Validation of Equivalents

A - Introduction

Charles McCathieNevile wrote:

"The notion of equivalence which we use allows and requires us to 
understand that some equivalents are better replacements than others. This 
applies to both textual and non-text equivalents. Perhaps we need to make 
the point more clearly in our definitions."

Of course the last-minute change to the definition of equivalent partially 
addresses this issue since it implies that the suitability of an equivalent 
may depend on the disability and the state of available technology: "In the 
context of this document, the equivalent must fulfill essentially the same 
function for the person with a disability (at least insofar as is feasible, 
given the nature of the disability and the state of technology), as the 
primary content does for the person without any disability." (Definition of 
"equivalent", 5 May 1999 WCAG document). 

B - Candidate Equivalent Versus Validated Equivalent

Yet it also may be worthwhile to note somewhere (not necessarily in the 
WCAG document) that for the sake of simplicity, the document does not 
distinguish between candidate equivalents and validated equivalents. I 
attempted to address this issue in a note in my revision of the 24 March 
1999 WCAG draft (http://etsr.digitalchainsaw.com/wcagpub/r990324a.htm). 
Following is an excerpt.

"Ignored Distinctions"

"For simplicity, usage of "equivalent" sometimes ignores distinctions that 
are worthwhile keeping in mind."

"Candidate Equivalents versus Validated Equivalents"

"There is a distinction between what we might term "candidate equivalents" 
and "validated equivalents."

"Basically, the issue, is: "At what point does a candidate equivalent 
become a validated (true or unqualified) equivalent?" For example, when 
text is placed in the "alt" attribute of the IMG element, it is perhaps 
formally _more_ correct to say that "The text serves as a _candidate_ text 
equivalent for the image" than to say "The text serves as the text 
equivalent for the image." The reason for this distinction is that, 
according to the formal definition of equivalent, one does not have an 
equivalent unless it is actually capable of being delivered in the 
requisite way(s) (in the case of text equivalents: synthesized speech, 
braille, visually-displayed text) and that it actually serves essentially 
the same communication purpose for the users. By this strict standard, one 
might argue, the text does not serve as a text equivalent until has been 
validated in some way with users. Thus, pursuing this line of reasoning, 
the Web content developer provides the text that serves as a _candidate 
text equivalent_ for the image. He should achieve a certain level of belief 
or obtain some amount of evidence that it will actually render properly in 
the requisite ways (synthesized speech, braille, or visually-displayed 
text) for the target users before he asserts it to be a validated, true, or 
unqualified "text equivalent".

"The use of the phrase "may serve" in the sentence: "An audio clip of a 
person speaking a text aloud may serve as an auditory equivalent for 
written text" reflects the conditional nature of the functional adequacy 
that the audio clip will have when presented to actual users. 

"In common usage, it is expected that this an other distinctions will be 
often be ignored. (They are sometimes ignored in current guidlelines 
document.) For example, the term "text equivalent" is used instead of 
"candidate text equivalent" even though there may be little or no evidence 
that to validate it as a true (or unqualified) text equivalent."

--
It seems to me that any conformance claim implies a claim that equivalents 
are either validated or validatable. It does not imply that any specific 
techniques (e.g., Appendix A - Validation) have actually been followed. Of 
course, use of the techniques may make the claim more easily defensible.

=============================
Eric G. Hansen, Ph.D.
Development Scientist
Educational Testing Service
ETS 12-R
Rosedale Road
Princeton, NJ 08541
(W) 609-734-5615
(Fax) 609-734-1090
E-mail: ehansen@ets.org

Received on Tuesday, 14 September 1999 15:51:44 UTC