Glyph Selection in Absence of Variant Selectors

From: Richard Wordingham (
Date: Mon Feb 27 2006 - 02:59:05 CST

  • Next message: Peter Constable: "RE: Glyph Selection in Absence of Variant Selectors"

    Dear List,

    I am having a problem understanding variant selectors. If the form of a
    character may be selected by a variant selector, is its form nevertheless
    specified to the same degree if no variant selector appears? One
    possibility is that the renderer is free (within the constraints set by
    specifications by other means, e.g. 'features', 'language', etc.) to choose
    the form according to the renderer's author's taste and inclination (e.g.
    programming time constraints). The other is that the Unicode standard
    should specify the selection. In the latter case, should the proposal for
    the addition of the variant selectors (and character, if appropriate)
    specify the selection rules?

    The question may be relevant to the Lanna proposal currently being reworked,
    with the glyph in question the AA dependent vowel. However, the general
    question needs to be answered first. If being 90% correct for most authors
    is good enough, then there may be no issue. However, it has been suggested
    that it is desirable that an author should be able to tell what alternative
    forms he has selected. (The notorious example is breaking a conjunct in
    Devanagari - an author cannot immediately see whether it was he or the
    font+renderer that chose to use a half-form.)


    This archive was generated by hypermail 2.1.5 : Mon Feb 27 2006 - 03:03:04 CST