Re: Dave Raggett: Re: IMG ALT attribute in HTML 4.0 (fwd)

Upon further consideration, I am genuinely not sure whether LONGDESC would
be a good transitional measure, and I am willing to be convinced by both
sides.

LONGDESC Advantages: it can readily be added to existing pages; it is
visually invisible, hence non-invasive; it should be relatively easy to
implement.

Disadvantages: it is not supported by existing user agents; the
description must reside either in a separate document, or in another part
of the same document if an URL fragment is used; the list of links
pertaining to an image map can not be integrated into the description as
with OBJECT; promotion of LONGDESC may entrench an inadequate and obsolete
solution into HTML authoring practices.

D-link advantages: it is supported by all existing user agents; it can
easily be recognised by a person who is using a screen reader, whether in
a text-only or graphical environment.

Disadvantage: it detracts from the visual appearance of a page and is
therefore unappealing to authors. Although d-links can be suppressed by
styles, the necessary feature of CSS is not available in current
implementations (is this still true)?

Another approach is to redesign the page to incorporate a visually
unintrusive link to a long description.

Advantages: this approach is compatible with existing implementations; it
is not detrimental to the visual presentation of the page.

Disadvantages: the presence of a link to a long description may not be
obvious to screen reader users unless appropriate guidelines are developed
and followed; in the absence of appropriate conventions (such as a
reserved value of CLASS), the link referring to a long description could
not be selected by CSS for the application of audio/braille styles; as
with d-link, the description is separated, in the case of an image map,
from the list of relevant links; we do not know how feasible this solution
would be in practice, and how easily it can be put into effect in a manner
that is satisfactory to both HTML authors and people with disabilities.

I have not attempted to assign comparative weightings to the
aforementioned advantages and disadvantages, although clearly some are
outwayed by others, and it is necessary to balance relevant considerations
in arriving at a decision.

Received on Friday, 12 September 1997 22:13:11 UTC