Formatting Page Layout Without Tables

"Sanders Kaufman" <bucky (AT) kaufman (DOT) net> wrote

Andy Dingley wrote:
On Tue, 21 Aug 2007 22:29:16 GMT, Sanders Kaufman <bucky (AT) kaufman (DOT) net
wrote:

But it works... and it works on all the browsers. So I don't see the
problem.

The problem is that it breaks the browser's ability to re-arrange the
page elements according to their best fit for a fluid design. It'll look
great once, for your computer. Anywhere else it probably looks dreadful.
In particular it is likely to fail if someone changes the size of text
relative to window size. Overlapping text onto adjacent elements is the
usual giveaway.

Ahh, okay - that's already dealt with.

URL?

There is a presumption in any design that if you tinker with it, it'll
break - unless it was designed to be so customizable.

Er, what?

So if the end-user turns off or modifies the CSS styles, they *should*
expect the page to appear broken.

Never heard of "gracefull degredation" have you?

Normally, I would look for a work-around, but I think if there was such a
thing, you guys would have recommended it.

Not for something so trivial as a "three column CSS layout", which has been
discussed here and elsewhere so many times. Eight hundred and sixty eight
thousand hits from google-)

What is it about this industry that there's so many nay-sayers?!

Fishes and being late for dinner.

--
Richard.

Sanders Kaufman wrote:

Gregor Kofler wrote:
Sanders Kaufman meinte:

Still - it would have been cool if someone could have addressed this
in the group, instead of just pointing to a web site.

Why? This extra-basic stuff. No need to post a 3-column page layout
with DIVs for the umpteenth time.

Gregor

Because this is a news group, not a web page... and this is news to me.

Assuming that people reading this group have a browser,
posting a link seems perfectly reasonable.

Rekeying would be dumb.

BugBear

Sanders Kaufman wrote:

Andy Dingley wrote:
On 18 Aug, 22:33, Sanders Kaufman <bu... (AT) kaufman (DOT) net> wrote:
I'm about to drop a decade-long, bad practice of formatting page layouts
with tables. I just got one question - how does that work in the real
world?

Read this newsgroup (and c.i.w.a.h and even alt.html). It's one of the
best resources around, good generally and excellent for the practice
of getting CSS good practice _right_, not just bearably competent for
one page on one browser. Search the archives. Search the archives for
tutorial recommendations.

Thanks - I found what I was looking for in the first place I looked...
after this one.

The crux of the answer was to use position:absolute.

If position:absolute achieves what you want,
you ought to stick with it.

Actually, if you're going to do that,
you may as well stick with tables.

BugBear

Am 22.08.2007 02:00 schrieb Andy Dingley:

On Tue, 21 Aug 2007 22:29:16 GMT, Sanders Kaufman <bucky (AT) kaufman (DOT) net
wrote:

But it works... and it works on all the browsers. So I don't see the
problem.

The problem is that it breaks the browser's ability to re-arrange the
page elements according to their best fit for a fluid design. It'll look
great once, for your computer. Anywhere else it probably looks dreadful.
In particular it is likely to fail if someone changes the size of text
relative to window size. Overlapping text onto adjacent elements is the
usual giveaway.

even if you use
position: absolute;
left : 10em;

etc., i.e. no absolute dimensions? I'm using position: absolute with
dimensions specified in ems quite a lot, and it scales ok in FF and IE
(it's an intranet application, so these two are all the browsers I'm
interested in).

--
Christian

bugbear wrote:

Sanders Kaufman wrote:

The crux of the answer was to use position:absolute.

If position:absolute achieves what you want,
you ought to stick with it.

That's poor advice to give to a CSS newbie. Any positioning method
should be avoided until you understand what it is *supposed* to do, the
pros and cons of its use, and how to deal with issues like scalability.

Many ready-made templates have addressed these already, but not all. And
an inexperienced person should never just pick a method because it seems
to work (for them in their particular browsing environment). Chances are
it will break in different circumstances.

Learn something about the technology you are attempting to use before
hacking away and making a mess.

Actually, if you're going to do that,
you may as well stick with tables.

ick

--
Berg

"Bergamot" <bergamot (AT) visi (DOT) com> wrote

bugbear wrote:
Sanders Kaufman wrote:

The crux of the answer was to use position:absolute.

If position:absolute achieves what you want,
you ought to stick with it.

That's poor advice to give to a CSS newbie. Any positioning method
should be avoided until you understand what it is *supposed* to do, the
pros and cons of its use, and how to deal with issues like scalability.

Many ready-made templates have addressed these already, but not all. And
an inexperienced person should never just pick a method because it seems
to work (for them in their particular browsing environment). Chances are
it will break in different circumstances.

Learn something about the technology you are attempting to use before
hacking away and making a mess.

Actually, if you're going to do that,
you may as well stick with tables.

ick

Bergamot, I read the bugbare post as if it were:

<tongue position=cheek attitude=firmly>
<sarcasm level=high pity=medium>
<post>...</post>
</sarcasm>
</tongue>

FWIW I agreee ;-)

--
Richard.

On 22 Aug, 13:15, Christian Kirsch <c... (AT) bru6 (DOT) de> wrote:

I'm using position: absolute with
dimensions specified in ems quite a lot, and it scales ok in FF and IE

If position:absolute; works for you, then fine, use it.

My concern is that a newbie goes from tables to position absolute in 5
minutes flat, then tells us all that we're wrong. Now as
position:absolute; certainly does have a whole pile of known ways to
break your site associated with it, then that's a very strong hint
they've almost certainly built a broken page that's non-fluid.

If you know about these risks, and you've used absolute within the
limited constraints that won't cause this, and you've tested to see
that you've got it all right, then it's fine to use it. I don't think
the OP has got that far.