4510, FIXED, finally! Simple test case.
Nice to see the current behavior fixed to match CSS2.1. However, is there any possiblity of the CSS behavior getting restructured so that any other properties than
width apply to table columns?
The problem here seems to lie in that table column elements do not contain their cells and thus cannot be inherited from by the table cells. Is there any method of fixing this?
The W3C wants "us" (I think Ian Hickson asked this on his weblog) to provide solutions for that problem.
Yes, I remember that post. In fact, it was what prompted my question. I can see someone making an XML language where table row and columns use exactly reverse of how XHTML handles them, or one where both rows and columns are specified separately, not containing any of the cell elements, where the cell elements are just listed in a flat list, depending on the rows and column tags for determining position in the table. or maybe one establishing a grid of rows and tables, and specifying cell placement by position and size relative to the grid. None of these would inherit the same way HTML tables does, and neither would they really work well with the current CSS tables model. Which leads mo to one conclusion: For tables, it would be useful to explicitly specify inheritance not based on element hierarchy relations, but on table model relations, somthing like this:
table-row table-row-group table-column table-column-group table