WebAIM - Web Accessibility In Mind

E-mail List Archives

RE: Relative vs. absolute units

for

From: John Foliot - bytown internet
Date: Mar 4, 2003 2:39PM


Julian,

sorry for not responding earlier... this one slipped through the cracks and
was only discovered during regular maintenance...

*Under normal circumstances* an em is 16px. I say normal, as this is based
on the fact that the end user generally does not change their default font
size. In IE, the only way to do that is to change the default font size
through the OS properties dialogue (Desktop>> right click>>> properties).
In Netscape this measurement can be changed via the Edit>> Preferences
dialogue within the browser. If you have either version 6 or 7 installed
and check, you will see that the default font is actually set to a pixel
size, and not the older point size (which was what NN 4 used). Note, this
doesn't work in NN4, as that browser doesn't know what an "em" is. However,
like all unknowns, if the browser does not know what "it" is it ignores
"it"; technically, those images in NN4 would not have declared Height and
Width attributes, or at least none that had any real meaning to the browser.

Now should a user go in and change their default font size to something
larger, then the image will scale larger - proportionate to the font size
still, but again, you will experience a slight degradation of image quality.
For this reason I tend to restrict it to "mission critical" imagery - icons,
text buttons, etc., whereas photos, "design" imagery, etc. I still size in
pixels. It's not a perfect solution, but a stab in the (I hope) right
direction.

cheers!

JF

> -----Original Message-----
> From: <EMAIL REMOVED>
> [mailto: <EMAIL REMOVED> ]
> Sent: Tuesday, January 07, 2003 11:18 AM
> To: <EMAIL REMOVED>
> Subject: RE: Relative vs. absolute units
>
>
> Despite the fact that the bytowninternet.com example you gave is quite an
> eye-opener, it raises the question, how does one measure in ems or convert
> from Photoshop pixels to HTML ems?
>
> Jules
>
> > -----Original Message-----
> > From: John Foliot - bytown internet [mailto: <EMAIL REMOVED> ]
> > Sent: Friday, January 03, 2003 5:21 PM
> > To: <EMAIL REMOVED>
> > Cc: <EMAIL REMOVED>
> > Subject: RE: Relative vs. absolute units
> >
> >
> > Jules,
> >
> > Your reading is both correct and reasonable.
> >
> > However, remember that with the latest versions of popular
> > browsers, they
> > also support relative sizing for images. See:
> > http://www.bytowninternet.com/examples/relative.html for
> > examples of what I
> > mean.
> >
> > While most images, when subjected to this treatment, may lose
> > a fair bit of
> > quality, it is an important consideration when developing
> > graphics such as
> > buttons or "clickable" icons. In these instances, making the
> > image scale
> > with the text may aid in accessibility, despite the
> > degradation of image
> > quality.
> >
> > Food for thought.
> >
> > JF
> >
> >
> >
> >
> > > -----Original Message-----
> > > From: <EMAIL REMOVED>
> > > [mailto: <EMAIL REMOVED> ]
> > > Sent: Friday, January 03, 2003 3:22 PM
> > > To: <EMAIL REMOVED>
> > > Subject: Relative vs. absolute units
> > >
> > >
> > > Hi:
> > >
> > > WCAG 1.0: Checkpoint 3.4
> > >
> > > "Use relative rather than absolute units in markup language
> > > attribute values
> > > and style sheet property values."
> > >
> > > I understand the purpose of using relative font sizes but WCAG 1.0
> > > Checkpoint 3.4 does not make the distinction that it is
> > referring to only
> > > font sizes or if it is refering to any time a measurement is to
> > > be used. For
> > > example, for usability issues, graphics should be sized
> > with height and
> > > width but graphics have a "physical" dimension and should be
> > > measured using
> > > pixels, an absolute unit of measurement. In an instance where the
> > > left cell
> > > of a layout table in which the left cell is the navbar for the
> > > page (and the
> > > site), if the links are graphical links, then the graphics
> > will have to be
> > > measured using pixels and it makes sense to make that
> > cell's width also
> > > measured in pixels. However, I do agree that the table as a whole
> > > should be
> > > sized using percentage (for example 100%) so that the right
> > cell with the
> > > content may expand or contract as the widht of the browser window
> > > dictates.
> > >
> > > Is this a reasonable interpretation of checkpoint 3.4?
> > >
> > > Jules
> > >
> > > ---------------------------------------------------------
> > > Julian Rickards
> > > Digital Publications Distribution Coordinator
> > > Publications Services Section
> > > Ontario Ministry of Northern Development and Mines
> > > Phone: (705) 670-5608
> > > Fax: (705) 670-5690
> > >
> > >
> > > ----
> > > To subscribe, unsubscribe, or view list archives,
> > > visit http://www.webaim.org/discussion/
> > >
> > >
> >
> >
> >
> > ----
> > To subscribe, unsubscribe, or view list archives,
> > visit http://www.webaim.org/discussion/
> >
>
>
> ----
> To subscribe, unsubscribe, or view list archives,
> visit http://www.webaim.org/discussion/
>
>



----
To subscribe, unsubscribe, or view list archives,
visit http://www.webaim.org/discussion/