29 Aug 2007


See also: IRC log


Shadi, CarlosI, Reinhard
Johannes, CarlosV


"Content" in EARL and in HTTP-in-RDF

<shadi> http://lists.w3.org/Archives/Public/public-wai-ert/2007Aug/0012

SAZ: comments related to Content in EARL and HTTP-in-RDF summarized

CI: when http:Content came in?
... AFAIK is not part of HTTP and so I think is not a good idea to have it at HTTP-in-RDF

SAZ: AFAIR the idea came in at last F2F
... probably it was named Content instead of Body because the property was already body
... one of the possible solutions is to get the Content "concept" outside of both into a new document and reuse from there as CarlosV argued

<shadi> http:Body subClassOf rfc822:Content

SAZ: The first question is, should we drop earl:Content or not?
... does anybody fell we need to keep it?

CI: so, the idea is to drop earl:Content and use TestSubject instead to point to a Content class defined anywhere?

<shadi> <earl:testsubject rdf:resource="content01"/>

SAZ: Yes

<shadi> <earl:Content rdf:ID="content01">

<shadi> http://www.w3.org/TR/EARL10-Schema/#content

<shadi> example 13

<shadi> <earl:context rdf:resource="#httpRequest"/>

SAZ: earl:Content seems to be a not necessary in-between class

<shadi> http://www.w3.org/TR/HTTP-in-RDF/#body

<shadi> http://lists.w3.org/Archives/Public/public-wai-ert/2007Aug/0003.html

SAZ: http:Content is defined in quite a generic way now and could be reused
... if http:Content is generic enough earl:Content becomes redundant
... the question then is if going on with the http:Content is a "clean" solution
... also need to think how to separate it: a new section, a new document...

Inferred test results

<shadi> http://lists.w3.org/Archives/Public/public-wai-ert/2007Jul/0052

SAZ: Seems that there is a use case for an "inferred" result
... some people don't like the term itself

<shadi> http://www.w3.org/TR/EARL10-Schema/#testmode

CI: composition of tests is more a test case definition question, not a result one
... for me the best solution is not to have a inferred result type and specify rules on how to compose different results
... e.g. automatic+automatic=automatic and automatic+manual=manual
... need to think about the possible combinations to see if could be tricky

RR: you'll lose the details, how many manual, how many automatic...

SAZ: you will still have the details in other results if you want

Requirements for Pointer Methods in RDF

<shadi> http://lists.w3.org/Archives/Public/public-wai-ert/2007Jul/0042

<shadi> http://lists.w3.org/Archives/Public/public-wai-ert/2007Jul/0049

CI: don't like the idea of PointerGroup not beeing a Pointer subclass
... more difficult to reuse outside EARL
... no difference between Char and Byte Offsets

SAZ: more flexible

CI: could update the document with the changes

SAZ: would like to focus on the overall model

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.128 (CVS log)
$Date: 2007/08/29 15:33:20 $