W3C home > Mailing lists > Public > www-style@w3.org > October 2008

Re: [css3-multicol] column overflow

From: Håkon Wium Lie <howcome@opera.com>
Date: Fri, 17 Oct 2008 00:20:10 +0200
Message-ID: <18679.48666.989479.4910@opera.com>
To: robert@ocallahan.org
Cc: "David Hyatt" <hyatt@apple.com>, "www-style@w3.org List" <www-style@w3.org>

Also sprach Robert O'Callahan:

 > It's worth trying, and fairly easy to implement.

Good to hear.

 > One issue is that normally, when you scroll down by a page, Firefox (and I
 > think other apps) scrolls down by a little less than the full page height so
 > that some content that was at the bottom of the previous page is visible at
 > the top of the new page --- this gives users some retained context. 

Opera 9.6 adds a "scroll marker" for this purpose. 

  Preferences->advanced->browsing->show scroll marker

 > So if you do the obvious thing, say html { height:100%;
 > column-width:20em; }, repeated page-down will not put the column
 > tops at a constant offset from the viewport top. We could
 > special-case the page down amount for, say, scrollable elements
 > that use columns and the viewport when the root element is using
 > columns, but that's not a perfect solution.

Right.

 > In general, when you want to put columns on an element that isn't the only
 > child of a scrollable container, I think this is not going to work all that
 > well. I don't know what would, but that's a problem I'd really like to
 > solve.
 > 
 > Another issue is that you're likely to end up with some completely empty
 > columns at the end of the element, which will be ugly, especially if the
 > first column of the last horizontal set is not full. I suppose we could
 > balance the last horizontal set, but then should we balance the first
 > horizontal set if it's the only one?

Assuming you use 'max-height' to constrain the height (as opposed to
'height') why wouln't the last set of columns be short and balanced?
That is the default behavior:

  http://www.w3.org/TR/css3-multicol/#column-fill

 > I suppose that if we chose this behaviour, authors could still opt-in to the
 > horizontal overflow behaviour by setting the element with columns to be
 > width:intrinsic-width. That would be nice.

Yes, I can see some uses for it. In most cases where you really want
many columns, tables are probably better, though.

-h&kon
              Håkon Wium Lie                          CTO °þe®ª
howcome@opera.com                  http://people.opera.com/howcome
Received on Thursday, 16 October 2008 22:21:01 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:55:15 GMT