[last call, S2] text content, etc.

There is a glossary entry 

   Text content, non-text content, text element, non-text element, text
          equivalent, non-text equivalent

The concepts presented in this glossary entry are incompatible with the sense
of the WCAG and should be purged from the UAAG.

Text content in not defined as anything that is comprehensible in the three
modes mentioned; on the contrary, text is _recognized_ as an act of technology
assessment as _sufficient_ for tri-mode comprehensibility, and is the only
medium which can generally be regarded as tri-mode usable in and of itself.

It would be possible to create a tri-mode accessible bundle of alternatives by
expressing the content in a tactile-graphic bump array for embossing, an audio
recording, and a print-image pixel array as well.  This contains no text, but
it is not readily accessible because the graphics modes of embossers are not
standardized and one is not guaranteed that the deaf-blind user will have an
embosser; only a refreshable Braille display.  Hence the only safe policy
is to
deliver text.

But this is by no means an equation, a by-definition equivalence.  It is a
one-way sufficiency condition.  Text is sufficient for tri-mode usability; it
is not equivalent to tri-mode usability.

Text is far from "by definition" tri-mode accessible.  It is only tri-mode
accessible by dint of consumers investing reasonable chunks of change in
screen
readers and refreshable Braille displays.  

But for the consumers who need this to access a Web page, they need it to
access anything on the computer, so it is reasonable for the Web content
provider to consider that this capability is readily available for the
transformation of text content.

But not by definition.

Text alternatives are required of the author in the WCAG because the author
can
tell the difference.

Text vs. non-text is not a distinction that the User Agent is required to
recognize; only the markup in the formats that frame alternatives or
equivalence groups. 

The format specifications create the equivalence group recognition rules for
the user agent.  The author recognizes non-text content and  provides text
equivalents; but the User agent only sees equivalents.  Only in the case where
the equivalents appear in attributes (such as ALT) can the User Agent discern
that they are text or non-text, and it does not matter at least for the
satisfaction of checkpoint 2.3 [which should be viewed in this paragraph as
having been generalized to include the first of two rules expressed in
checkpoint 2.4 -- see comment to 2.4].

Al
--
Usage in headers.  Comments in response to the last call request for comments
have been classified S1, S2, or E based on the following rough scale:

S1: Substantive matter of the first (highest) criticality or importance to the
mission of the document.  The standard set is ineffective, the document is
self
contradictory, etc.

S2: Substantive matter of a somewhat lower criticality.  The document is hard
to comprehend, does not align well with related WAI documents, etc.

E: Editorial matters.  Not regarded as substantive.

Re: 

User Agent Accessibility Guidelines 1.0

W3C Working Draft 23 October 2000

   This version:
         
[9]<http://www.w3.org/TR/2000/WD-UAAG10-20001023>http://www.w3.org/TR/2000/W
D-UAAG10-20001023

Received on Monday, 13 November 2000 00:57:25 UTC