Re: Roman Numerals (was Re: Improper grounds for rejection of proposal N2677)

From: Philippe Verdy (verdy_p@wanadoo.fr)
Date: Fri Oct 28 2005 - 13:47:26 CST

  • Next message: Philippe Verdy: "Re: Roman Numerals (was Re: Improper grounds for rejection of proposal N2677)"

    From: "Kenneth Whistler" <kenw@sybase.com>
    > Philippe,
    >
    >> The only approximate alternative is to not use the existing Roman
    >> numerals
    >> at all, and revert to Latin letters, and then use C, I, and OPEN O (which
    > ^^^^^^
    > no
    >> looks quite similar to the turned C, except that the serif on is missing
    >> on
    >> the bottom leg, when drawn with serif fonts),
    >
    > It shouldn't be surprising that U+2183 ROMAN NUMERAL REVERSED ONE
    > HUNDRED looks *exactly* like a reversed C, because that is what it
    > is.

    Oh God. I forgot it was absent only from the fonts I use. Sorry.

    > Unlike the East Asian compatibility characters in the ranges
    > U+2160..U+217F, the ligated forms and the reversed C in the
    > range U+2180..U+2183 *are* intended for general use with the
    > Latin alphabet in forming the types of Roman numerals that you
    > are talking about.
    >
    >> or to replace the sequence
    >> <I,TURNED C> by <D>, and possibly add joiner controls between them to
    >> request (and may be force) their ligature.
    >>
    >> So to represent 888,888, ...
    >
    > Please recast this in terms of the characters encoded for such
    > high numeric value expressions, and you'd get much closer to
    > the intent.
    >
    > The convention of using rulings over strings of Latin letters
    > to indicate higher values should be handled by styles, rather
    > than by individual insertion of combining lines over single characters.

    How do you do that in HTML or CSS? It's quite difficult to emulate (notably
    the stretched M over groups of roman digits, but the macron/rulings are as
    much difficult to place as it requires playing not only on the span of text,
    but also playing with the whole formatting of the paragraph to change the
    line height), and there are plenty of similar issues.

    Once you start using those tricks, the text looses its accessibility, and
    requires specific classes of devices to be correctly rendered. This is not
    only a stylistic convention (how would you do if your renderer is not
    visual? The renderer would have no other choice than changing the
    semantics).

    Arguably, the thousand multiplier has a plain-text meaning that should be
    encodable as such. OK the ten multiplier is encoded with the reversed one
    hundred roman numeral (what a bad name that does not reflect its effective
    semantics!), but not the thousands.

    Only some ligatures (namely the ONE THOUSAND CD and FIVE HUNDREDS), based on
    only one of the rendering conventions, are encoded, and these ligatures are
    much less known and used than the other conventions based on explicit and
    distinct 10 or 1,000 multipliers.



    This archive was generated by hypermail 2.1.5 : Fri Oct 28 2005 - 13:49:31 CST