RE: Glyphs for German quotation marks

From: Kent Karlsson (
Date: Fri Jun 16 2006 - 17:56:46 CDT

  • Next message: Richard Wordingham: "Re: U+0345 COMBINING GREEK YPOGEGRAMMENI not usable in other scripts as "hook below"?"

    John Hudson wrote:

    > > That's what happens now all the time in German texts with
    > > Courier [New] and Verdana! The German Usenet and German web pages
    > > are full of complaints about this since many years! That's how
    > > this thread started!
    > Yes, and to solve the German problem you want to force
    > everyone else to perform this kind of text encoding hacks.

    No, not at all. Only you seem to suggest that.

    > This is not a solution. This is a glyph design problem and should
    > be resolved at the glyph level. This means either redesigning
    > the quote glyphs in the problem fonts so that they are clearly
    > differntiated at small sizes and low resolutions without needing
    > to rely on slant angle,

    I don't think that the rotated form needs to be different
    from the unrotated form in all fonts. Note that many languages
    use the SAME character both for opening and closing quote marks.

    I do think that the mirrored form (rarely used, and should
    perhaps not be used) needs to be different from the unmirrored
    (&unrotated) form. Only the ASCII quotes should be straight.

    > or providing language-specific glyph variants.

    No thanks. That is bound to fail.

    > By the way, we already do the latter for the Sindhi comma,

    I think that is a very bad idea.

    > which has the same encoding as the Arabic comma but is a
    > mirrored form of the Latin comma. There is neither a need nor a
    > desire for separate encoding for these glyph variants.

    I think very much that is desired. Otherwise we could use just plain
    ASCII comma for the Arabic comma, Armenian comma, and Ideographic

    > point that Adam made right at the beginning of this
    > discussion: that Verdana and Tahoma
    > etc. are fonts intended primarily for use on screen, not at
    > the high resolutions of print
    > and not at the large sizes you are showing them in all your
    > examples. So the question is
    > the differentiation of these glyphs *when they appear as
    > oblique monoline strokes*.

    So? The rotated form would then look like the unrotated (high)
    form, but that is expected. It is NOT expected that that the
    rotated form would suddenly become a mirrored form instead.

    > So what are you saying? Again, we're talking about small
    > sizes and low resolutions -- the
    > situations for which the Verdana and Tahoma fonts were
    > designed. There are only these two
    > pixel patterns to work with \ / -- which one do you think
    > should be used for which characters?

    Both should look like /. Only the (largely unused) mirrored
    form would look like \. As mentioned, there is no absolute
    need to distinguish the opening and closing quote marks
    by glyph shape, indeed they are not distinguished in ASCII,
    and many use (high) [double/single] "right" quote marks
    both for opening and closing of quotes.

    > > I try again:
    > >
    > > Make an apostrophe as you like it and put it into U+2019.
    > > Rotate the apostrophe and put it into U+2018.
    > > Mirror the apostrophe and put it into U+201B.
    > > Lower the apostrophe to the base line and put it into U+201A.
    > And if your U+2019 apostrophe looks like / at typical text
    > sizes on screen, rotating it for U+2018 makes it look like... /


    > That is the design issue that the makers of these fonts were
    > dealing with.

    I don't see a problem in making these to look alike at small
    type sizes (or in very rare cases, even large type sizes).
    > Yes, I agree, and if your apostrophe looks like / at small
    > sizes, then you have limited
    > options for how to differentiate the rotated U+2018 form. As

    They don't need to be differentiated. They do need to
    be 180 degree rotations of each other, even if that makes
    them look alike.
    > I wrote before, this is a
    > glyph design problem when these characters are used in the
    > German context. We agree on

    I don't think there is anything particularly "German" about
    this issue.

    > this. What we disagree on is the solution. Any approach that
    > involves hacking the text
    > encoding to achieve a desired glyph display -- whether for
    > German or for other languages -- is a bad idea.

    That would be a bad idea. But no one but you suggests that.

    > The entire existence of U+201B is a bad idea.

    Maybe. I don't suggest actually using it, except perhaps
    in rare contexts like mimicking somebody's handwriting
    (including idiosyncratic abbreviations and misspellings).

                    /kent k

    This archive was generated by hypermail 2.1.5 : Fri Jun 16 2006 - 18:00:47 CDT