[thelist] testing older verions of IE for css (was css menu problem)

[Barney Carroll]

No way. IE6 is still has a regrettably significant market share and requires
testing for responsible universal web design.  IE5.5 can go (and for the
most part has gone) to hell.

I work with firebug and like to keep all the CSS for any given element in
one place and one rule where possible, so I use hacked CSS per-property, eg:

selector {
property: regular value;
_property: ie6 value;
 *property: ie7+ value;
}

I personally find this is much better for my workflow. Whenever a problem
arises in any browser I know exactly where to look. Likewise for the 10+
other people who may need to edit my code down the line will see the code
where it works. Commenting each line, eg _property: value; /* ie6-specific
value */ may also help for master files depending on how educated the rest
of your team is.

I have a peeve with conditional comments in that you need to edit code in
what are technically comments in your actual markup to fix any given
problem, which in my mind is not the way to resolve a styling bug. This way
I end up with invalid CSS, but I have one file, one http request for each
user, and everything in one place.


Regards,
Barney Carroll

barney.carroll at gmail.com
07594 506 381


2009/10/30 Birdie <bird at koolfish.com>

>
> "The reason IE screws up is because it interprets elements differently.
> Read this article:
>
>
> http://www.456bereastreet.com/archive/200612/internet_explorer_and_the_css_b
> ox_model/
>
> I use suggestion 3, conditional comments for all websites that I create
> now.
> You can tweak the CSS for each version of IE, since they all suck in
> different ways. "
>
> Do the majority of you do this?
> I mean do you still test on 5.5 and lower for IE for websites that do not
> target a specific audience and could be viewed by anyone?
>
> Lisa
>
>
>
> --
>
> * * Please support the community that supports you.  * *
> http://evolt.org/help_support_evolt/
>
> For unsubscribe and other options, including the Tip Harvester
> and archives of thelist go to: http://lists.evolt.org
> Workers of the Web, evolt !
>