Re: CSS1 and tables

On Oct 6, 10:51pm, Terry Crowley wrote:

> >  OK, so these browsers do error correction on a document which has
> >  a block  level element inside a phrase level element, and internally
> >  generate

> I believe you give the internal implementation of the early browsers too much
> credit.

No, but perhaps I should have made more use of the phrase "or behave as
if they do so". I have no illusions about the codebase of early browsers,
having used XMosaic since its first beta and Netscape since version 0.87.

However, the browsers are clearly shifting to a more formally correct
tree-oriented model since the cost of only pretending to operate like that
while trying to support document object models and selector mechanisms is
greater than the cost of doing it in a more straightforward fashion. The
"speed hacks" of old have become millstones around the necks of developers.

> Notice for example how the color attribute on
> the <FONT> tag in NS3 leaks into the table, but neither size nor face do.  If
> that's by design, it's not a pretty one.  What we're living with is the
> of bad code.

Sure. But the code is getting better. IE4 is lots better than IE3 in terms
of CSS compliance; NS4 is a first implementation of CSS and I have
confidence that compliance will also improve in subsequent releases. It
is unrealistic to expect companies to deliver a 100% compliant application
with their first bite at the problem.

Chris Lilley, W3C                          [ ]
Graphics and Fonts Guy            The World Wide Web Consortium              INRIA,  Projet W3C                       2004 Rt des Lucioles / BP 93
+33 (0)4 93 65 79 87       06902 Sophia Antipolis Cedex, France

Received on Tuesday, 7 October 1997 08:25:28 UTC