be encoded as such, I think, if justified for
    compatibility with legacy standards (but at least the standard body that
    created or supported this character in its legacy encoding should ask for
    it...), but really it should not be used in any new documents (anyway if you
    need subscripts, you'll need them for all characters, and all possible
    numbers or expressions).

    So may be, you should request this to the Russian and German representants
    at ISO to see what is the status of this character in their encoding (or if
    its encoding is already deprecated and no more a standard since long).

    This archive was generated by hypermail 2.1.5 : Wed Jan 16 2008 - 10:58:35 CST