Absolute and relative font sizes; was RE: Response to comments

[Thomas Dowling]

> At 6:16 PM 1998/05/18, Robert Rasmussen wrote:
> >What is NOT possible in HTML, up until CSS, is PRECISE control
> of font size -
> >making certain text display at 12 point (just like TV
> programmers can't make
> >your TV produce exactly 65 decibels).
>
> I hope that is never the case. As a reader _I_ want to control the font
> size and face that I read not what some designer thinks looks "good" to
> them. As my eyesight declines this will become of increasing
> importance to
> me.

Perhaps we should rephrase the claim.  CSS gives a precise language for
*suggesting* font size, by point size, percentage, etc.  Like any powerful
tool, it can be misused to detrimental effect.

>
> I understand due to some representations made from the Sunrise Labs in
> Australia that the standard will be to allow for the reader to override
> whatever style that is specified in the document. This is to be
> applauded.
>

Definitely.  The "C" in CSS does stand for "Cascading" and determines who
wins when style rules conflict.  The CSS spec assumes the user has access
to a personal style sheet; the order of precendence from lowest to highest
priority is:

  Normal user settings
  Normal author settings
  Author settings marked "!important"
  User settings marked "!important"

The last two rules are reversed from CSS1, which is a Good Thing.  Users
always have the trump card.

Several other rules determine precendence within each of these categories:
CSS2 section 6.4.1 spells them out,
<URL:http://www.w3.org/TR/REC-CSS2/cascade.html#cascading-order>.


Thomas Dowling
OhioLINK - Ohio Library and Information Network
tdowling at ohiolink.edu