Re: Glyphs for German quotation marks

From: John Hudson (john@tiro.ca)
Date: Thu Jun 15 2006 - 18:37:22 CDT

  • Next message: John Hudson: "Re: Glyphs for German quotation marks"

    Andreas Prilop wrote:

    > But we do speak about code positions here: U+2018 and U+201B.
    > This is a question of standards. You can make a Latin letter "g"
    > according to your own taste, but you cannot put it into an
    > arbitrary code position.

    > Nobody objects to the *glyphs* in Verdana, I believe.
    > The objection is against the abuse of code position U+2018.

    ...

    > The Unicode standard says:
    > 2018 left single q.m. = single turned comma q.m.
    > 2019 right single q.m. = single comma q.m.
    > 201B single high-reversed-9 q.m. = single reversed comma q.m.
    >
    > "turned" means "rotated", certainly not "mirrored".
    > "reversed" means "mirrored".
    >
    > If you do not agree, please give us *your* definition of U+201B.

    I don't disagree with this. The point I have been making is that in the context of type
    designs that use oblique monolinear quotations marks, the direction of slant may be either
    to the left or right for U+2018. That is to say, this is considered by designers a
    legitimate glyph variation for most languages in this kind of design.

    U+2018 is identified as a *left* single quotation mark. But what does left mean in the
    context of an oblique monolinear quote, where the only possible distinguishing
    characteristic between left and right quotes is slant direction? Do you see the design
    ambiguity?

    In 6/9 style quotes, or wedge-shaped quotes, left-right directionality is determined by
    relative position of the heavy and light portions of the glyph, e.g. ‘ ’, and they are
    *differentiated* in this way. If the design is an oblique monolinear stroke, how can you
    differentiate the left and right quotes? The obvious solution is to slant them in opposing
    directions, i.e. with the left quotes slanted to the left \ and the right quotes slanted
    to the right /. If you slant them both the same way, you have lost the visual
    differentiation. This is why \ is considered by type designers to be a legitimate and
    useful glyph form for U+2018.

    Again, I'm not entering the debate on whether this is acceptable or not for German, only
    explaining why \ angled quotes make sense as legitimate glyph forms for U+2018 in a
    broader context. You may have a very good case for these not being appropriate for German,
    but I don't think you have a good case for saying that this glyph form is an 'abuse' of
    the left quote codepoint.

    > No, it is not about language specific glyph variants, but about
    > code positions U+2018 and U+201B. The mechanism is provided by
    > the Unicode standard. However, I think the design of the quotation
    > marks in Verdana etc. is just lazy because current software
    > (programs & keyboard layouts) does not provide access to U+201B,
    > which is outside cp1252 and MacRoman.

    Leaving aside the question of whether texts should have differently encoded quote marks
    depending on the form desired from a particular font...

    It isn't laziness, it is because the left \ form is not only a perfectly legitimate glyph
    for U+2018 in the majority of contexts but is a *preferred* form in the majority of
    contexts because it is differentiated from the right / form. This is the point Adam was
    making earlier: a design decision is made to differentiate these two different codepoints,
    U+2018 and U+2019.

    You are objecting saying that, even for oblique monoline quotes, U+2018 and U+201B should
    never be identical, that the former should be / and the latter should be \. But surely you
    see that saying these two characters should not be identical means that U+2018 and U+2019
    will be identical instead. That might suit the Germans better, but does it suit everyone
    else better?

    Personally, as a designer now aware of this issue, I would avoid using oblique monoline
    quotes in any font intended for broad multilingual use. The limitations of the form are
    too great.

    John Hudson

    -- 
    Tiro Typeworks        www.tiro.com
    Vancouver, BC         john@tiro.ca
    I am not yet so lost in lexicography, as to forget
    that words are the daughters of earth, and that things
    are the sons of heaven.  - Samuel Johnson
    


    This archive was generated by hypermail 2.1.5 : Thu Jun 15 2006 - 18:38:48 CDT