W3C home > Mailing lists > Public > www-international@w3.org > January to March 2006

Re: Justification in Arabic, Korean

From: Martin Duerst <duerst@it.aoyama.ac.jp>
Date: Wed, 15 Mar 2006 19:01:02 +0900
Message-Id: <6.0.0.20.2.20060315183940.036ce7e0@localhost>
To: "'WWW International'" <www-international@w3.org>, Paul Nelson <paulnel@winse.microsoft.com>, "'WWW International'" <www-international@w3.org>

At 06:55 06/03/13, fantasai wrote:
 >
 >CCing i18n people in case they have input...
 >
 >Paul Nelson (ATC) wrote:
 >
 > >
 > > 1. Korean typography uses white space justification. They never function
 > > as ideographic text. I have confirmed this with two different people,
 > > but am waiting for bood resources.
 >
 >The test case for this would be in these mixed-script cases:
 >   If there is a line where the Korean has inter-character expansion
 >     while the Latin does not, then it is definitely as ideographic
 >   If there is an article where Japanese or Chinese has inter-character
 >     expansion while Latin and Korean do not, then it is definitely not
 >     as ideographic
 >   If there is an article where Japanese or Chinese *and* Korean have
 >     inter-character expansion but Latin does not, then it is
 >     definitely not as ideographic.

The few books I have here (mostly on lettering, dated in the 1980es)
have spaces between words, but they also use what one might call
ideographic line breaking (breaking at any point between two
characters (except for certain exceptions such as before punctuation,...).
Inter-letter spacing didn't seem to occur. The spaces were about
half the size of a Hangul character, in one book considerably
smaller. There were usually enough spaces on a line, and because
a word can be broken everywhere, the problem of wide white spaces,
as one can often see it in narrow columns in western languages,
didn't show up.

There were a few examples of Latin words in Korean, but mostly,
these were just single words, and often with a Korean ending
attached, without a space. Hanja also showed up, but mostly
just one or two, in parentheses. That was the case that most
often lead to expanded inter-word space, because breaking
lines just inside parentheses is disallowed.


 >If we can't find any of these three, then it is hard to say: it
 >might just mean that Korean prefers choosing 'inter-word' to
 >'inter-ideograph', not that the 'inter-ideograph' scheme would not
 >expand Korean.
 >
 >Our best bet would be to find Korean embedded and justified in a
 >Chinese or Japanese paragraph, and see which of the last two cases
 >it matches.

The big problem with things like this is that these cases are rare,
and therefore typographic practice isn't very well established, and
one or the other typographic practice often dominates in a mixed
script situation, due to the typographer not knowing enough of
the other script's practice, because of technology limitations,
or because of a concious decision to not mix practices.

One other point: According to JIS X 4051, justification in Japanese
text is first done in the 'space' after punctuations before using
true inter-ideographic spacing. For its description, JIS X 4051
actually describes full-width punctuation marks such as 、and 。
as consisting of a half-width glyph and a half-width space.


 > > 2. Yes. There is a need for Newspaper style justification that stretches
 > > Arabic scripts evenly. This is used with Arabic and Uighur.
 >
 >And this definitely
 >   - expands just the connections like lam-reh, not the disconnected
 >     sequences like reh-lam?
 >   - takes precedence over stretching word spaces?

For the examples I seem to remember, definitely yes on both accounts.
What I seem to remember from explanations is that there are specific
points in a word where the stretching is done, not every connection
is used. This is what the kashida/tatweel character in Unicode is
there for, to indicate these stretching points.

 > > 3. I have not had time yet to contact Chinese type people. This week or
 > >    next week I may have a chance.
 >
 >Ok. Feel free to ask them about my scans. :)
 >   http://fantasai.inkedblade.net/style/discuss/emphasis-marks/
 >I wish I could find that book with x-shaped marks...

I can remember having seen some of these, but I have no clue where
that was (but it was probably a Japanese context).

If I have a chance, I might have a look at some University
entrance examination problem collections. These are the
cases where typography goes farthest in terms of things
like double underlining with different styles, various kinds
of marks, and so on.

Regards,     Martin. 
Received on Wednesday, 15 March 2006 10:03:04 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 19:17:07 GMT