RE: Use of interum PUA encodings for 85 letters

From: Philippe Verdy (
Date: Sat Oct 13 2007 - 14:21:50 CDT

  • Next message: Rick McGowan: "Re: Use of interum PUA encodings for 85 letters"

    If you want to avoid possible conflicts during the time of experimentation,
    you may look at the “Conscript” registry, asking them to reserve you some
    range for your 85 letters (request a block of 96 code points and describe
    your intent, including the expected time of completion of your complete


    Then start encoding your font there. If you need to experiment privately
    before sending a formal proposal, just look at the Conscript registry, make
    sure you’re somewhere in the middle of the unallocated area, and create a
    alpha font for your private use only.


    When you’re ready for a proposal, create some PDF to avoid the problem of
    using codepoints in the published document (it won’t conflict with any other
    implementation, because the internal encoding of PDF documents remains
    private to each document, as long as you don’t bind the glyphs to reserved
    names glyph names: check these names on the Adobe website.) Use this
    published document to ask for support in the Conscript registry. At the same
    time you may submit your proposal to the UTC and ISO, using the procedures
    described, but then be prepared to justify all your choices when answering
    to the questions in the required submission form.


    But if your script encoding is too far to be usable by several independent
    third-parties, you can use any PUA you want, without even asking to
    Conscript to assign you a temporary block.


    Note that NO block in the Conscript registry is permanently assigned (like
    in Unicode). If the experiment fails or does not produce any formal
    proposal, it will be deallocated after some time and reusable for something
    else. If you register to Conscript, this means you want to be contacted for
    discussing your proposal, and must be ready to accept changes and signal
    changes in your proposal (you are not required to keep your initial PUA
    assignments for the whole time of your project)…


    Note also that many fonts use the end of the PUA block of the BMP for
    mapping contextual glyph variants. This may permit you to reuse the
    behaviour of some other scripts supported on some system using fonts with
    these PUAs.


    Some of the problems you may be exposed could be related to the default
    behaviour of renderers that treat PUAs like left-to-right symbols. They
    won’t be able to perform normalizations, or interpretation as combining
    characters, so you’ll need to tweak some renderer to allow giving the
    correct character properties (notably for directionality)


    Finally, as long as your encoding will remain in PUA, you won’tbe able to
    produce plain-text only documents that are usable without some required
    fonts. For keeping the association between plaintext and the required font,
    you will most often need a rich-text format (RTF, HTML…). To experiment with
    advanced properties, you may tweak an existing version of ICU and of some
    open-sourced renderer and layout engine, to add your needed properties in
    its properties data files (don’t restart these needed tools from scratch,
    such development is really complex and very long).


    If designing fonts is not your best skill, you may use the PUA’s that are
    mapped in ideographic encodings: you’ll get immediately a usable input
    editor, a virtual keyboard to start with, and a simple bitmap glyph editor
    for your private ideographs. You won’t get superb rendering, but at least it
    will permit you to discuss your proposal easily with others, including
    qualified font designers. Then create your own keyboard using these PUAs
    (the MSKLC tool will help you on Windows, similar support for custom
    keyboard maps exist for other platforms including Unix/Linux).



    De : [] De la
    part de PCL Institute
    Envoyé : samedi 13 octobre 2007 03:11
    À :
    Objet : Use of interum PUA encodings for 85 letters




    I'm involved with a project that is creating a new alphabetic orthography.
    The orthography is quite new, and not another exists for the language. I'm
    about to float the completed proposal summary for the script to this
    newsgroup for review. The process will minimally take two, and be more like
    3 years before the script shall be able to be incorporated into the Unicode
    standard. Can someone fill me in on Interum Solution employment of the
    Private Use Area? What does one do in order to use encodings for 85 new
    letters via the PUA, in the mean time?





    This archive was generated by hypermail 2.1.5 : Sat Oct 13 2007 - 14:23:56 CDT