W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > July to September 2001

Re: RE Checkpoint 3.4 again

From: Anne Pemberton <apembert@erols.com>
Date: Tue, 31 Jul 2001 12:52:01 -0400
Message-Id: <>
To: Kynn Bartlett <kynn@reef.com>, "gregory j. rosmaita" <oedipus@hicom.net>
Cc: <w3c-wai-gl@w3.org>
Kynn and Gregory,

         Take a step back from the mathematical concept of "equal", and 
take a gander at the social science concept of "equal" .... More Equal and 
Less Equal have no mathematical meaning, but they have precise meanings in 
social science ... history, philosophy, politics, etc. Unlike the meaning 
of "equal" in math, which is either a right or wrong answer (unless we're 
playing games in other bases), in the social sciences "equal" is a very 
subjective term which can and is hotly debated in interesting times and 
places. "Equal" is a perception, as well as an absolute truth.


At 07:44 AM 7/31/01 -0700, Kynn Bartlett wrote:
>At 11:46 PM 7/30/2001 , gregory j. rosmaita wrote:
> >GJR: take it from someone for whom the issue isn't theoretical: a
> >book-on-tape produced from a print edition with requisite quality control
> >(as performed by NLS or RFB&D, in the states, where the reading of the text
> >is monitored for accuracy during the recording process, then double-checked
> >through a read-along) is an equivalent...
>Except you have to remember that most people in the states aren't
>aware of this, and part of the -value- of book-on-tape to them is the
>fact that someone famous is reading it.  Thus there's an inherent
>inequivalency in the whole process; different emotional responses,
>different ways of using the book, different economics.
> >second, you are implying that only an "absolute equivalent" is acceptable;
> >that may be the case for someone to whom the question of equivalencies is
> >theoretical, or, phrased less combatively, less than a P1 issue; [...]
>quote "These two are more equal than those two." endquote
>See, the problem is with natural language understanding of the term
>"equivalent," and that's where a lot of people get hung up in WCAG
>1.0 and other docs.  How often have we seen questions over "what
>should the right text equivalent be for this image?"
>The difference between equivalents and parallels:
>1.  An "equivalent" is meant to be "equal", and most people think in
>      terms of mathematics where equality is an absolute.  It's thought
>      of as crazy to say something is "more equal" than something else.
>      With the term "equivalent" we have to battle peoples' preconceived
>      notions of what "equal" means.
>2.  The term "equivalent" also puts too much emphasis on one "sainted"
>      format as being the primary, and the rest have to be "equivalent"
>      to that one.  This hides the true semantics and content, which is
>      really what we are looking for.  It's less important that alt
>      text be equivalent to the _image_ as much as it's equivalent to
>      the _content_ and _semantics_ of the image.
>3.  The term "parallel" rightly puts the emphasis back on the content
>      and semantics by giving a sense that BOTH of these things (the
>      image and the text, say) -- or ALL of them -- are just expressions
>      of the underlying information and information architecture.  A
>      "parallel" to an image serves the same function and conveys the
>      same information as the image, while an "equivalent" of an image
>      focuses on trying to "equate" to that particular image, which is
>      just an expression of the concepts BEHIND the image.
> >let's stop couching the conversation in terms of absolutes -- WCAG is not,
> >nor could it ever be "absolute" -- no individual or group could ever
> >definitively permute all of the potential use scenarios...
>But this isn't about absolutes, it's about how we look at "equivalents",
>text or otherwise, and it's about how we write our document so that it
>can be understood better.  I think -- especially when we are dealing
>with issues _more_ complex than alt text -- that a sense of "parallel"
>may be more understandable and may emphasize the important parts (the
>semantics, e.g.) better than "equivalent."  That's all.
>Kynn Bartlett <kynn@reef.com>
>Technical Developer Liaison
>Reef North America
>Accessibility - W3C - Integrator Network
>Tel +1 949-567-7006

Anne Pemberton

Received on Tuesday, 31 July 2001 13:01:50 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 16 January 2018 15:33:38 UTC