Using the conditional comment method, to provide older browsers with
the correct styles, makes maintainance, and cascading styles more
difficult to manage.

w3c blog has this to say about their oocss site [http://www.w3.org/
Help/about-redesign#limitations]:
"Because of known interoperability issues, we have accepted to use CSS
that does not validate with the CSS validator. Over time we hope to
evolve towards valid CSS."

http://www.webstandards.govt.nz/technical/, only has this to say about
stylesheets:
"Stylesheets. Sites must work properly (be navigable, present
information intelligibly) with stylesheets disabled."

It doesn't say that the css must validate, it just says that the site
must work correctly without css.

I can't find any mention of css validation in the WCAG2.0 (AA)
Guidelines either.

What is our stance on css validation?

Hi Murray,

I've written up my reasons for choosing the _ and * hacks over other  
options. You can check that out here.

http://wiki.github.com/stubbornella/oocss/faq#hacks

Cheers,
Nicole

On Nov 8, 2009, at 11:36 PM, Murray wrote:

Object Oriented CSS  http://wiki.github.com/stubbornella/oocss
Smush it  http://smush.it
Website  http://stubbornella.org
Tweet!  http://twitter.com/stubbornella
Book  Even Faster Websites with O'Reilly

Yeah, I'm thinking this might be a good way to put it.

The trouble is that OOCSS uses (very minimal) hacks, to deal with  slight
display mismatches between older and newer browsers, rather than conditional
comments. A couple of reasons are explained here,
http://wiki.github.com/stubbornella/oocss/faq#hacks, but personally I like
that:

·         It makes the code easier to maintain for other developers where
the slight variation of values sit side by side rather than in another file.

·         Fewer http calls makes it more efficient.

*The WCAG2.0(AA) Guidelines doesn’t say anything about css validation.*

OOCSS also uses bits of CSS3 which also don’t validate, but so do many
“lightbox” solutions. The reason CSS3 may be present is that, in an ideal
world, older techniques, such as background images for rounded corners,
would be served using a javascript which could be removed when CSS3 is
accepted in all browsers (when IE6,7,and 8, are dead).

So far I have spoken to one of the guys at webstandards.govt.nz who agrees
that *there is currently nothing in the NZ standards which states that CSS
must validate*. He said that the Web Standard Working Group no longer felt
that it was necessary.

However I have written him a long email, about my approach, and some
questions, for him to discuss with his colleague in the next day or so. I am
interested to hear what they think but I’ll be sure to summarise it for you
when I know more.

In the mean time the W3C (http://www.w3.org) are using OOCSS for their site.
This is their stance on the subject:

Is the CSS invalid? The CSS does not validate with the W3C CSS validator. We
mentioned this as one
limitation<http://www.w3.org/Help/about-redesign#limitations>of the
site back in March. As we wrote then, “Because of known
interoperability issues, we have accepted to use CSS that does not validate
with the CSS validator. Over time we hope to evolve towards valid CSS.”

They are *The* Web Standards Authority, and I took a sigh of relief when
they released the new OOCSS site because it means we’re all heading in a
good direction.

m.

On Mon, Nov 9, 2009 at 3:15 PM, Nicole Sullivan <nic...@stubbornella.org>wrote: