W3C home > Mailing lists > Public > www-style@w3.org > October 2007

Re: [CSS3 Text] Thoughts on hanging-punctuation property

From: Daniel Aleksandersen <aleksandersen+w3clists@runbox.com>
Date: Wed, 10 Oct 2007 00:00:17 +0200
To: fantasai <fantasai.lists@inkedblade.net>
Cc: W3C Emailing list for WWW Style <www-style@w3.org>
Message-Id: <200710100000.17903.aleksandersen+w3clists@runbox.com>

On 2007-10-09, fantasai wrote:
> Daniel Aleksandersen wrote:
> > On 2007-10-09, fantasai wrote:
> >> Daniel Aleksandersen wrote:
> >>> These are the once I would like to see instead:
> >>> none | [ start || left-edges || edges || end || right-edges ]
> >>>
> >>> As everyone can see I use plural in ‘edges’ to clarify that it will
> >>> apply on multiple edges.
> >>
> >> Multiple edges?
> >
> > Yes, the edge of every line. I call that more than one. Sorry if my
> > English is a problem for my you. ;-)
>
> Ah. I'd consider that one edge. :) It's the left/right edge of the
> paragraph.
>
> >>> I also changed it from start and end to left and right
> >>> edges; to further clarify which edges will get hanging‐punctuation.
> >>> Another reason for doing this is that ‘left hanging‐punctuation’ is a
> >>> common term in typography.
> >>
> >> The reason for using 'start' and 'end' instead of 'left' and 'right'
> >> is that it automatically works correctly both for right-to-left and
> >> left-to-right scripts.
> >
> > Yes. I actually understood there was a internationalisation reason. But
> > I still thinks using left and right is better. It is basically the same
> > thing. But since CSS addresses almost every other direction using left
> > and right, I though it was best to use it here to. And as I said, it
> > makes things more simple.
>
> No, it makes things more complicated. 'start' is the better option
> because it works in all cases and handles e.g. automatic translation as
> well. I see no good reason to make a less-good option available as well.
>
> ('text-align' will also be taking a 'start' value: we're trying to
> transition towards start/end rather than left/right for these things.)
>
> > If you look trough the two character blocks I proposed—try using
> > http://www.fileformat.info/info/unicode/block/index.htm—you will see
> > that it makes sense for almost every character. Except the U+2052
> > COMMERCIAL MINUS SIGN and other glyphs that appear separated from other
> > characters using U+20 SPACE and other spacing characters.
> >
> >> Can you post examples (e.g. scans) of where this is applied to other
> >> punctuation, or where "left-edge" ('start-edge') hanging punctuation
> >> is used?
> >
> > http://livedocs.adobe.com/en_US/Illustrator/13.0/images/tp_36.png
> > Left hanging punctuation to the right. (Image courtesy of Adobe.)
>
> Ok, that could be handled with the existing definition for
>    hanging-punctuation: start;
> since the emdash is on the first line of a block.
>
> > http://www.artlebedev.com/mandership/120/
> > Showing left (first illustration) and right (second illustration)
> > hanging punctuation to the right. Try hovering the two images! (Images
> > courtesy of Artemy Lebedev.)
>
> Hmm, the hanging quotation mark at the start would work with
> hanging-punctuation: start (that's what the 'start' value was added
> /for/), but the parenthesis on the fifth line would indeed require a
> 'start-edge' value.

‘Start-edge‘ sounds goofy.

I did not notice it, but Artmey had some thoughts on what characters should 
be hanged as well: ‘Quotation marks, brackets, and bullets [...]’. I would 
have added: hyphens, dashes, and parentheses; as well as periods, commas,  
interobangs (a non-common glyph combining a question mark and an 
exclamation mark), [double] question marks, and [double] exclamation marks.

> How common is that effect? It looks a bit weird to me.

Here in Norway we see it in the large news papers, school books, and other 
books. It is actually common in print in general. I saw it on a poster on 
the train today too.
-- 
Daniel Aleksandersen
Received on Tuesday, 9 October 2007 22:03:00 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:55 GMT