Um...all the options on the main menu such as main, problems, contests, etc are vertical and blend into the main page.

like this at the top of the page. Not sure if it's intended.
.main
.blar
.balrr
.xglbfr

Browser?

Chrome

I think jargon had the same problem? He was using the beta developer version of Chrome.

Anyway, the menu is a CSS hack according to Hanson and appears to be tailored to Firefox. So I'll try to get this fixed (so we can add the valid CSS icon too, lol) and then it should be browser-independent.

Try refreshing the cache with ctrl+f5? (In the future please post your browser/version whenever reporting an error)

It happened to me ONCE (on the latest version of Firefox), but when I refreshed it went away...

OK, this is done. Just about every page should now be valid XHTML 1.0 Transitional. However, I have undoubtedly made a few mistakes along the way. For example, you might see the fancy less-than-or-equal sign &#8804; in code in analyses, even though it should be <=. My fault; if you do see such a thing, make sure you point it out. You may also see things like "cout ≤≤ x ≤≤ endl;"-- same idea here.
It would've been impossible for me to get everything done in such a short time if I had not enlisted the help of the UNIX utility sed for automatic editing. However, automatic search-and-replace often replaces items which actually shouldn't be replaced. For those UNIX-inclined, here is the script used:
[code]#!/bin/bash
sed -e 's|<br>|<br/>|g' -e 's/width=\([0-9][0-9]*\)/width="\1"/g' -e 's/height=\([0-9][0-9]*\)/height="\1"/g' \
-e 's/cellspacing=\([0-9][0-9]*\)/cellspacing="\1"/g' -e 's/border=\([0-9][0-9]*\)/border="\1"/g' \
-e 's/class=\([a-zA-Z0-9][a-zA-Z0-9]*\)/class="\1"/g' -e 's/align=\([a-zA-Z0-9][a-zA-Z0-9]*\)/align="\1"/g' \
-e 's|<PRE>|<pre>|g' -e 's|</PRE>|</pre>|g' -e 's|<P>|<p>|g' -e 's|</P>|</p>|g' \
-e 's|<SUB>|<sub>|g' -e 's|</SUB>|</sub>|g' -e 's|<I>|<i>|g' -e 's|</I>|</i>|g' \
-e 's|<UL>|<ul>|g' -e 's|</UL>|</ul>|g' -e 's|<LI>|<li>|g' -e 's|</LI>|</li>|g' \
-e 's|<H2>|<h2>|g' -e 's|</H2>|</h2>|g' -e 's|<H3>|<h3>|g' -e 's|</H3>|</h3>|g' \
-e 's|<CODE>|<code>|g' -e 's|</CODE>|</code>|g' -e 's|<BR>|<br/>|g' \
-e 's|<i>\([^<>]*\)<sub>\([^<>]*\)</sub></i>|<i>\1</i><sub><i>\2</i></sub>|g' \
-e 's|<i>\([^<>]*\)<sub>\([^<>]*\)</i>\([^<>]*\)</sub>|<i>\1</i><sub><i>\2</i>\3</sub>|g' \
-e 's|<B>|<b>|g' -e 's|</B>|</b>|g' -e 's|<=|\&le;|g' -e 's|<hr>|<hr/>|g' \
< $1 > /tmp/$1
cp /tmp/$1 $1[/code]
If you see weird stuff like "&le;<=", it's my fault for making a mistake with the script. Again, make sure you point it out.

#!/bin/bash 
sed -e 's|<br>|<br/>|g' -e 's/width=\([0-9][0-9]*\)/width="\1"/g' -e 's/height=\([0-9][0-9]*\)/height="\1"/g' \ 
-e 's/cellspacing=\([0-9][0-9]*\)/cellspacing="\1"/g' -e 's/border=\([0-9][0-9]*\)/border="\1"/g' \ 
-e 's/class=\([a-zA-Z0-9][a-zA-Z0-9]*\)/class="\1"/g' -e 's/align=\([a-zA-Z0-9][a-zA-Z0-9]*\)/align="\1"/g' \ 
-e 's|<PRE>|<pre>|g' -e 's|</PRE>|</pre>|g' -e 's|<P>|<p>|g' -e 's|</P>|</p>|g' \ 
-e 's|<SUB>|<sub>|g' -e 's|</SUB>|</sub>|g' -e 's|<I>|<i>|g' -e 's|</I>|</i>|g' \ 
-e 's|<UL>|<ul>|g' -e 's|</UL>|</ul>|g' -e 's|<LI>|<li>|g' -e 's|</LI>|</li>|g' \ 
-e 's|<H2>|<h2>|g' -e 's|</H2>|</h2>|g' -e 's|<H3>|<h3>|g' -e 's|</H3>|</h3>|g' \ 
-e 's|<CODE>|<code>|g' -e 's|</CODE>|</code>|g' -e 's|<BR>|<br/>|g' \ 
-e 's|<i>\([^<>]*\)<sub>\([^<>]*\)</sub></i>|<i>\1</i><sub><i>\2</i></sub>|g' \ 
-e 's|<i>\([^<>]*\)<sub>\([^<>]*\)</i>\([^<>]*\)</sub>|<i>\1</i><sub><i>\2</i>\3</sub>|g' \ 
-e 's|<B>|<b>|g' -e 's|</B>|</b>|g' -e 's|<=|\&le;|g' -e 's|<hr>|<hr/>|g' \ 
< $1 > /tmp/$1 
cp /tmp/$1 $1

Happened to me too. (Chrome) Refreshing solves the problem.
(Maybe coz both W3C and Chrome are open source ;D )

W3C isn't a program, it's an organization. Although the validator is open source.

Use this to see anything unusual on a particular page : [url=http://browsershots.org/]http://browsershots.org/[/url]

There are A LOT of browsers, you might want to check ONLY the most used ones because if you check them all, it can take very long time.