RE: Contracted words in Hebrew

John M Slatin writes:
 > 
 > Gregg,If disambiguation is the problem we're trying to solve, that's
 > fine. But it won't helpsolve that problem if we describe it incorrectly,
 > and the term "contracted words" is not an accurate term for Hebrew words
 > in which vowels are indicated by diacritical marks.
 > 
 > Contractions in English, such as "isn't," "It's," and so forth, are
 > formed when *two* words are joined together and some of the letters are
 > delted and replaced with an apostrophe.  That's not the ame thing at all
 > as the case of single words in languages that use diacritics to indicate
 > vowels, etc.
 > 
Correct. What we essentially want, I think, is that every word in the
document be unambiguously identifiable so that it can be looked up in
a dictionary, pronounced by a text to speech system, or otherwise
processed. As Gregg mentioned, this may demand nothing of the author;
but where the orthography is ambiguous, and there is no reliable
automated means of disambiguating it, authorial intervention is needed
to solve the underlying problem. This point, incidentally, generalizes
to all of our guidelines at level 1: the author need only supply that
information which tools need in order to adapt the content and thereby
make it more accessible to the user.

Another way of making this point is that WCAG is the "Web Content
Accessibility Guidelines", not the "what the author has to do"
guidelines. Many of the requirements are destined to be implemented by
software, particularly authoring tools; but they can equally well be
met by the author directly in the case of "hand-coded" content, or
where the author is using an editor that offers a high degree of
control over the eventual coding.

What the guidelines become, in practice, via the Authoring Tool
Accessibility Guidelines, is a series of prompts, options, warnings,
error messages and so on, in an authoring system.

Received on Monday, 10 May 2004 01:18:46 UTC