Re: How is it possible to devise such a feeble system?

--- Vadim Plessky <> wrote:
> On Wednesday 24 October 2001 21:05, Sampo Syreeni
> wrote:
> |   On Wed, 24 Oct 2001, Jesse McCarthy wrote:
> |   >I was referring to a legitimate way of doing
> it, without involving
> |   >tables at all.  See this document:
> |   > .
> |
> |   How are CSS tables not legitimate? AFAIK, the
> legitimacy issue (at least
> |   as far as WAI goes) is about using HTML tables,
> and not about marking
> |   something as a table-cell in CSS.
> In CSS3, TABLEs is just one module.
> Browser can claim CSS3 conformance while not
> supporting tables at all.
> BTW: I think this would be really great idea.
> Code handlings tables and all hacks around it can
> take easily 20%-20% of all 
> you *rendering* code.
> TABLEs is evil!  We should avoid them, either in
> HTML or in XML.
> DIV/SPAN, with display: block, inline, inline-block,
> is much more clean way 
> to do formatting.
> Especially in simple cases, like Jesse exposured.
> See also my testcase.

Tables are not evil.  A table is a valid structure for
describing data that is correlated in two-dimensions. 
Lists are a similarly valid structure for
1-dimensional correlations.  3 and greater dimension
correlations unfortunately are not easily described in
2-d, so we usually take a cross-section that shows two
variables with all other parameters held constant...
also know as a table.

It is considered bad practice to use tables to layout
a web site, but not to describe table data.  Support
for tables should not be discouraged because web
develoeprs have resorted to exploiting them when
browser vendors struggled to provide a non-buggy
alternative.  There is a valid need for a table data
structure, just as there are needs for lists.

-Dylan Schiemann

Do You Yahoo!?
Make a great connection at Yahoo! Personals.

Received on Wednesday, 24 October 2001 18:30:32 UTC