From: Eric Muller (emuller@adobe.com)
Date: Wed May 21 2008 - 23:58:02 CDT
Philippe Verdy wrote:
> - But the presentation HTML page for French
> (http://www.unhchr.ch/udhr/lang/frn.htm) is still using the ASCII quote.
>
The pages at the unhchr.ch website are not under the control the UDHR in
Unicode project. There is an HTML version on the unicode
> And another problem:
> - Also in articles 4, 18, 21.3, 22, 25.1, 26.1 there's a semi-colon missing
> a non-breaking space before it (NNBSP is possibly preferable, but NBSP is
> widely used in most French sources, SPACE is always incorrect
Please stop making authoritative-looking statements when they are not.
It is true that the rendered appearance should have some white space
before a semicolon, in essentially all the modern typesetting traditions
(an probably many older ones). But it is much less clear how to
represent that in Unicode texts. There are at least three options:
1. no space character of any sort, rendered white space materialized by
the rendering software, much like Japanese akis
2. a U+0020, which the rendering software can materialize with the
appropriate width, and treat as non-breaking
3. some space character which carries all the necessary semantics of
width, stretchability and line-breakability
I agree that 1. is not very good. It is used in the unicode.or/udhr
text, and should probably be changed.
I personally think that 3. is not viable because a) Unicode does not
currently contain the necessary assortment of characters, b) keyboards
would probably not include all them if they existed, c) I don't think we
can expect users (at least those who are not professional typesetters)
to enter the "right" character, even if keyboarding is easy.
2 seems the best choice to me. The determination of appropriate width,
stretchability and line-breakability seems amenable to automatic
computation. You can experiment for yourself such an implementation in
Flash Player 10 which just went in beta.
>
> - This is standard rule for all punctuation signs drawn using more than one
> glyph,
glyph -> multiple pieces of ink (as in ":;!?«»"). Actually a better
characterization is "high punctuation", i.e. punctuation that extends
well above the baseline, although the precise amount of space, its
stretchability and its location varies with the signs.
> a semicolon is
> used in article 26.1 in French due to the problem when translating
> "generally"
It seems to me that the French is far from being a translation. Given
the history of that text and the folks involved, I suspect it was
written simultaneously in English and French. Incidentally, I find the
style of the French text very elegant and very well suited to that
particular text.
>
> Hmmm.... How does the reviewing works?
>
http://www.unicode.org/udhr/contributing.html
Eric.
This archive was generated by hypermail 2.1.5 : Thu May 22 2008 - 00:02:35 CDT