Re: Proposing a DOUBLE HYPHEN punctuation mark

From: Philippe Verdy (verdy_p@wanadoo.fr)
Date: Mon Jan 29 2007 - 13:21:25 CST

  • Next message: Chan, Florence \(IT\): "ISO 15924 and ISO 639"

    From: "Jukka K. Korpela" <jkorpela@cs.tut.fi>
    > On Wed, 24 Jan 2007, Philippe Verdy wrote:
    >
    >>>>> Tam wisi czerwono-
    >>>>> -niebieska flaga.
    > - -
    >> Isn't this example encodable simply by using a regular SHY followed by
    >> an the orthographic hard HYPHEN?
    >
    > As mentioned in another message, a normal ("hard") hyphen tends to allow a
    > line break after it, so the break will probably appear in a wrong
    > position.
    >
    > However, you could use SOFT HYPHEN followed by NON-BREAKING HYPHEN. This
    > seems to work, in a sense, in some programs, but there's a presentational
    > problem. The glyph for NON-BREAKING HYPHEN often differs from the glyph
    > for SOFT HYPHEN when visible. The difference is rather noticeable. In
    > Times New Roman for example, NON-BREAKING HYPHEN is considerably longer,
    > almost like an en dash.

    If you have read the list, this was an old suggestion, and in fact I proposed something else:

    * SHY (soft hyphen, normally invisible if no break occurs there)
    * the normal orthographic HYPHEN, followed by a ZWNBSP (deprecated) or WJ (word joiner)

    This provides the ligation only on the right of the regular hyphen, which keeps its encoding as well as its associated glyph. The problem is to have WJ or ZWNBSP supported (note that they are not in ISO 8859-1/15 used very commonly with Spanish, and these characters are sometimes displayed as empty boxes for a missing glyph in very commonly selected fonts or default fonts of many systems).

    Well, in fact, it also happens that SHY is visible even if no breaks occurs, in some old systems (despite the SHY behavior was specified since very long in ISO 8859-1 and even before in prior character sets...)



    This archive was generated by hypermail 2.1.5 : Mon Jan 29 2007 - 13:23:29 CST