W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > January to March 2002

RE: Pixels vs Points as a unit for sizes

From: Charles McCathieNevile <charles@w3.org>
Date: Tue, 8 Jan 2002 18:29:46 -0500 (EST)
To: "DERRICK, Monique" <Monique.Derrick@qed.qld.gov.au>
cc: <w3c-wai-ig@w3.org>
Message-ID: <Pine.LNX.4.30.0201081814440.16514-100000@tux.w3.org>
Hi Monique,

yes, using ems does open the way for the sizing to change. This is an
important feature of the Web as a medium, and a reason why people use the Web
when they have difficulty using things like printed material.

Essentially in designing for the Web it is important to take this possiblity
into account.

An example like http://www.w3.org/Jigsaw demonstrates a problem that can
arise, where the background is required to ensure sufficient contrast, and
doesn't resize according to the text in the sidebar.

(this could be fixed by applying a background colour to the div used for the
sidebar. The width of that div could be expressed in ems - it is known how
many letters the longest line is...)

By contrast http://www.tropicasa.com avoids this with its design, and
changing the font to 3 or 4 times normal size doesn't present much problem at
all.

cheers

Charles McCN

On Wed, 9 Jan 2002, DERRICK, Monique wrote:

  What I want to know about this is if we use em's then that completely
  changes the design if the viewer uses larger sizing, how do you take this
  into account when designing a site.

  Thanks

  Monique Derrick

  -----Original Message-----
  From: Charles McCathieNevile [mailto:charles@w3.org]
  Sent: Tuesday, January 08, 2002 4:29 PM
  To: WAI IG
  Subject: Pixels vs Points as a unit for sizes


  I think this is part of a Frequently asked Question...

  Using px (pixels) as a unit of measure for font sizes isn't a great idea. It
  takes no account of the size the user prefers to have their text so it can
  be
  easily read. It also means your text is likely to bo very different sizes on
  different kinds of system, so it doesn't provide control anyway.

  Using pt (points, or 72nds of an inch) isn't too good either. Not only does
  it ignore the size that the user would like to work with, but because of
  strange bugs in some systems it isn't actually reliable in all browsers, and
  you text may come out a lot smaller than expected.

  There are a couple of cases where it seems reasonable to use these measures:

  For images that do not scale well (e.g. gif, png or jpg formats) it is not a
  bad idea to set sizes in terms of pixels.

  For a narrow decorative border or horizontal line, it doesn't make that much
  difference what the units are.

  just a thought - my personal opinion.

  cheers

  chaals

  Vadim Plessky recently wrote:

  Now I realize that during 1.5 months that I am subscribed to this list, none
  was asking "how using pixels instead of points affects site accessibility"
  I am not visually impared, but have some preblems with accessing
  microsoft.com site (and several other sites produced with genius "Microsoft
  technology")
  Can someone take care about me, please? :-)

  What I found is that even if you use 'pt' font sizes - there is no warranty
  that your system (Windows / PC or Macintosh) has correct dpi (dots per inch)
  settings. And without correct dpi settings, there is no way how browser can
  recalculate 'pt' (or 'cm', 'mm') to number of pixels...
  I have read in some book /article that both Windows/PC and Macintosh are
  broken in terms of reporting correct dpi settings. Has someone opposite
  info?

  [ good news: recent versions of XFree86, used on Linux, can auto-detect dpi
  on many video adpaters/display combination, so we finally have at least one
  platform working ok]





-- 
Charles McCathieNevile    http://www.w3.org/People/Charles  phone: +61 409 134 136
W3C Web Accessibility Initiative     http://www.w3.org/WAI    fax: +1 617 258 5999
Location: 21 Mitchell street FOOTSCRAY Vic 3011, Australia
(or W3C INRIA, Route des Lucioles, BP 93, 06902 Sophia Antipolis Cedex, France)
Received on Tuesday, 8 January 2002 18:29:50 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 19 July 2011 18:14:00 GMT