RE: Suggestions in Unicode Indic FAQ

From: Kent Karlsson (
Date: Fri Jan 31 2003 - 11:23:32 EST

  • Next message: Mete Kural: "How is glyph shaping done?"

    Keyur Shroff wrote:
    > No fallback rendering is coming into picture with your explanation.

    Yes, there is. A character sequence <FULL STOP, VOWEL SIGN E> (say)
    is very unlikely to have a ligature, specially adapted (and fitting)
    adjustment points, or similar. The rendering would in that sense
    need to use a fallback mechanism that renders an "approximation"
    for this rare combination.

    > Here is the para you are talking about.
    > [Quote]
    > should be rendered as if they had a space as a base character."
    > [/Quote]
    > In the text there is no mention of explicitly inputting space character
    > before any combining mark that is defective combining character.

    The text says "as if". Which I also emphasised before.

    > Also, the word "should be rendered" implies that it is recommendation.

    Yes. A rather good one.

    > > By removing that particular fallback mechanism from implementations
    [inserting dotted circle glyphs for allegedly "invalid" combinations]
    > > as well as the TUS text! (I'm serious!) This particular fallback
    > > mechanism is NOT recommended as it stands.
    > Note that the text has been written in the section "Implementation
    > Guidelines". Can't it be considered as recommendation?

    That particular one, no. Just an example [that isn't very good,
    outside of a general "show invisibles" mode].

    > > But since its mention is erroneously taken as a recommendation, I'd
    > > suggest removing also its mention.
    > This is disastrous! What will happen to the systems which already
    > implemented this recommendations!?

    It's not a recommendation.

    > Will they be considered invalid
    > implementation afterwards? What is about stability?

    They are ugly implementations as they are. And will stay ugly
    implementations. Stability is good ;-).

                    /Kent K

    This archive was generated by hypermail 2.1.5 : Fri Jan 31 2003 - 12:10:22 EST