Re: [css3-multicol] propose "column-width:minimum"

[snip]

> Yeah now that you mention it, I think min-width:???em is what designers
> want in this case. This is not the same as setting column-width:???em,
> because column-width can not:
>
> 1. expand or contract the column width (constrained to minimum or
> min-midth) so as to maximize total width while minimizing height. Note we
> might need a column-max-width then too?


What I am trying to eliminate is the unused space between the inline
direction edge of multi-column element and the last column in that
direction.  The column-width setting prevents the browser from expanding
or reducing the column-width to maximize the use of content 2D real
estate.

The more I think about this, reducing column width is self-defeating,
because the content height probably increases.

The main thing trying to be achieved is to give a target column-width, but
let the browser shrink or expand this (subject to any min-width minimum
and content width constraints) so as to optimize use of content 2D real
estate.

Thus I no longer see much value in min-content.  Rather I want a setting
to make column-width flexible.


> 2. the min-width is overriden by the content itself. Btw, this is why is
> very important to be able to wrap long text sans spaces (e.g. urls).

Actually this applies to column-width too?  I forgot what happens if the
content overflows the inline direction of the column-width?  Too busy to
go figure it out.

Someone is going to need to write a user-friendly guide on CSS columns.


>
> Tangentially, I wish there was some CSS support for automatically
> displaying long unwrappable text shortened with ellipsis (that expand on
> hover or something like that) depending on the need to avoid overflow in
> the dynamic layout?
>
>
>> plus a "width:min-content" rule to ensure that the
>> column doesn't get smaller than the largest word / replaced element.
>
> I thought that was always the case?  Or is that just the default?  I
> wasn't aware of and haven't yet studied that setting. Will do later.
>
>
>

Received on Friday, 15 October 2010 16:33:10 UTC