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

From: Doug Ewell (
Date: Wed Aug 06 2003 - 18:24:35 EDT

  • Next message: Kenneth Whistler: "Re: Conflicting principles"

    Peter Kirk <peter dot r dot kirk at ntlworld dot com> wrote:

    >> Or it may not. It may be a deficiency in the level of Unicode
    >> support afforded by the fonts and rendering engines. ...
    > If there are such deficiencies in fonts and rendering engines which
    > purport to be Unicode compliant, that suggests a lack of clarity in
    > the standard which should be rectified.

    I wish I had a dollar for every "Unicode-compliant" font, rendering
    engine, or other software that was in some way less compliant than
    advertised. Only a fraction of the non-compliances are traceable to
    ambiguities or deficiencies in the Unicode Standard.

    >> ... It may simply reflect a difference between your "requirements"
    >> and what the standard promises, and doesn't promise.
    > If Unicode doesn't promise what I require, surely it is at least
    > reasonable for me to ask on this list whether it ought to be extended
    > or clarified to do so. The UTC may choose not to make any changes, but
    > I don't see why they shouldn't even be asked to.

    Absolutely, you are allowed to ask. Go ahead. I wasn't trying to
    prevent questions from being asked, only trying to state why I think the
    problem is out of scope for Unicode.

    >> The standard doesn't say anything about width in this case. It
    >> leaves it up to the display engine, which is as it should be.
    > The standard does say, section 2.10 of 4.0, that "In rendering, the
    > combination of a base character and a nonspacing character may have a
    > different advance width than the base character itself".

    I apologize for missing this reference.

    > And any intelligent typographer will realise that this "may" is a
    > "must", with regular character designs but not of course in monospace,
    > in some cases like the example given of i with circumflex. This
    > sentence applies to spaces with diacritics as space is a base
    > character, as we have been informed. The subsection of 2.10 entitled
    > "Spacing Clones of European Diacritical Marks" (by the way, why
    > "European" when the text appears to apply to all diacritical marks?)
    > should suggest to any intelligent typographer that the sequence space,
    > diacritic is intended to be spaced as the diacritic and not as a
    > space, but it would help for this to be clarified as not all
    > typographers are very intelligent and some may not be aware that this
    > space has actually lost most of the properties of a space e.g. line
    > breaking and is being used only "By convention".

    Like Freud's cigar, sometimes a "may" is just a "may." And I suspect
    the phrase "any intelligent typographer" MAY generate some flak from
    typographers on this list who consider themselves "intelligent enough"
    yet have a different opinion.

    I'm not a typographer (intelligent or otherwise), but I'm having a tough
    time seeing how Section 2.10 *requires* fonts and rendering engines to
    give a space-plus-combining-diacritic combination the exact minimum
    width of the diacritic alone, or to leave equal space before and after
    such a combination. All I think it is saying is that, for example, the
    combination i-plus-tilde may be wider than i alone, because tilde is
    wider than i.

    >> When the specific alignment of isolated glyphs is important to me, I
    >> use markup. I'm a big supporter of plain text, as many members of
    >> this list know, but the exact spacing of isolated combining marks
    >> seems like a layout issue to me.
    > OK, what kind of markup should I use, in any well-known markup
    > language, to ensure that an isolated diacritic is centred in the space
    > between the words before and after it?

    All right, you've got me there. I'll have to think about it. But I
    still think this is a layout problem, a problem having to do with glyphs
    and not characters.

    -Doug Ewell
     Fullerton, California

    This archive was generated by hypermail 2.1.5 : Wed Aug 06 2003 - 19:00:43 EDT