Re: CSS multi-column layout (was Re: From CSS to DSSSL)

At 8:33 PM -0500 4/17/97, Liam Quinn wrote:
> On 17 Apr 97 at 15:23, Todd Fahrner wrote:
> > (if only css
> > could accommodate multi-column layout....)
> What about the following, using CSS Positioning:
> div.col1 { position: absolute; left: 1%; width: 31% }
> div.col2 { position: absolute; left: 35%; width: 48% }
> div.col3 { position: absolute; left: 70% }
> and then
> <DIV CLASS=col1>
> <!-- Column 1 -->
> </DIV>
> <DIV CLASS=col2>
> <!-- Column 2 -->
> </DIV>
> <DIV CLASS=col3>
> <!-- Column 3 -->
> </DIV>
> Or is that not what you meant?

Not quite what I meant. Unlike nsml's <multicol>, this requires markup to
terminate/begin each column. Not good - columns would not balance properly
in many cases, and I don't need to tell *you* about the drawbacks of
intensive presentational markup, graceful degradation or no. Like
<multicol>, however, columns are likely to be clipped if the canvas is too
short. It's one thing to have to scroll down; another to scroll up *and*

I am imagining a "pseudo-paged" model in which text flows from one box (for
lack of a better word) to another, where the box dimensions and positions
relate geometrically to the aspect/area of the window, taking height into
account as well as width. Somewhere would be some unobtrusive (or
invisible) mechanism to queue up (or scroll) the next segment. The "keep
together" and "break before" bits from the Printing Extensions draft would
be helpful here, which is another reason I don't want to see "print" and
"screen" as media selectors, but instead descriptions of the rendering
environment in terms of aspect, dpi, color model/depth, etc. (A printer is
just a tree-powered screen with a very low refresh rate, a high resolution,
and a troublesome color model, after all).

Todd Fahrner

The printed page transcends space and time. The printed page, the
infinitude of books, must be transcended. THE ELECTRO-LIBRARY.

--El Lissitzky, 1923