Using CSS positioning for layout?

What is the current school of thought in using CSS-P? As I understand it current browsers offer pretty good support for it. Is it to soon to be using it as a primary layout method? Dreamweaver MX seems to offer an easy method the implement CSS-P and the code is straightforward. I was thinking of investing some time to learn to use it properly, but I?m not sure if browsers are ready yet.

Here is a web site that uses CSS-P. You tell us if you think current browsers are ready to support it. :-)

http://www.meyerweb.com


MiKE.
Looking for freelance work | (email address temporarily removed)
$i = "fun"; while($i > $possibilities) { $code; $code++; }

or http://www.csszengarden.com/

"mzanime.com" <webforumsuser (AT) macromedia (DOT) com> wrote

Here is a web site that uses CSS-P. You tell us if you think current
browsers are ready to support it. :-)

http://www.meyerweb.com

MiKE.
Looking for freelance work | (email address temporarily removed)
$i = &quot;fun&quot;; while($i > $possibilities) { $code; $code++; }

If you go back one browser version (IE 5.0 for example) does the page display with a ton of errors?

Actually I guess I should go get 5.0 and find out.

"aca_dia" webforumsuser (AT) macromedia (DOT) com wrote:

If you go back one browser version (IE 5.0 for example) does the page display with a ton of errors?

Actually I guess I should go get 5.0 and find out.





Its not that simple at the moment given the many different hacks

required to support css-p x-browser.

I would learn it so you are in a strong and advanced position to use it
when browser support becomes more stable.

Simple tables styled with css is still a more stable option to use at
present and will take you less time to implement.

Also if you look at the two url examples provided they fall short if you
need to support NN4x visually. Plus some examples in the zen garden are
not well constructed, they just look pretty and serve no real world purpose.

Its not a case of go for it and use every css-p positioning attribute in
the book. Its a case of determining if its reliable for the job you need
it to do. That takes some thought and planning.

Only fools rush in.

aca_dia wrote:

If you go back one browser version (IE 5.0 for example) does the page
display with a ton of errors?

Actually I guess I should go get 5.0 and find out.

Eric Meyer's brilliant Slantastic blows apart with IE 5 & 6. It is an example
of the problems that can arise when using CSS (although I wouldn't actually
call it CSS positioning in this case):
http://www.meyerweb.com/eric/css/edge/slantastic/demo.html

I found a workaround that largely solved it, but when I tried to use it to
have rounded corners using CSS but no images at all, it blew apart a little
with IE 5 on W2000 and a lot with IE 5.2 on Mac. Once again not really CSS
positioning:
http://www.barry.pearson.name/articles/corners/index.htm

And I found an interesting bug with IE 6 on one of the my CSS positioned
layouts that I haven't found any reference to before:
http://www.barry.pearson.name/articles/layout_tables/ie6_bug.htm

It is scary knowing that a bug in some browser somewhere can actually mean
that parts of your page can disappear!

--
Barry Pearson
http://www.Barry.Pearson.name/photography/
http://www.BirdsAndAnimals.info/
http://www.ChildSupportAnalysis.co.uk/

aca_dia wrote:

What is the current school of thought in using CSS-P? As I
understand it current browsers offer pretty good support for it. Is
it to soon to be using it as a primary layout method? Dreamweaver MX
seems to offer an easy method the implement CSS-P and the code is
straightforward. I was thinking of investing some time to learn to
use it properly, but I?m not sure if browsers are ready yet.

It can be very hard to start with a layout concept and make it work properly
with CSS positioning. Yet there are many very attractive and workable
CSS-positioned layouts.

I've been trying to see how these can both be true. I have a hypothesis, and
I'm currently looking at pages "out there" to see if it is true. In summary, I
hypothesise that there are two dramatically different approaches to web page
layout:

1. One approach is that used by visual-design experts, perhaps experts in the
way human beings explore the visual field and extract information from it.
These people may have generic expertise, that applies to the vision-scape in
general, or expertise that applies to to paintings, to photographs, or to the
specific case of web pages. The typical attitude of such a person is "X needs
to be here, Y needs to be there". In spite of what some claim, I don't believe
the current CSS standards were ever intended to handle those cases. It isn't
just a problem with browser support. If you set out to specify CSS handle
those cases, you would end up with a different CSS from what we have.

2. The opposite approach is that used by experts in CSS positioning. These
people know what can be achieved with CSS positioning, and how to do so. The
typical attitude of such a person is "X can be positioned here, Y can be
positioned there".

In other words, if you start with a layout concept unconstrained by
implementation concerns, you may well end up with a layout that cannot easily
be handled by CSS positioning. Our brains often work best with horizontal and
vertical lines, and for those a simple layout table may be by far the best way
of translating that layout concept onto the web. Just have a look at the many
attempts to take the 5-box 3-column layout and do all the things with CSS
positioning that are trivial with a layout table.
http://css-discuss.incutio.com/?page=ThreeColumnLayouts

But if you want are willing to be steered by what CSS positioning can and
can't do, and are willing (say) to have background colours of columns that
match what CSS positioning, eg. using floats, can do, then you can typically
go with CSS positioning. If you pick up someone else's implementation of (say)
a 2-column layout that has already been tested on various platforms in various
browers, you may be safe even if you don't test it in all combinations
yourself.

Until we get a proper page layout language, we are stuck with 2 techniques,
neither of which was really intended for page layout: simple layout tables,
and CSS positioning. I have previously posted here:

- Understand the strengths & weaknesses of simple layout tables and CSS
positioning. This includes understanding what you can do yourself, not just
what a guru can do.

- Keep them both in your toolkit - like having both a hammer and a
screwdriver. And recognise that sometimes you might use a bit of each, in a
hybrid layout. If you use a layout table to handle 3 columns, that doesn't
mean you need to continue to use a layout table to handle what is in the
columns. They are totally separate decisions. And vice versa. Step back from
the holy war.

- Don't feel guilty whichever way you go. If you use simple layout tables,
the sky doesn't fall, you don't destroy the web, and you won't lose all your
blind users. If you use CSS positioning, at least your content will all be
there even if it looks a bit strange in some browser on some platform that you
haven't managed to test on.

- Don't ever try to use CSS positioning instead of tables where the content is
in a plausible sense "tabular data". That would be very silly use of CSS. (But
recognise that there are some other good tools too, such as various types of
list, including nested lists).

- Whatever you do, use CSS anyway. That is the single "given" - use CSS. In
case the message didn't get across - use CSS. Even when using layout tables,
keep them very simple and control them with CSS.

--
Barry Pearson
http://www.Barry.Pearson.name/photography/
http://www.BirdsAndAnimals.info/
http://www.ChildSupportAnalysis.co.uk/

The current school of thought on CSS-P has changed dramatically since the
below web page was published.

Tables versus Full CSS Integration
http://www.decloak.com/Dev/CSSTables/CSS_Tables_01.aspx
(i.e. replacing all <table> tags with <div> tags and CSS positioning)

Some of the most die-hard pro-CSS people are either still in their denial
stage OR have turned 180 degrees around because of it.




"aca_dia" <webforumsuser (AT) macromedia (DOT) com> wrote

What is the current school of thought in using CSS-P? As I understand it
current browsers offer pretty good support for it. Is it to soon to be

using it as a primary layout method? Dreamweaver MX seems to offer an easy
method the implement CSS-P and the code is straightforward. I was thinking
of investing some time to learn to use it properly, but I?m not sure if
browsers are ready yet.

dc wrote:

The current school of thought on CSS-P has changed dramatically since
the below web page was published.

Tables versus Full CSS Integration
http://www.decloak.com/Dev/CSSTables/CSS_Tables_01.aspx
(i.e. replacing all <table> tags with <div> tags and CSS positioning)

Some of the most die-hard pro-CSS people are either still in their
denial stage OR have turned 180 degrees around because of it.

That's just not true. There are idiots blabbering about CSS as if it
were religion and there are idiots blabbering about tables as if they
were the antidote for lack of CSS understanding. The truth is somewhere
in the middle. Fact is, no one will learn anything unless they are
taught reasonable lessons by reasonable people. Creating an adversarial
atomosphere is not conducive to learning (or to making a business
successful).


--
Al Sparber
PVII
http://www.projectseven.com

I second that.

Alex
"Al Sparber- PVII" <adSTRApmin (AT) projectseven (DOT) com> wrote

dc wrote:
The current school of thought on CSS-P has changed dramatically
since
the below web page was published.

Tables versus Full CSS Integration
http://www.decloak.com/Dev/CSSTables/CSS_Tables_01.aspx
(i.e. replacing all <table> tags with <div> tags and CSS
positioning)

Some of the most die-hard pro-CSS people are either still in their
denial stage OR have turned 180 degrees around because of it.

That's just not true. There are idiots blabbering about CSS as if it
were religion and there are idiots blabbering about tables as if they
were the antidote for lack of CSS understanding. The truth is
somewhere
in the middle. Fact is, no one will learn anything unless they are
taught reasonable lessons by reasonable people. Creating an
adversarial
atomosphere is not conducive to learning (or to making a business
successful).


--
Al Sparber
PVII
http://www.projectseven.com