Re: Extension property: "using"

On Tue, 12 Apr 2005 00:20:45 +1000, Chris Ridpath  
<chris.ridpath@utoronto.ca> wrote:

>> Sidar adds a property to its EARL report called using...
>>
> How is this different from the "earl:tool" element?

"Tool" is a Class, or a type of thing, and "using" (http://...#usando is  
the URI, but it has an english label which is what I will use) is a  
Property, or a type of relation.

Because in RDF the basic model is that things have relations with other  
things (rather than XMLs basic model where elements have content including  
other elements - why XML has predictable structure and RDF doesn't, but  
RDF vocabularies can be put together as is and XML can't), the difference  
between a Class and a Property is that oe describes the type of a thing,  
and one describes the relations.

Tool is a subclass of Assertor, which means anything that is a Tool is an  
Assertor. Person s also a subclass of Assertor, which means that any  
earl:Person is an Assertor (foaf:Person doesn't have that distinction - a  
foaf:Person is not necessarily a kind of Assertor, although they can be.

The domain of "using" is defined as Assertor. In other words, anything  
that has the relation "using" to something is by definition of type  
Assertor. (A thing can of course be of many types at once - for example  
foaf:Person and earl:Assertor and cia:TerroristSympathiser and  
alQaeda:WarriorOfGod and so forth...).

I had thought about making the range of "using" an Assertor too - in other  
words, anything that an Assertor is "using" is also an Assertor, but then  
it struck me that someone might do a quality evaluation using a couple of  
browsers, which are not themselves Assertors. (This case becomes  
interesting when you are making assertions like "this page complies with  
the house style" across different platforms, and just doing a rough test).

So my typical case is that an Assertor who is a Person is "using" a tool  
to help them make decisions - whether it is a browser, a simulator, or  
Kynn's famous "20 pints of beer to simulate certain vision, motor and  
cognitive difficulties in combination" (although I am not sure of the URI  
for that last one :-).

However a tool might want to note that it is using a different tool for  
something - e.g. the W3C code validators. I am not sure of this - the  
normal idea would be to simply have the W3C validator itself make the  
particular relevant assertion.

So, a piece of EARL with this extension looks like the following:

<?xml version="1.0" encoding="utf-8"?>
<rdf:RDF xmlns:earl="http://www.w3.org/WAI/ER/EARL/nmg-strawman#"
  xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
  xmlns:foaf="http://xmlns.com/foaf/0.1/"
  xmlns:sidar="http://www.sidar.org/EARL/mas-earl.rdf#"
  xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#">

    <earl:Assertor rdf:about="#assertor">
    <rdf:type rdf:resource="http://xmlns.com/foaf/0.1/Person"/>
       <foaf:name>charles</foaf:name>
       <earl:name>charles</earl:name>
       <sidar:usando>
          <earl:Tool rdf:about="http://www.sidar.org/hera/">
             <earl:name>HERA</earl:name>
             <earl:email rdf:resource="mailto:hera@sidar.org"/>
          </earl:Tool>
      </sidar:usando>
    </earl:Assertor>

   <earl:Assertion rdf:about="#wcag1cp1.1">
    <earl:subject rdf:resource="#subject" />
    <earl:message>Looks like they're all fine to me</earl:message>
    <earl:result  
rdf:type="http://www.w3.org/WAI/ER/EARL/nmg-strawman#pass"/>
    <earl:mode  
rdf:resource="http://www.w3.org/WAI/ER/EARL/nmg-strawman#manual"/>
    <earl:testcase  
rdf:resource="http://www.w3.org/TR/WCAG10/#tech-text-equivalent"/>
    <earl:assertedBy rdf:resource="#assertor" />
   </earl:Assertion>

In other words a real person is making the claim (this is what mode manual  
says) and the assertor is indeed a person, who is using a tool.

(the about="http://...hera" doesn't tell you where to find the tool,  
although it happens to live at that URI. It's just an identifier.  
dc:location would more explicitly note where the tool can be found. It is  
there so if we have a second person using the tool and want to combine  
their results, as we can do in our beta version, we ust say
   <sidar:usando rdf:about="http://www.sidar.org/hera/"/>
Anrely on the fact that the rest of the characteristics have been defined  
in the first declaration.

(Exercise - make up a set of assertions with two assertors doing this, and  
look at the graph...)

Hope that covers it :-)

cheers

Chaals

> ----- Original Message ----- From: "Charles McCathieNevile"  
> <charles@sidar.org>
> To: "Charles McCathieNevile" <charles@sidar.org>; <public-wai-ert@w3.org>
> Sent: Sunday, April 10, 2005 2:14 AM
> Subject: Re: Extension property: "using"
>
>
>>
>> The RDF schema for this is at http://www.sidar.org/EARL/mas-earl.rdf -  
>> the relevant bit for this property is below:
>>
>> <rdf:Property rdf:about="http://www.sidar.org/EARL/mas-earl.rdf#usando">
>>    <rdfs:isDefinedBy  
>> rdf:resource="http://www.sidar.org/EARL/mas-earl.rdf#"/>
>>    <rdfs:domain  
>> rdf:resource="http://www.w3.org/WAI/ER/EARL/nmg-strawman#Assertor"/>
>>    <rdfs:label xml:lang="es">Usando</rdfs:label>
>>    <rdfs:label xml:lang="es">Using</rdfs:label>
>>    <rdfs:comment xml:lang="es">lo que ayuda al evaluador para decir que  
>> resultado poner</rdfs:comment>
>>    <rdfs:comment xml:lang="en">anything the evaluator uses to help  
>> deciding what result to put</rdfs:comment>
>> </rdf:Property>
>>
>> cheers
>>
>> Chaals
>>
>> On Sun, 10 Apr 2005 15:40:12 +1000, Charles McCathieNevile  
>> <charles@sidar.org> wrote:
>>
>>> Hi folks,
>>>
>>> Sidar adds a property to its EARL report called using, to identify  
>>> that although the results are done as mode manual, there is a tool  
>>> involved (in presenting the test cases and asking the question).
>> ...
>> -- Charles McCathieNevile                      Fundacion Sidar
>> charles@sidar.org   +61 409 134 136    http://www.sidar.org
>>
>



-- 
Charles McCathieNevile                      Fundacion Sidar
charles@sidar.org   +61 409 134 136    http://www.sidar.org

Received on Monday, 11 April 2005 17:20:51 UTC