W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > October to December 2005

RE: Navigational features

From: Gregg Vanderheiden <gv@trace.wisc.edu>
Date: Sat, 3 Dec 2005 23:39:57 -0600
To: "'Gez Lemon'" <gez.lemon@gmail.com>, "'Bailey, Bruce'" <Bruce.Bailey@ed.gov>
Cc: "'Guide Lines list'" <w3c-wai-gl@w3.org>
Message-ID: <00e001c5f895$2a619160$ee8cfea9@NC6000BAK>

Well -- that isn't really correct.

Validity is very much more than parsing unambiguously.   And conformance to
a standard is much more than validity.  

This SC is not as described below.  It is worded that way to check for a
known problem for accessibility (parsing) while not including full validity
which goes beyond access.  As does full conformance.  

It is interesting that you say that parsing unambiguously is meaningless
when it was cited as one of the major accessibility reasons for requiring
validity. 
 
The definition of 'programmatically determined' is also quoted wrong.


Please everyone - if you want to know the meaning of terms - use the
definitions that are provided in the guidelines' glossary at the bottom and
linked to from each of the success criterion.

 
Parsed unambiguously is also not an 'undefined' term as implied.  The term
comes from SGML. 


Gregg

 -- ------------------------------ 
Gregg C Vanderheiden Ph.D. 
Professor - Ind. Engr. & BioMed Engr.
Director - Trace R & D Center 
University of Wisconsin-Madison 


-----Original Message-----
From: w3c-wai-gl-request@w3.org [mailto:w3c-wai-gl-request@w3.org] On Behalf
Of Gez Lemon
Sent: Saturday, December 03, 2005 10:14 PM
To: Bailey, Bruce
Cc: Guide Lines list
Subject: Re: Navigational features


On 03/12/05, Bailey, Bruce <Bruce.Bailey@ed.gov> wrote:
>  As a point of reference, I have a bachelors degree in computer 
> science from a decent school, and have been posting to the IG since at 
> least 1998.  I regret that I have not sufficiently grokked the meaning 
> behind "programmatically determined."  I find this a little a little 
> frustrating, but perhaps it is just me?  In particular, how is that 
> phrase differentiated from my second favorite gem, "parsed unambiguously"?

I think I may be able to help you with this one, Bruce. Both terms are an
attempt to be technology neutral, but parsed unambiguously goes a step
further in trying to avoid the words conformance or validity in a guideline
that at a glance might fool some people into believing that validity and
conformance might be best practice in adhering to that guideline.
Programmatically determined means that elements and attributes can be
programmatically extracted from the document structure, whereas
unambiguously parsed is an effort to acknowledge that invalid content can
choke user-agents, but avoids the word validity at all costs.

The weakness of the term programmatically determined is that it's defined to
be technologically agnostic. The following is programmatically determinable,
but no known user agent would know what to make of it:

<summary id="overview">The Importance of Being Valid</summary>

The weakness of unambiguously parsed is that it's meaningless. It doesn't
require that markup conforms to any kind of schema or DTD, or that any other
type of technology renders correctly in a user agent that supports the
technology; merely that the resulting data structure is parseable.

The following snippet of HTML markup is not programmatically determinable,
as user agents wouldn't know what to do with it, as HTML is well-defined. It
can be parsed unambiguously, as the rules of parsed unambiguously are
non-existent, and any 999 experts would conclude the same meaning.

<label for="day month year">Date:</label>

Best regards,

Gez

--
_____________________________
Supplement your vitamins
http://juicystudio.com
Received on Sunday, 4 December 2005 05:40:24 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:47:40 GMT