W3C home > Mailing lists > Public > public-wai-evaltf@w3.org > April 2012

Re: Conceptual Model of Web Accessibility Evaluation link - fyi

From: Detlev Fischer <detlev.fischer@testkreis.de>
Date: Wed, 11 Apr 2012 18:21:31 +0200
Cc: "public-wai-evaltf@w3.org" <public-wai-evaltf@w3.org>
Message-Id: <CA3D965B-9775-40FD-807C-AE8E9C7066D8@testkreis.de>
To: "Boland Jr, Frederick E." <frederick.boland@nist.gov>
Hi Tim, hi list,

This is an interesting model and quite complex, too. I wonder to what  
extent it can aid us in the Evaluation Methodology. The conceptual  
object of "accessibility rule" introduces a specific requirement on a  
level *below* success criteria, but probably *above* the level of  
individual WCAG technique (in the sense that the WCAG Quickref often  
specifies "use GXX AND one of the following techniques"). Then, there  
is a meta object "ruleset" which references WCAG SC as external  
reuqirement ID. All this is probably the result of quite a thorough  
ordering/ categorisation activity but I simply wonder whether it makes  
sense to take these extra concepts on board.

Having said that, I think might be a good practical case for combining  
atomic tests related to a SC in a checkpoint. (For SC 1.3.1, think of  
inspecting the source code for lack, or wrong use, of semantic mark-up  
in editorial content).

So how can we evaluate the usefulness of the conceptual model for WCAG- 
EM? It might be best to think of an typical evaluation scenario  
(something simple to start with) and then think of how the AInspector  
conceptual model would operate in testing and reporting, and to what  
extent in could be prescribed (if at all).

One reason for my caution is that the link between "accessibility  
rules" and specific techniques is not an easy one. All atomic test  
results must live with the caveat that the SC might be met another  
way, which means the accessibility rule must aggregate different  
options / techniques for meeting it. So I wonder what "required rule"  
really means.

Down the line, rule results are themselves subject to aggregation,  
both on the level of rule result and evaluation result across target  
HTML-objects, and this introduces further complexity. It reminds me a  
bit of the UWEM 1.0 score function - will anyone understand what goes  
on under the bonnet? How is criticality reflected? What confidence can  
we have in the aggregated result?, etc. But this is a another can of  
worms...

Sorry, I really don't want to come across as pessimistic - I encourage  
everyone to look at the AInspector conceptual model. Then we might  
discuss its potential application in WCAG-EM tomorrow.

Regards,
Detlev

On 11 Apr 2012, at 16:22, Boland Jr, Frederick E. wrote:

> Shared by permission of Jon Gunderson of the Web Accessibility Best  
> Practices Group (NOTE: This is a draft and may change over time):
> https://trac.ainspector.org/design/wiki/interaction-design/conceptual-model
>
> Thanks and best wishes
> Tim Boland NIST

-- 
Detlev Fischer
testkreis - das Accessibility-Team von feld.wald.wiese
c/o feld.wald.wiese
Borselstraße 3-7 (im Hof)
22765 Hamburg

Tel   +49 (0)40 439 10 68-3
Mobil +49 (0)1577 170 73 84
Fax   +49 (0)40 439 10 68-5

http://www.testkreis.de
Beratung, Tests und Schulungen für barrierefreie Websites
Received on Wednesday, 11 April 2012 16:15:03 GMT

This archive was generated by hypermail 2.3.1 : Friday, 8 March 2013 15:52:13 GMT