Can anyone tell me why this works in Firefox but not in IE ?

[Sunday8pm]

Here is the problem website

The tables are really messed up... so other than telling me that IE is shit can anyone come up with a techinical explanation about what might be wrong?

[Androme]

Your coding's shit.

Open up Dreamweaver and re-code from scratch.

[LHC]

Why would Sunday be making a website in whatever language that is?

[sayonara]

From a cursory glance I notice the following:

- You have not specified a DOCTYPE so browsers are forced to guess which spec they are meant to use.

- Some of your attributes are not quoted, which is no longer valid HTML.


I'll look more closely in a sec, just busy elsewhere for the moment.


ps - those rollovers would be less resource-heavy and more accessible if you used CSS. Grab the source from my homepage or check out the horizontal version on ALA.

[Sunday8pm]

Leshy has taken a bash at this on # forums

Have him add # before all color values in the CSS.

Was his answer.

As for the website itself, some guy with crappy gaelic was originally hired to do the site and wrote up a template, they got rid of him and handed a cdr of what had been done already to my mate who now has this abomination on his hand.

He was thinking of going from scratch on dreamweaver again but asked me if I knew anyone.

[Sunday8pm]

Sayonara what would be your homepage?

[LHC]

www.sayonara.info

p.s. where is the banana page?

[Sunday8pm]

Banana?

Appreciating all this input very much btw.

[sayonara]

The banana page?

That's not been there for yeeeeaaaaaaaars (well, probably at least 1 year)

[sayonara]

Ahhhh...

Those library items contain link tags that fetch the CSS file. That is probably not messing up the page but it is still going to be screwing with IE's parser in some way.

[edit] also there is some invalid nesting, such as:

<span class="body"><br></span>

[Leshy]

I reckon it was the border: 2px 336699 solid; that sodded things up in MSIE.

Having said that, both the CSS and HTML files are horrible, and a recoding would probably be in the best interest of future maintenance.

[Androme]

Originally Posted by Leshy

both the CSS and HTML files are horrible, and a recoding would probably be in the best interest of future maintenance.

I recommended this but was ignored.

Popularity bites

[Leshy]

Originally Posted by sayonara on his website

So whereas you would once have done this:

<a href="page.htm#section1">Jump to Section 1</a>

... some page content...

<a name="section1"></a>
<h1>Section 1</h1>


...you now do this:

<a href="page.htm#section1">Jump to Section 1</a>

... some page content...

<h1 id="section1">Section 1</h1>

I was painfully unaware of this. Muchos gracias, and I shall be removing several anchors from my page shortly

[sayonara]

Originally Posted by Leshy

I was painfully unaware of this.

I know, it's so very simple yet so shockingly absent from any of the specs, tutorials, blogs etc I have seen, including all the endless pages of "oh wtf do I do with these anchors?" threads.

I was pretty chuffed when I figured that out

[meglamaniac]

Originally Posted by Leshy

I was painfully unaware of this. Muchos gracias, and I shall be removing several anchors from my page shortly

Thats probably because afaik it IS still valid HTML (at least it's still valid HTML 4.01 Transitional, which is what most browsers render to unless told otherwise by a doctype).
What it ISN'T is valid XHTML, which is what is being brought in now (it's an XML specified version of HTML).

In XHTML, all attributes must be quoted, all tags are specified in lower case, and all tags must be closed.
So HTML's <br> becomes <br />
and <img src="blah.jpg" alt="something"> becomes <img src="blah.jpg" alt="something" />
and so on.

The id attribute has been introduced mostly to aid with integrating CSS 2.0 into XHTML pages, but it also replaces the name attribute (although at the moment using both is acceptable).
The class attribute still works too, so:

p.special { definitions }
would apply to <p class="special">

and
p#special { definitions }
would apply to <p id="special">

and so on.

http://www.w3schools.com/xhtml/default.asp
http://www.w3.org/TR/xhtml11/

[Androme]

In old browsers you still have to use id & name to ensure maximum compatibility with your users but it's not necessary since most people don't use ie: MSIE 5.5 etc.

www.alistapart.com is an excellent site - offering tutorials on CSS & XHTML practically most other sites don't - I keep advertising it because it is actually very good and too few people know about it

[Leshy]

Originally Posted by meglamaniac

What it ISN'T is valid XHTML

The W3C validator does not report any errors on XHTML 1.0 pages containing classic anchor links, which - combined with the absence of any word about XHTML anchors everywhere - is why I was unaware of these having been replaced by the method mentioned above.

As for classes and id's, I am quite well aware of how they work - you may want to check the link in my signature

[sayonara]

Originally Posted by meglamaniac

Post #15

Errr... if it was a problem for Leshy, it's because he needed the accessible XHTML solution.

Originally Posted by Androme2

www.alistapart.com is an excellent site

Indeedeo.

[Leshy]

Originally Posted by sayonara

Errr... if it was a problem for Leshy, it's because he needed the accessible XHTML solution.

Not as much an actual problem - the anchor links were functional - as a desire to not do things half-assed

[sayonara]

Originally Posted by Leshy

Not as much an actual problem - the anchor links were functional - as a desire to not do things half-assed

OK, how about this:

"Errr... if it was a desire of Leshy, it's because he wanted the accessible XHTML solution. "

?

[Leshy]

Now that I'm actually awake, I'll just go with what you said

[djbass]

Everything it mentions in here needs fixing

[sayonara]

Originally Posted by djbass

Everything it mentions in here needs fixing

Like I already said, there's no DOCTYPE specification so the rules that the validator is using are fairly arbitrary. From the look of it the original author was using HTML 3 or 4.1 Frameset.

Other than the randomly-placed link tags, which I already mentioned, there's nothing in that report that would screw up IE's parsing so badly.

[at0mic.c0w]

also whoever wrote this is an idiot for not testing it in IE, rewrite = the win.