W3C home > Mailing lists > Public > www-style@w3.org > September 2003

Re: Scrollbar Code

From: Ernest Cline <ernestcline@mindspring.com>
Date: Fri, 12 Sep 2003 18:25:44 -0400
Message-ID: <410-220039512222544796@mindspring.com>
To: www-style@w3.org

> From: Charles Kendrick <charles@isomorphic.com>
> 
> That's a good point - but that's what "!important" rules in user
stylesheets are for.
>
> It is already possible for a designer to create problems for people with
motor difficulties, poor 
> visual resolution, partial or total colorblindness, etc.  I think the way
to avoid this to educate 
> designers about accessibility problems.  Arbitrary limits on what can and
cannot be styled won't 
> really help given that accessibility problems can be created anywhere by
a designer who isn't aware 
> of accessibility concerns.
>
> Also, if I can't style the scrollbars, I can't offer novice users a high
contrast / poor motor
> skills skin for my site or web application.  I'm not saying that just to
say it, by the way.  Our
> product does offer skinnability, and we have had requests for a
high-contrast skin.  At the moment
> we can accomodate that via our DHTML scrollbars, but not via plain CSS.

I'd have to agree with Charles here. User agents that don't allow one to
have user stylesheets
or to turn off the default stylesheet, or use alternative stylesheets cause
problems beyond this
one area.  With them, allowing CSS to suggest a presentation for scrollbars
is no more subject
to potential abuse than other style capabilities such as the following CSS
which reverses a
common convention for visited and unvisited links .

:link {color:purple}
:visited {color:blue}

There are plenty of non-web programs (mostly games I admit) that use 
non-standard, fancified
UI widgets. If a user can't stand them, then they won't use the program. 
The same goes for
webpages.  Since CSS doesn't control anything outside the viewport, I see
no new problem
in having CSS being able to suggest the style of scrollbars.
Received on Friday, 12 September 2003 18:25:37 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:23 GMT