- From: Carlos Iglesias <carlos.iglesias@fundacionctic.org>
- Date: Wed, 21 Jun 2006 14:22:31 +0200
- To: "Shadi Abou-Zahra" <shadi@w3.org>
- Cc: <public-wai-ert@w3.org>
Hi Shadi, > > "The foaf:Agent class is the class of agents; things that do stuff." > > [http://xmlns.com/foaf/0.1/#term_Agent] > > Yes, just like earl:Software or dctype:Software. > > > > So, I don't find it proper to use it as TestSubject (they > do nothing). > > As I said before, I think that this class fits better as an > "Allowable > > Type" for earl:Assertor. > > Why don't do TestSubjects do anything? A Web applications > also do something... > > More importantly, imagine a browser that is being tested to > comply with UAAG or an authoring tool being tested against > ATAG. These are all examples of things that do stuff and that > may be a test subject in a specific context. Right now we are > restricted to earl:Software only (or dctype:Software if we > decide to swap). What I mean is that earl:Software (or dctype:Software) is not playing an Agent role when it is used as TestSubject because Test Subjects always play a "passive" role (although earl:Software could play an "active" role as Assertors for example) So, in brief: TestSubjects are ALWAYS passive roles --> can't be Agents Assertors, for example, are active roles --> can be Agents Software can be TestSubject (passive --> not an Agent) or Assertor (active --> an agent) IMO again, Agent could be an allowable type for earl:Assertor but not for earl:TestSubject Regards, CI.
Received on Wednesday, 21 June 2006 12:23:17 UTC