There are actually two "relative" issues being dealt with here,
which may be part of the confusion.  When a page is laid out using
relative proportions, the size of table cells or graphical elements will
automatically change with the size or resolution of the display screen.
If one viewer looks at the page with a 17 inch monitor set to 1024x768
resolution, and another views it with a 13 inch monitor and 640x480
resolution, you can assure that the text occupies the same proportion of
the visible screen.  This is considered good practice, because your
table won't hang off the side of the small screen, requiring frequent
scrolling.

       The em and ex elements don't scale with the screen size, but with
the font size.  If I need to have a larger font on the screen because of
low vision, the relative size of margins and some in-line graphics need
to change to accommodate that change.  For example, I use a CSS
"drop-cap" on some of my pages, and that should change in proportion to
the text around it so that the overall layout remains the same.

       So, tables and graphics have a relative measure that is based on
the size of the screen, and in-line elements have a relative measure
that is a function of the font size that is around them, and to some
extent, to the font, since the proportion of ex height to em height
differs between fonts.

       I trust you were being facetious about not caring about border
widths, because you were blind.  That is the mirror of the attitude that
has led to so many inaccessible pages.  "I can see, so I don't care
about blind access."  When we design pages, they aren't for *us* to look
at, but for our audience.

       Denis Anson, MS, OTR
       Computer Access Specialist
       College Misericordia
       301 Lake St.
       Dallas, PA 18612
       email: [log in to unmask]
       Phone: 570-674-6413


       > -----Original Message-----
       > From: * EASI: Equal Access to Software & Information
       > [mailto:[log in to unmask]] On Behalf Of Ross Eadie
       > Sent: Monday, October 28, 2002 12:37 PM
       > To: [log in to unmask]
       > Subject: Re: Question about absolute vs. relative sizes
       >
       > I am by no means an expert, but It would seem to me that
elements
       > related
       > to the spacial layout of the screen should be created by using
       > percentages.
       >  I think you might be getting confused with the dynamic
relationship
       > of
       > relative percentages of the screen verses the affect a screen
size
       > can have
       > on an element using pixcel size.  On the other hand, someone
could
       > make the
       > pixcel size relative to the screen pixcel size variable if one
can
       > create a
       > script or program to extract that information from every user
agent
       > hitting
       > the page.  I am not overly familiar with CSS 2.? or 1.0 for
that
       > matter.
       > Does it provide a utility to extract the screen size variable
from a
       > user
       > agent's computer?  It is just so much easier to use
percentages.
       > As for using pixcel size for borders and similar elements, I
would
       > suggest
       > the element use dynamic relative measurements as well to avoid
       > visual
       > distortions.  I am totally blind, and I honestly don't care
about
       > the size
       > of borders anyway <grin>.
       >
       > At 09:18 AM 10/28/02 , you wrote:
       > >Hello EASI Colleagues,
       > >
       > >I would appreciate hearing comments from web accessibility
experts
       > on this
       > question I received from a web developer who is retrofitting a
site.
       > >
       > >Alan
       > >
       > >The question:
       > >
       > >Which length measurements are relative?
       > >
       > >The WAI examples list percentages (%) and em (base font
height) as
       > examples of relative length measurements (slide 32). BUT, the
CSS1
       > and CSS2
       > >specifications also list ex (the x-height) and px (pixels)
       > >as relative measurements. "px" is relative because
       > >
       > ><blockquote>
       > >Pixel units are relative to the resolution of the viewing
device,
       > i.e.,
       > most often a computer display. If the pixel density of the
output
       > device is
       > very different from that of a typical computer display, the
user
       > agent
       > should rescale pixel values. It is recommended that the
reference
       > pixel be
       > the visual angle of one pixel on a device with a pixel density
of
       > 90dpi and
       > a distance from the reader of an arm's length. For a nominal
arm's
       > length
       > of 28 inches, the visual angle is therefore about 0.0227
degrees.
       > (CSS2 spec)
       > >
       > ></blockquote>
       > >
       > >Since the pixel is defined in terms of a visual angle and a
       > distance,
       > isn't it really an absolute value just like in (inches) or cm
       > (centimetres)? Also, it's not resizeable by the user like font-
       > relative
       > measurements are, is it?
       > >
       > >On the other hand, it's very useful for defining things like
       > border-width
       > or image sizes. Is it OK (i.e. accessible) to use px for
certain
       > properties
       > and not for others?
       > >
       > >
       > >
       > >Alan Cantor
       > >Project Manager
       > >Strategic e-Government Implementation
       > >e-Government, OCCS
       > >416-212-1152
       > >[log in to unmask]
       > >
       > ---
       > Ross Eadie
       > Voice:  (204) 339-5287