W3C home > Mailing lists > Public > www-style@w3.org > June 2007

Re: Stylings only possible with Tables

From: Spartanicus <mk98762@gmail.com>
Date: Sun, 24 Jun 2007 19:12:11 +0100
Message-ID: <n2m-g.dv8t735h033jfhuach07h8mbmfdb75u15i@4ax.com>
To: www-style@w3.org

Daniel Beardsmore <public@telcontar.net> wrote:

>> CSS 2.x doesn't really address layout (a complex issue).
>
>I don't think layout need be that complicated at all. CSS is the 
>Complicator's Styling System and we need to try to move away from this. 
>Remember how simple frame layout was? That's all we need. A lot of layout is 
>conceptually trivial and should be able to be expressed just as easily on a 
>computer.

Creating a comprehensive and flexible layout mechanism that is able to
work well in a media independent manner is a complex task. I'd agree
that the resulting complexity of the mechanism potentially forms a
problem given that most authors only want to create a layout grid that
looks ok on their screen. If the mechanism is too complex it will be
ignored by many, but for those unsophisticated authors there are HTML
tables.

Imo that does not mean that a more sophisticated mechanism shouldn't be
developed for those that are willing to study and use it.

>Has anyone considered, though, whether inline blocks could be used for 
>columns? It's not widely supported (just iCab and KTHML/WebKit I think) but 
>might work somewhere :)

inline-block cannot be used to create auto scaling columns.

For what it is worth, inline-block is supported by most modern browsers:
IE5.5 - 7, Opera7+, WebKit derivatives, iCab, Konqueror, sadly with the
exception of release quality Gecko based browsers.

>However, I remind you that the CSS3 Advanced Layout Module is virtually left 
>for dead. It seems like no-one actually cares about advanced layout, yet I 
>am sure most real-world CSS users would be delighted if they could have a 
>mechanism for just getting layout to work in a simple, sane way. Surely the 
>powers-that-be are still adequately in touch with the real world enough to 
>see how pressing this is?

I agree that this issue has received to little attention and too late.
The complexity of a comprehensive solution probably played a part in
that. Plus it suffers from the lack of development momentum that imo CSS
in general suffers from. My guess is that implementors aren't keen to
contribute resources to create solutions if there is an existing method
that works. HTML tables used for layout isn't really broken to many, the
problems that arise from that type of usage mainly feature in the
awareness of the few.

I think authors (myself included) also share a certain blame for the
current state of affairs. I noted the lack of reviews from authors of
the Advanced layout proposal when it was published. As you note it is
one of the issues most struggled with in practice today, so why aren't
we authors showing more interest? For me it is a lack of time that would
be required to review the proposals properly.

>Well, assuming Internet Explorer would go wander off and die and leave 
>people to only code for vaguely standards-compliant browsers.

Layout is a fundamental issue to web development, the lack of a CSS2
mechanism means that authors won't be able to use a comprehensive CSS
solution for many, many years even if one were published today.

Yes currently authors need to work around a significant, although
decreasing number of implementation bugs and lack of support, but the
more skilled amongst them can manage today, although I'll gladly agree
that it shouldn't be necessary to have the level of skill and experience
current implementations require.

>As long as we have to write code to run on browsers whose CSS implementation 
>never got past a partial implementation of CSS2 (i.e. IE) then all the 
>innovation in the world won't save you.

IE is being developed again, so things will improve in time.

>Where do we go from here?

Things will improve but it will take a long time. In the mean time
skilled developers can use CSS for non columnar layouts, and for
columnar layouts the use of a simple HTML table used for layout isn't
the huge problem some proclaim it to be.

-- 
Spartanicus
Received on Sunday, 24 June 2007 18:11:43 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:51 GMT