W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > January to March 1998

Re: WD-WAI-PAGEAUTH-0203 and the use of ALT text

From: Colin F Reynolds <colin@bespin.demon.co.uk>
Date: Mon, 9 Feb 1998 10:51:01 +0000
Message-ID: <QJd8uPAV+t30EwCo@the-net-effect.com>
To: w3c-wai-ig@w3.org
Cc: "Alan J. Flavell" <flavell@a5.ph.gla.ac.uk>
Since the way in which ALT content is defined, and therefore interpreted
by both browser developers and content providers is, IMO, crucial to the
effective rendering of information in a meaningful cross-platform

I have reservations as to whether the discussion of the use of ALT
content should be restricted to a forum which appears to be devoted to
the concerns of "Accessibility", which is bound to be seen as of
minority concern.

However, AFIACS this forum appears to be the only one addressing the
problem - or should this be discussed in w3c-wai-gl, -ui, or -au
instead? (In fact, how can one hope to come up with a concensus if the
three interested groups are discussing the issues independently of each

I am keen to have the usage of ALT properly defined in an unambiguous
way, at the earliest possible juncture, in order that browser developers
and content providers may both work towards a common goal. The present
situation is that content providers ("Web Authors") have to constantly
struggle to markup inline content such that it still makes sense when
"viewed" in a non-graphical environment.

Recently, this situation became *worse*, instead of better, due to the
implementation by some browsers of ALT text as popup tooltips in a
graphical environment. I would argue that this functionality is totally

Consider, for instance, the common usage of such constructs as:

  <IMG src="foo.gif" ALT=" ">

When rendered by a browser that displays ALT text as tooltips, the user
interface is spoilt.

In this example, it can be argued that it is a simple matter to alter
the markup such that it reads

  <IMG src="foo.gif" ALT="">

However, AIUI there are user agents which require the use of a space,
not a null string, in order to render sensible content.

I have written a page which deals with this matter in some more detail,
and gives examples to illustrate my concerns:



In article <Pine.OSF.3.96.980205192149.10146C-100000@a5.ph.gla.ac.uk>,
"Alan J. Flavell" <flavell@a5.ph.gla.ac.uk> writes

[much useful commentary snipped]

>I note the advice given at "Provide descriptive titles for all images used
>as links." and the example 
>        <A href="home.html" title="XYZ company home page">
>        <IMG src="logo.gif" alt="XYZ logo">
>        </A>

I am in total agreement with Alan here. Where the IMG is unavailable to
the user agent (for whatever reason), rendering "XYZ logo" in its place
as recommended in this example is IMO worse than useless. Of what use is
this information to a blind user? A sighted user would be better served
by the presentation of simply "XYZ".

>I'm thinking this one over.  I had always been accustomed to marking
>this up something like
>        <A href="home.html">
>        <IMG src="logo.gif" alt="[XYZ company home page]">
>        </A>
>but I can see the point - though again I have the same reservations about
>using ALT to _describe_ the information, rather than to substitute for it.

The example that Alan cites here hides another can of worms in the
context of tooltips. Consider the following, where the markup has been
employed in the manner which I feel is appropriate (ALT as a
_replacement_ for the missing image, TITLE as _supplementary
information_ for the element):

       <A href="home.html" title="XYZ company home page">
       <IMG src="logo.gif" alt=" XYZ " title=" XYZ logo ">

Unless there are clear guidelines as to how tooltips are to interpret
such markup, the possibilities for future confusion and poor backward
compatibility are endless :(


A short while ago, on www-html:

In article <01ISU2RGUNG200877V@SCI.WFBR.EDU>, Foteos Macrides 

[discussion snippage]

>       So what you want is:
>       <IMG SRC="foo.blah" ALT="alternative_string" TITLE="tooltip_string">
>and broader implementation and understanding of this provision in HTML 4.0.
>                               Fote
> Foteos Macrides            Worcester Foundation for Biomedical Research
> MACRIDES@SCI.WFBR.EDU         222 Maple Avenue, Shrewsbury, MA 01545


Colin Reynolds
"I know you believe you understand what you thought I said, but
I'm not sure you realize that what you heard was not what I meant!"
Received on Monday, 9 February 1998 07:17:58 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 13 October 2015 16:21:00 UTC