Re: Proposal: :column pseudo-class

On 7/2/05, Lachlan Hunt <lachlan.hunt@lachy.id.au> wrote:
> Orion Adrian wrote:
> > Couldn't this problem simply be solved by removing the presentational
> > attributes "colspan" and "rowspan".
> 
> They're not presentational at all.  They're structural, and it is not
> always semantically correct to simply duplicate the values in each cell
> in place of colspan and/or rowspan.  For example, marking up a timetable
> of classes.  The time of the day along one axis, and the days of the
> week along the other.
> 
> ____|_09:00_|_10:00_|_11:00_|_12:00_|_13:00_|_14:00_|_15:00_|_16:00
> Mon | Maths |    English    |Science| Lunch |   Computers   |
> Tue | ...
> 
> Notice how some classes, English and Computers, span two hours each.
> They're not two seperate consecutive 1 hour classes — they're both 2
> hour classs.  To mark that up without colspan by duplicating the name in
> each cell would be semantically incorrect.

Why would it be semantically incorrect? Each is a label for a
parcticular class. In high school when I had my 2-period classes
listed out, they were listed as two instances of the same class.
Presentation can be used to improve the accessibility of semantics (in
readability for example). It's not just done with size, but also with
spacing and organization.

I would however propose a series of properties (which I'm working on),
that would show adjecent cells as one cell, but leave the underlying
data alone.

This is very much presentation to me and removing colspan and rowspan
open a whole range of behaviors and possibilities.

Orion Adrian

Received on Saturday, 2 July 2005 18:25:02 UTC