W3C home > Mailing lists > Public > www-style@w3.org > February 1999

CSS3 suggestion: make physical measurements invalid for screen media!

From: Braden N. McDaniel <braden@shadow.net>
Date: Mon, 8 Feb 1999 17:43:16 -0500
To: "W3C Style Sheets Mailing List" <www-style@w3.org>
Message-ID: <003501be53b4$6b04d4d0$01000080@bonezero>
I think this has come up before, but I don't remember seeing it on the list
of suggestions for CSS3, so I thought I'd bring it up again.

Physical measurements--in, cm, pc, pt--don't make sense in the context of
screen media, since the dimensions of the output device are never known (or
at least, CSS cannot be made aware of them). Thus I suggest that these
measurements be syntactically invalid inside @media rules that specify
screen media, and stylesheets imported with @import rules that specify
screen media. Rules within these contexts that included properties with
physical measurements would be seen as invalid, and ignored by a conforming
CSS3 parser per the rules in 4.2 of CSS2.

Why do this? As stated, these physical measurements don't make sense in the
context of screen media. But this doesn't keep naive authors from using
them; and worse, it doesn't keep naive software from generating HTML/CSS
with physical measurements which naive users will duly post to the Web.

Pixel-density can vary quite a bit, and while a pixel is a pixel, an inch is
also an inch. And if your monitor has a lower pixel density than mine, then
fonts I specify in points will be rendered with fewer pixels on your monitor
than they are on mine--meaning, usually, that they'll be harder (even
potentially impossible) to read.

Pixel measurements obviously don't have this problem. If I specify 20px,
you'll see 20 pixels and so will I, even if yours are bigger than mine. Ems,
ex, and percentages don't really have the problem either, since they will
all be functions of the readability of the base font size (which, if
expressed in pixels, should render consistent results for screen media).

Braden
Received on Monday, 8 February 1999 17:43:04 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:53:58 GMT