| |
 he first accessibility principle is
presenting your information in alternative modalities---that
is, using as many different media and different input/output
techniques as possible. For example, the audio channel may
work for the blind, while the visual channel would duplicate this
information for the deaf. This principle is especially
important for the "bottlenecks," that is, the elements that are
critical for delivering your message, for understanding of what
follows, or for navigating to other parts of the site.
In practice, the alternative modalities principle most often
amounts to the imperative of providing textual parallels for all
non-textual information. This is because plain text is the
most natural informational medium for the computer; it is easy to
display with any sort of computer interface, easy to index and
process automatically, and relatively easy to convert to
speech. By presenting information as text, we take advantage
of the "inherent accessibility" of this format and may not be
concerned any more about alternative modalities.
However, a page doesn't need to be text-only to be
accessible. The principle of alternative modalities and the
provisions of HTML allow you to produce pages that are both
presentationally rich and accessible at the same time. If you
need an audio clip, accompany it with a text transcript; if you want
an image (as you most probably do), give it an alt
text conveying as much of the image's information as possible (for
more on this, see "Graphics" later in the
chapter); and if you feel an urge to employ a Java applet or
Shockwave object, make sure it is not critically necessary
for navigating your pages or reading the text (and if that's not the
case, provide an alternative access mechanism for the same
information).
In other words, an accessible web page may be multi-modal
(i.e. may use various input/output techniques) provided that all its
critical parts are sufficiently alt-modal (i.e. no important
part of its content is locked within one modality). The art of
creating alt-modal web pages while keeping them efficient, rich, and
elegant enough to be competitive in today's Web might seem tricky, but
actually it's only a matter of obeying some simple and obvious
guidelines (most of which are summarized in this chapter).
The second fundamental principle of accessibility is using
logical, not presentational, markup for your content.
(You may be familiar with these concepts from Chapter 4, "The HTML
Document Type Definition" that deals with the SGML roots of HTML.)
Logical markup means that for every element of your page, you should use HTML
tags to describe what it is rather than how it's supposed to
look (or sound, or print, or whatever).
The importance of logical markup reaches far beyond the problem of
accessibility for the disabled; in fact, it is the best method to
ensure the general accessibility, processibility, and, in the
bottom line, longevity of your information. When the computer (not
only your human reader) is aware of the logical structure of your data, the
chances of its being able to successfully convert it into another
format or media are very much improved. And accessibility implies, for the
most part, exactly this: the ease of transforming the document into
another media or presentation mode, in order to be able to communicate
it to people with disabilities.
One example is the controversy of the I tag versus the
EM tag. The former one, being a tool of
presentational markup, directs the browser to switch to an italic
font and is thereby limited to visual presentation mode only.
A speech browser has no "italic font" and is therefore likely to
ignore this piece of markup altogether. Conversely,
EM is a logical tag meaning emphasis; while a visual
browser may interpret this element by switching to an italic font, a
speech browser may read it aloud with a different tone of voice
("acoustic emphasis"). An additional advantage is that
EM allows distinguishing italics for emphasis from other
uses of italics, such as citations (to be marked up by the
CITE logical tag).
Naturally, web authors often need tools to adjust presentational
aspects of their documents as well---for instance, to control font
face, or size, or voice pitch. HTML's answer to this demand is
style sheets, an external mechanism capable of specifying a
multitude of presentational parameters in a flexible manner (see
http://www.w3.org/TR/REC-CSS1).
The current specification of HTML style sheets supports only visual
presentation parameters, but there exists a proposed extension for
handling aural (speech) parameters as well (http://www.w3.org/TR/WD-acss).
| |
