Summary of teleconference on Wednesday 22 August, 2007

Dear Group,

Please find below a summary of today's ERT WG teleconference discussion.

Attendees: Johannes, Reinhard, Shadi

Topic: how to describe content in HTTP-in-RDF and in EARL

Open Questions:
  - Should we drop earl:Content?
  - Should http:Content go into a different namespace?


Summary:

#1. http:Content
While http:Content is actually quite specific in HTTP-in-RDF and refers 
to the body of an HTTP message, it is actually quite generic and could 
be reused in other contexts as well. For example as an RFC822 body, for 
an XML document, or just a text flat-file.


#2. Using http:Content in EARL
In EARL we can use http:Content directly as a test subject ("this is the 
content that was tested"):

...
   <earl:testsubject rdf:resource="content01"/>
...
   <http:Content rdf:ID="content01">
...

Note that the (optional) headers to the content are not part of the 
assertion but they could still be in the report and queried quite simply 
using an RDQL statement like "select * from (? http:body 'content01')".


#3. What if the HTTP headers are relevant to the EARL assertion?
In EARL we can but do not have to point to http:Content. We could just 
as well point to the http:Response which would contain the content with 
its headers:

...
   <earl:testsubject rdf:resource="response01"/>
...
   <http:Response rdf:ID="response01">
...

Note that this means the assertion is made on the whole response which 
includes the headers. This is important for any pointer methods used, 
they are all relative to the earl:TestSubject.


#4. earl:Content
The intention of earl:Content was to have a class that is more specific 
than earl:TestSubject yet general enough to address any Web content such 
as HTTP, FTP, etc. The problem is that by doing so, it becomes redundant 
to http:Content and creates compatibility issues.


#5. Should we drop earl:Content?
If it is clear from HTTP-in-RDF that http:Content is generic enough, 
then it can be used directly from EARL. Since http:Content seems to 
satisfy all the requirements there may be little need to keep 
earl:Content as well. -Anyone have objections to removing it?


#6. Should http:Content go into a different namespace?
Is there a need to move http:Content (and all its properties and 
siblings etc) into a different namespace? Argument: since it is so 
generic and represents more than an HTTP message body, it could be 
clearer to move it elsewhere. -Anyone have an objection to this?


All open for discussion...

Regards,
   Shadi


-- 
Shadi Abou-Zahra     Web Accessibility Specialist for Europe |
Chair & Staff Contact for the Evaluation and Repair Tools WG |
World Wide Web Consortium (W3C)           http://www.w3.org/ |
Web Accessibility Initiative (WAI),   http://www.w3.org/WAI/ |
WAI-TIES Project,                http://www.w3.org/WAI/TIES/ |
Evaluation and Repair Tools WG,    http://www.w3.org/WAI/ER/ |
2004, Route des Lucioles - 06560,  Sophia-Antipolis - France |
Voice: +33(0)4 92 38 50 64          Fax: +33(0)4 92 38 78 22 |

Received on Wednesday, 22 August 2007 16:30:34 UTC