W3C home > Mailing lists > Public > www-style@w3.org > November 1996

Re: Relative weight - reader vs. author styles

From: David Perrell <davidp@earthlink.net>
Date: Thu, 15 Nov 2096 11:42:53 -0800
Message-Id: <199611151946.LAA14638@spain.it.earthlink.net>
To: "Sherm Pendley" <sherm@infoboard.com>, <www-style@w3.org>
Sherm Pendley wrote:
> To allow for this, I propose an addition to the "important"
declaration:  A
> "required" declaration, which would be guaranteed to override all
other
> settings.

It would be simpler but effectively the same to say that 'important'
has more weight on the reader side, such that any author rule will
override a reader rule with normal weight but an important reader rule
will override all author rules. But how does a reader know what classes
an author has defined? And what about style overrides within a
document? You could wind up with _less_ legibility.

It would be more useful if a user style sheet could be easily toggled
on or off with a button bar control in the UA, completely overriding
all other style. That way a visually impaired user need only apply
their stylesheet if a page is illegible or offensive. I don't consider
my sight to be unusually impaired, but I've seen many sites with dark
type on a busy dark background that I can hardly read. I'd love to have
an instant override.

This could be implemented regardless of the CSS1 cascade spec.

There are other ways to handle visual impairments on the UA-side. Soon
all GUIs will have color management systems, and color mapping can
accommodate the color-blind globally. Authors who care about the
legibility of their pages will spec type in points, not pixels, so that
type size on the client side will depend on the pixels/inch of the
display. Some video drivers in Win95 give control over display
pixels/inch, allowing visually impaired people to set default type size
to up to 1000% of normal (for TT and T1 fonts). Hopefully this will
become the norm, and fixed size bitmap 'system' fonts will die off
completely.

David Perrell
Received on Friday, 15 November 1996 14:49:32 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:53:46 GMT