Re: Defined Private Use was: SSP default ignorable characters

From: Ernest Cline (ernestcline@mindspring.com)
Date: Tue Apr 27 2004 - 12:27:18 EDT

  • Next message: John Hudson: "Re: Proposal to add 2 Romanian characters"

    > [Original Message]
    > From: Doug Ewell <dewell@adelphia.net>
    >
    > Ernest Cline <ernestcline at mindspring dot com> wrote:
    >
    > > Others, such as Line Break, Bidi Class, and Casing are important,
    > > are used by existing software, and unlike Cursive Joining cannot
    > > simply be handled at present by putting out a Private Use font which
    > > is the current way that Private Use characters can be most easily
    > > and portably implemented. It is for support of these properties,
    > > that having a more precisely defined set of Private Use characters
    > > would be of use.
    >
    > I'll do the Unicode Character Properties for everything in ConScript if
    > others will agree to specify the properties for their own private-use
    > thingies.

    As others have pointed out in the past, because there exist Private
    Use scripts that expect the current existing Private Use defaults,
    it would not be a good idea for Unicode to change the defaults of
    existing Private Use characters. It would be slightly helpful if there
    was a means to include guidance as to which set of Private Use
    characters was being used. It would be quite a while before
    applications would be able to take advantage for such information,
    but it could serve as the basis of a long term solution.

    U+E0002 PRIVATE USE REGISTRY TAG

    might be a possible long-term solution that would enable
    such information to be stored in-band, altho give the lack
    of acceptance for in-band tags, the most I would expect
    from it would be to help define a common standard for tagging
    the set of private use characters in use that could be adopted
    by markup rather than the use of the tag characters themselves.

    > (Michael, John... I might need a little help with Tengwar.)

    TENGWAR DUODECIMAL DIGITS TEN and ELEVEN
    present an interesting problem. They are digits, but not
    decimal digits. Should the concept of General Category
    Nd be expanded to include non-decimal number systems?

    Or would
    E06A;TENGWAR DIGIT TEN;Nl;0;L;;10;10;10;N;;;;;
    be sufficient?

    This archive was generated by hypermail 2.1.5 : Tue Apr 27 2004 - 13:16:11 EDT