- From: Håkon Wium Lie <howcome@opera.com>
- Date: Tue, 8 Feb 2011 22:19:54 +0100
- To: Alex Mogilevsky <alexmog@microsoft.com>
- Cc: "www-style@w3.org" <www-style@w3.org>
Also sprach Alex Mogilevsky:
> Users already don't see any content when the normal formula hits
> zero-with (unless content is positioned), and we've just agreed
> it's ok...
True.
> With this
> (17) if ((column-count - 1) * column-gap < available-width) then
> (18) N := column-count;
> (19) W := (available-width - ((N - 1) * column-gap))/N;
> (20) else
> (21) N := column-count;
> (22) W := 0;
> (23) column-gap := (available-width/column-gap);
> (24) fi
>
> It goes like this:
>
> column-gap: 49; column-count: 3; width: 100; /* 3 columns of 0.66 width */
> column-gap: 50; column-count: 3; width: 100; /* 3 columns of 0 width */
> column-gap: 51; column-count: 3; width: 100; /* 3 column of 0 width, 50 gap */
> column-gap: 100; column-count: 3; width: 100; /* 3 column of 0 width, 50 gap */
> column-gap: 100; column-count: 3; width: 50; /* 3 column of 0 width, 25 gap */
Right. I can live with this. The upside, compared to your other
suggestion, is that there are no dramatic phase changes; once columns
get to zero width they stay there (as the window narrow).
In pracice, this would probably be discouragement to using
column-count (at least for values of, say, 4 or more) and a
corresponding encouragement to use 'column-width'. That's good, I
think.
-h&kon
Håkon Wium Lie CTO °þe®ª
howcome@opera.com http://people.opera.com/howcome
Received on Tuesday, 8 February 2011 21:20:36 UTC