Re: Targeting CSS3 only (evil?), either with pseudoclass or an extra syntax for properties.

Barry,

Since you advocate conditional comments so avidly. What is the
advantage for you (solely in the conext of CSS, not talking HTML) for
having a "conditional" syntax based on user agent name/version (like
you describe), over having a syntax that enables conditional styles
based entirely on the user agent's support for a certain, specific
subset of styles which you're using in a use case?

Ben

On Apr 4, 2005 11:23 PM, Barry <wassercrats@hotmail.com> wrote:
> 
> > The only way for web designers to stay sane, is to stick to standards.
> 
> I stick to valid HTML and CSS, but I still like conditional comments.
> 
> 
> > Every layout engine will have bugs. If you're looking for
> > pixel-perfection, HTML is the wrong medium. A strong core of standards
> > gives wiley web designers the tools they need.
> 
> I don't know if you'd call my needs pixel-perfection, but I could only get
> what I want with conditional comments or scripting. If conditional comments
> create pixel-perfection, all the better.
> 
> 
> > And if you want to see a good example of innovative web browser design
> > that doesn't flaunt web standards, check out Omniweb[1]. It's features are
> > impressive and original.
> >
> > [1] http://www.omnigroup.com/applications/omniweb/
> 
> That webpage has invalid HTML. If all that was "wrong" with it were
> conditional comments, it would validate. It makes me wonder how much the
> authors of Omniweb really care about standards.
> 
> Omniweb might be innovative, but it would be better if it supported
> conditional comments.
> 
> If everyone had my standards for creating webpages (proper use of tables,
> browser compatibility, and valid HTML and CSS), there would be more demand
> for conditional comments. I don't need them to take the place of the
> proposed CSS property sniffers. I just need them for what I'm using them for
> now, for example to fix the blink in IE before the background position
> changes, which isn't a bug and wouldn't be caught by a property sniffer.
> Others have brought up problems with the proposed property sniffers, and
> conditional comments could help there too.
> 
> 
> David Woolley wrote:
> > I think the particular point here was that the HTML should reflect the
> > meaning of the document, not its appearence, so should not change.
> 
> I prefer what ever works best, and conditional comments work best, at least
> for me. I wouldn't mind having conditional comments for CSS only. CSS is the
> only thing I'd need them for if either Firefox or Opera had them.
> 
> 


-- 
http://www.ben-ward.co.uk

Received on Monday, 4 April 2005 23:13:54 UTC