W3C home > Mailing lists > Public > public-wai-ert-tsdtf@w3.org > November 2006

Linking 'location' and 'technique': models and examples

From: cstrobbe <Christophe.Strobbe@esat.kuleuven.be>
Date: Wed, 29 Nov 2006 15:44:10 +0100
Message-ID: <1164811450.456d9cbabac05@webmail2.kuleuven.be>
To: public-wai-ert-tsdtf@w3.org
Hi,

I had an action item to provide examples of two ways to link a 
'location' with one or more 'technique' elements. This issue has been 
discussed a few times before [1] [2], and there was another discussion 
on the last conference call. We did not reach consensus, so here's a 
description of the three models that were discussed.
(Note that we need to allow references to more than one technique for 
some success criteria [3], so the first proposal in [1] is not 
appropriate.)


Model A: using ID and IDREFS

Keep the sequence 'locations' (containing one or more instances of 
'location') and 'techniques' (containing one or more instances if 
'technique') as in the current schema, but link them by means of ID and 
IDREFS type attributes.
(And rename the 'id' attribute on the 'rule' element to 'xlink:href'.)
In practice, this would mean that we add an 'id' attribute to 
'technique' (do we need a naming convention for this ID?) and a 
'techrefs' attribute to 'location'. (These new attributes would be 
optional in TCDL 2.0, but we can make them obligatory in our usage 
document.)
The attachment sc3.1.1_l1_001_20061129_modelA.xml illustrates this 
model.


Model B: nesting 'locations' inside 'technique'

The rationale for this change is that locations can be seen as 
properties of a technique.
Compared to model C (nesting 'techniques' inside 'location'), model B 
also has the advantage that it is not necessary to repeat the 
'technique' element for each location where the technique is used. It 
was also pointed out however, that there should be only one instance of 
each technique or failure, because test samples should be atomic.
The attachment sc3.1.1_l1_001_20061129_modelB.xml illustrates this 
model.


Model C: nesting 'techniques' inside 'location'

The rationale for this model is that locations identify where barriers 
occur and that 'techniques' provides additional (outside TSD TF: 
optional) information about these locations. Outside TSD TF, not very 
location or barrier will map to a technique or failure documented by 
WCAG 2.0, so other users may not want to nest 'locations' inside 
'technique'.
(I have not created an example for this model.)


[1] http://lists.w3.org/Archives/Public/public-wai-ert-tsdtf/2006Nov/
0029.html
[2] http://lists.w3.org/Archives/Public/public-wai-ert-tsdtf/2006Oct/
0024.html
[3] "Re: Minimum number of techniques in metadata": http://lists.w3.org/
Archives/Public/public-wai-ert-tsdtf/2006Oct/0072.html

Best regards,

Christophe Strobbe

-- 
Christophe Strobbe
K.U.Leuven - Departement of Electrical Engineering - Research Group on 
Document Architectures
Kasteelpark Arenberg 10 - 3001 Leuven-Heverlee - BELGIUM
tel: +32 16 32 85 51
http://www.docarch.be/ 

Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm




Received on Wednesday, 29 November 2006 14:44:33 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:18:34 GMT