Sent by Ian Hickson on 3 December 2003 02:02
On Tue, 2 Dec 2003, Aleksandar [WINDOWS-1250] Vaci=E6 wrote:
>
> Now...if you haven't found the solution for 4 years, I can hardly find
> it for one evening :). For what you said about W3C proposed flow and
> Microsoft's way of doing it, I could suggest that you combine the two
> worlds - insert another step between yours 3 (layout doc) and 4 (paint
> doc) which will do additional CSS property setup, depending on the
> layout and relations between elements.
>
> This sounds rather easy to say, which is why I think you have probably
> thought about that. If you have, what was the reason against it?
The reason against that solution is that it is a performance nightmare.
User agents do not want to add another step to the rendering process,
since doing so would add tens of milliseconds to page load times, at a
time where they are all trying to reduce page load times by hundreds of
milliseconds.
It's also rather hard to envisage how it would look in the CSS.
We may be doing some work on the CSS3 Tables module shortly after the
CSS2.1 specification reaches Candidate Recommendation stage. I'll see if
we can find a solution to this problem once and for all, even if it does
require a special case syntax just to handle columns. I realise that a lot
of people want this desperately.
> So, in this case, IE's way of doing it is God send.
Of course, if you were to remove the tables and replace the markup with
more semantically correct markup (e.g. <h1>, etc), and then tried to make
the same layout happen, using display:table, display:table-row, etc, as
CSS was intended to allow, it would fail dismally in IE because IE doesn't
implement that feature. And it's only because it doesn't implement that
feature that it _can_ implement the styling columns thing. :-)
--=20
Ian Hickson )\._.,--....,'``. fL
U+1047E /, _.. \ _\ ;`._ ,.
http://index.hixie.ch/ `._.-(,_..'--(,_..'`-.;.'
______________________________________________________________________
css-discuss [EMAIL-REMOVED]]
http://www.css-discuss.org/mailman/listinfo/css-d
Supported by evolt.org -- http://www.evolt.org/help_support_evolt/