Re: Arabic encoding model (alas, static!)

From: John Hudson (tiro@tiro.com)
Date: Thu Jul 07 2005 - 01:47:06 CDT

  • Next message: John Hudson: "Re: Arabic encoding model (alas, static!)"

    Richard Wordingham wrote:

    > The Latin alphabet precedent is not promising - diacritics combine badly
    > in many fonts. There is a technical note on how to display them
    > (http://www.unicode.org/notes/tn2/), but it didn't seem to me that it
    > could be implemented in OpenType other than possibly by massive tables.
    > I may be wrong, but it looks as though it may be extremely laborious to
    > position Hebrew holam properly on all Latin consonants.

    Aside from the problem of not knowing what would constitute 'proper' holam positioning on
    a Latin consonant, I don't think there is any problem in this at all. Generally speaking,
    one needs only two anchor attachments for Latin letters -- one above and one below -- and
    then one anchor on each of the above and below marks if one wants to support mark-to-mark
    positioning. There are some language specific variations, but in general it is not very
    complex. It is something of a new area for most font developers, with limited tool support
    -- not to mention application support --, and adding combining mark positioning to a large
    library of existing fonts would be daunting for any foundry, but it isn't a hugely
    laborious undertaking as an aspect of new font development.

    I'm afraid that I have not followed all of this discussion, so I'm a little hazy about
    just what is being debated in terms of Arabic encoding. Mete Kural made some reference to
    a scholarly need for generative marks in Arabic, though, and I agree with this. It seems
    to me that, entirely independently of the Arabic encoding used for typical language
    processing, it would be desirable to have in Unicode a set of dotless archigraphemes (to
    use Tom Milo's phrase) and appropriate generative marks to be able to encode the text of
    Arabic manuscripts in a way that accurately reflects the writing conventions of the script
    in the early period.

    John Hudson

    -- 
    Tiro Typeworks        www.tiro.com
    Vancouver, BC        tiro@tiro.com
    Currently reading:
    Truth and tolerance, by Benedict XVI, Cardinal Ratzinger as was
    War (revised edition), by Gwynne Dyer
    God's secret agents, by Alice Hogge
    


    This archive was generated by hypermail 2.1.5 : Thu Jul 07 2005 - 01:48:25 CDT