alt131 - 8:55 am on Jun 2, 2011 (gmt 0)
Lucy that would be a long sleep Laura's last post was 47.5 hours ago!
For anyone following, this isn't a secret club, I've stickied a link to code to those interested.
Yup - recalling the reason for looking at the code is try to find the source of the error and this is very good compared to what we frequently see here. There is an unclosed <p> and <a>, which the browsers are self-closing. The remainder are missing "alts", and deprecated properties/values that won't be having an effect. That leaves the entity references and given one of my tests was to remove all child elements (which also means all invalid code), they aren't having any effect either. So while validation is really important, in this case invalidities do not seem to be the problem.
but have you taken a closer look at the html?
However here's an interesting side-bar: The opening of the document is as Laura posted above. Yet the first (and sometimes second) validation using jigsaw at default settings validates as HTML 4.01 - not XHTMl. Hitting "revalidate" brings up validation as xhtml. Using Unicorn this validates as xhtml first off. Has anyone experienced jigsaw doing this before?
Clever thinking about add-ons. To keep narrowing this down - are you seeing Laura's reported layout variation between ff and others - or like me, not seeing that at all?