Getting mad about CSS - Example from Bulletproof Webdesign

[Thomas Bügel]

Hi!

Out of the book from Dan Cederholms (Bulletproof Webdesign - german ed.) I
tested chapter 4 (creativ floating). For no CDR is included I wrote the code
from the text in the book (images and according text is taken from
tv-spielfilm.de). But it don't work for me in the same way as Dan tells in
the book.
I tested it with different browsers on Windows 2000. It should show three
pictures with a white frame (build though the padding 4px) and a closing
border with 1px and an according text with a heading on top of the text
positioned opposite of the picture neither left or right. Dan uses
Definition lists for the markup.
Now it looks allmost as it should in Opera 7.1, Netscape 6,7.1+8.0 and
Firefox. Only the heading is moved appart from the text. So that not the
heading is positioned on top only as it should.
But in IE 6.0 allmost nothing is correct: The main div is not positioned in
the middle of the browsers window as it should with margin-left/right: auto.
The fontsize is one step bigger as in the other browsers (to big). The white
border of the images is missing (ignoring padding 4px). Its awfull! Is there
a hack that IE6 needs to works correctly? And whats wrong with the heading?
Is there anyone, who please could test the example?
Here is the complete example including the CSS:

http://www.asys-it.de/webtest/defliste_o.html

Thanks for help! Tom

[Thomas Bügel]

"Thomas Bügel" <no****@am3d.de > schrieb im Newsbeitrag
news:44******** **@news.arcor-ip.de...

Only the heading is moved appart from the text. So that not the
heading is positioned on top only as it should.
I had a lttle mistake in the css declaration! The position of the heading is
now correct!
But in IE 6.0 allmost nothing is correct: The main div is not positioned in the middle of the browsers window as it should with margin-left/right: auto. The fontsize is one step bigger as in the other browsers (to big). The white border of the images is missing (ignoring padding 4px). Its awfull! Is there a hack that IE6 needs to works correctly?

The faults of the IE6 are further valid!

[Thomas Bügel]

OK! I discovered the fault!
I'm using Golive CS2 for building websites. Golive writes in the first line
of new XHTML-pages: <?xml version="1.0" encoding="iso-8859-1"?> and that
line makes that IE6 renders the page in the wrong way as I wrote it in my
posting from yesterday. I discovered that by testing of different settings
of that page, but don't know neither why Golive CS2 sets that code in the
first line of a new XHTML document nor why IE6 don't like it (the other
browser ignore it). Has someone an explanation?

With regards - Tom

[Chris Lieb]

Thomas Bügel wrote:

OK! I discovered the fault!
I'm using Golive CS2 for building websites. Golive writes in the first line
of new XHTML-pages: <?xml version="1.0" encoding="iso-8859-1"?> and that
line makes that IE6 renders the page in the wrong way as I wrote it in my
posting from yesterday. I discovered that by testing of different settings
of that page, but don't know neither why Golive CS2 sets that code in the
first line of a new XHTML document nor why IE6 don't like it (the other
browser ignore it). Has someone an explanation?

With regards - Tom

XHTML is an application of XML. XML requires an XML declaration as the
first line in any conforming document. Since XHTML is an application
of XML, it must also follow this rule. The problem with this is that
IE6 incorrectly renders XHTML pages that have an XML declaration as
their first line.

Chris

[Johannes Koch]

Chris Lieb wrote:

XML requires an XML declaration as the
first line in any conforming document.

In some circumstances, but not always.
--
Johannes Koch
Spem in alium nunquam habui praeter in te, Deus Israel.
(Thomas Tallis, 40-part motet)

[dingbat]

Chris Lieb wrote:

XML requires an XML declaration as the first line in any conforming document.

Untrue - and distinctly harmful in the case of XHTML.

An XML declaration is only required in XML where it's used by a
side-effect to sniff out the encoding of the document. On the web you
have HTTP and HTTP headers, so there's never any need to sniff for
this.

Don't use it for XHTML.