Re: Display of Isolated Nonspacing Marks (was Re: Questions on ZWNBS...)

From: Mark Davis (
Date: Tue Aug 05 2003 - 17:40:11 EDT

  • Next message: Peter Kirk: "Re: Display of Isolated Nonspacing Marks (was Re: Questions on ZWNBS...)"

    Where did you get the notion that space is not a base character? And
    base characters include those that are not control or format
    characters. Space is neither one.

    The standard specifically states in a number of places that to exhibit
    a combining mark in isolation you use a space (or NBSP).

    ► “Eppur si muove” ◄

    ----- Original Message -----
    From: "Peter Kirk" <>
    To: "Jim Allan" <>
    Cc: <>
    Sent: Tuesday, August 05, 2003 13:47
    Subject: Re: Display of Isolated Nonspacing Marks (was Re: Questions
    on ZWNBS...)

    > On 05/08/2003 09:42, Jim Allan wrote:
    > > Peter Kirk posted:
    > >
    > >> If I want to do this, should I explicitly encode a dotted circle,
    > >> should I encode nothing and expect the font to generate the
    > >> circle, as it often does?
    > >
    > >
    > > I think that practise of a font or application automaticaly
    > > a dotted circle under an orphaned combining character is dubious
    > > compliant with Unicode specifications.
    > >
    > > ...
    > >
    > >
    > Thanks, Jim, for all this data, but now I am totally confused. Well,
    > least it seems clear that if I want a dotted circle I should
    > encode it. But if I don't...
    > Suppose for example I want to write a sentence like "In this
    > the diacritic ^ may appear above the letters ...", but instead of ^
    > want to use a combining character, a regularly positioned centred
    > the letter diacritic, which does not have a defined spacing variant.
    > don't want a dotted circle. And I want it to be spaced as here, i.e.
    > with one space before the diacritic and one after it. It seems to me
    > that at one place in the standard I am told to encode space -
    > mark - space, for the combining mark will not combine with the space
    > because the space is not a base character; and in another place I am
    > implicitly told to encode space - space - combining mark - space,
    > because the second space acts as a carrier for the combining mark.
    > I hope that wanting to display this correctly is not another place
    > I "have stepped over the boundaries of what is reasonable to expect
    > plain text to convey", but that this too can be "grist for the
    > 5.0 mill to grind very finely" - both quotes from Ken Whistler
    > today. And I think that if this issue is clarified it will also
    > clear what should be done about string initial holam and alef etc.
    > Perhaps a simple way ahead would be to define a new character
    > like COMBINING MARK HOLDER with no glyph, which is defined
    > for this purpose, is a base character and not a format character,
    and is
    > expected to be just as wide as is necessary to display the combining
    > mark. Then we could say that a spacing accent is equivalent
    > even canonically if made a composition exclusion?) to COMBINING MARK
    > HOLDER plus a non-spacing accent, and remove the misleading
    > compatibility equivalences to SPACE plus a non-spacing accent.
    > --
    > Peter Kirk

    This archive was generated by hypermail 2.1.5 : Tue Aug 05 2003 - 18:18:32 EDT