W3C home > Mailing lists > Public > www-style@w3.org > February 2004

Re: [css3-page] LCWD issue 22 -- [22] Section 3.3.2 <length>

From: Boris Zbarsky <bzbarsky@MIT.EDU>
Date: Tue, 17 Feb 2004 15:54:18 -0500
Message-Id: <200402172054.i1HKsIUO018021@no-knife.mit.edu>
To: "BIGELOW,JIM (HP-Boise,ex1)" <jim.bigelow@hp.com>
Cc: ernestcline@mindspring.com, W3C CSS List <www-style@w3.org>

> Points and picas are units of absolute length rather than relative units.
> The other relative unit, px (pixel) is device dependent.  Images printed in
> a  300 dot per inch device may not be the same size as one rendered on a 600
> dpi device. The same pitfall of device dependence holds for specifying the
> page size in pixels.

Actually, if I understand the definition of "pixel" in CSS correctly, that is
precisely what should NOT happen if pixels are implemented correctly.  In CSS
"pixel" does not refer to a dot but rather to a solid angle in the field of
view.  Therefore a "300px" image should have the same physical dimensions (as
measured with a ruler) no matter what the medium, as long as the viewing
distance is held constant (modulo the possily low resolution of the imaging
device, actually).

This is not a problem with desktops yet, since they almost all have basically
the same DPI.  But with the appearance of 200+ dpi LCD displays on the market
will expose the fact many implementations do _not_ in fact implement CSS pixel units
correctly and should hopefully lead to said implementations being corrected
(for the simple reason that on such a display a CSS implementation which
directly maps pixels to screen dots will show all images 1/2 to 1/3 of
the size they really should be, and the discrepancy will be glaringly obvious).

Boris
-- 
A computer lets you make more mistakes faster than any invention in 
human history, with the possible exceptions of handguns and tequila.
Received on Tuesday, 17 February 2004 15:54:26 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:26 GMT