Re: Tifinagh - extension for complete common Berber alphabet isomorphic with Latin

Date: Tue Feb 16 2010 - 19:40:44 CST

  • Next message: "Re: Tifinagh - extension for complete common Berber alphabet isomorphic with Latin"

    From: Hans Aberg ( on Tue Feb 16 2010 - 14:23:40 CST

    > On 16 Feb 2010, at 15:54, Doug Ewell wrote:
    >> Fair enough. Do you feel there is an analogy between the math
    >> compatibility alphabets and the proposed new Tifinagh letters?
    > I just noted that characters have been added on the principle on
    > planned future common use.

    > As for the experimental characters, my guess you might do the same if
    > they get some such status. It's a bit hen and egg problem: you don't
    > add them because nobody uses them, and they are not used because they
    > are not added.
    > Hans

    It would be a chicken and egg problem if computer representation were the only
    ubiquitous means of representing text. Fortunately, there is this thing called
    handwriting, and even though it's old-fashioned, you don't need to take up room
    in the Unicode Standard to incorporate a new character. The PUA is for the
    quite understandable desire to use the new character(s) on computers right now:
    you standardize in the PUA until you can demonstrate usage in the wild, then
    you do a simple one-to-one mapping to the eventual official Unicode allocation.

    This gives you two, very easy means of generating the proper documentation for
    a UTC proposal: 1) people find the proposed characters useful and use it in
    their handwritten correspondence 2) people find the proposed characters useful
    and use them from the PUA. Each demonstrates usage and the necessity of
    standardized encoding.

    My suggestion is to find yourself some good PUA codepoints (suggestion: E650-
    E67F) after checking with the Conscript Unicode Registry (the only fairly well
    supported Private Use registry I know of), and staying out of the first 256
    code points to avoid conflicting with the most common location of REALLY
    private use characters. Get a font to support this PUA assignment, and see if
    anyone actually uses it. Also, get scholars, students, and others with interest
    to use them in private correspondence, and use the handwritten letters as
    documentation. Write it up as a scholarly proposal for the larger community,
    and again, see if anyone uses it. It may be the most logical script extension
    ever conceived, but if it is not used, it has no place in Unicode.

    This also enables you to see what use others will put the concept. You may be
    suprised at what happens when others get ahold of this, it may blow your
    encoding model out the window, in which case, Unicode assignment was not only
    premature, but was subverted by actual usage. This is why proposals require
    documentation, and why some proposals need to demonstrate that the encoding
    model actually represents the text properly.

    -Van Anderson

    This archive was generated by hypermail 2.1.5 : Tue Feb 16 2010 - 19:46:21 CST