W3C home > Mailing lists > Public > w3c-wai-er-ig@w3.org > June 2002

Re: Experiments With Abbreviated EARL As An Authoring Tool

From: Charles McCathieNevile <charles@w3.org>
Date: Sat, 22 Jun 2002 06:14:31 -0400 (EDT)
To: "Sean B. Palmer" <sean@mysterylights.com>
cc: <w3c-wai-er-ig@w3.org>
Message-ID: <Pine.LNX.4.30.0206220608050.22462-100000@tux.w3.org>

Well, this form is clearly going to be easier to use for writing some EARL,
although I am not convinced that the generality is as easily amintained as
the example suggests - do you have an expanded example to show how it works
with different types of subjects, and with multiple results?

One question is whether you can convert RDF EARL into this form. (Most EARL
is likely to be produced by tools, so the exact details of the syntax are
somewhat less important than they are for people authoring by hand). Beyond a
more compact syntax, what would that enable?

Although it seems that EARL as XML can be validated against a schema, it
loses some of its flexibility and extensibility. It would seem to that you
also preclude the possiblity of merging EARL results with detailed
information about the tests, which we could have if we stuck with RDF. (It is
possible to write an XML schema for recording the tests, but those I have
seen so far seem fairly limited).

Certainly I think this is worth pursuing.



On Sat, 22 Jun 2002, Sean B. Palmer wrote:

  I noticed that 95% of EARL uses the same fraction of features that EARL has
  to offer:-

  * Assertor name
  * Assertor email
  * Test subject+date
  * Test subject type
  * Validity property
  * Test case ID
  * Optional: Test case label
  * Optional: note

  but nevertheless the XML representation of this is too verbose; it puts me
  off, let alone newcomers:-

  <rdf:RDF xmlns="http://www.w3.org/2001/03/earl/0.95#"
     xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#" >
     <name>Bob Bobbington</name>
     <email rdf:resource="mailto:bob@example.org"/>
     <asserts rdf:parseType="Resource">
       <rdf:subject rdf:parseType="Resource">
           <documentation rdf:resource="http://www.w3.org/TR/xhtml11"/>
       <rdf:object rdf:parseType="Resource">
         <id rdf:resource="http://www.w3.org/TR/xag#cp2_3"/>
       <note>It was an architectural decision</note>
  <rdf:Description rdf:about="http://www.w3.org/org/TR/xag#cp2_3">
     <rdfs:label>2.3 Use XLink and XPointer</rdfs:label>

  Of course, there is always N3, but people don't like the fact that it's not
  XML based. Here's the equivalent of the above in N3:-

  @prefix : <http://www.w3.org/2001/03/earl/0.95#> .
  @prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
  @prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .

  [ a :Assertor; :name "Bob Bobbington";
    :email <mailto:bob@example.org>;
    :asserts [ rdf:subject [ a :Language; :date "2002-06-15";
                 :documentation <http://www.w3.org/TR/xhtml11> ];
               rdf:predicate :fails;
               rdf:object [ :id <http://www.w3.org/TR/xag#cp2_3> ];
               :note "It was an architectural decision" ] ] .
     rdfs:label "2.3 Use XLink and XPointer" .

  So, after a little hacking, I was able to come up with an XML syntax for
  these features that was just a third of the size of the M&S XML RDF, and
  half the size of the N3:-

  <EARL xmlns="http://infomesh.net/2002/earlxml">
  <Assertor name="Bob Bobbington" email="mailto:bob@example.org" >
   <asserts Language="http://www.w3.org/TR/xhtml11" date="2002-06-15" >
     <fails id="http://www.w3.org/TR/xag#cp2_3"
         label="2.3 Use XLink and XPointer"
         note="It was an architectural decision" />

  Not only that, but in about half an hour I had produced an XSLT sheet that
  can convert the abbreviated syntax into XML RDF. The XSLT sheet is attached
  as <<xml2rdf.xsl>>, and a test run is given:-

  Abbreviated input: <<abbrev-xml.xml>>
  M&S XML RDF output: <<abbrev-out.xml>>

  I just used saxon:-

     saxon abbrev-xml.xml xml2rdf.xsl > abbrev-out.xml

  So there you have it. If anyone finds this useful and wants me to come up
  with some sort of schema for it (check out the transformation for the basic
  semantics of the abbreviated form for now, or ask me), I may cobble one

  Kindest Regards,
  Sean B. Palmer
  @prefix : <http://purl.org/net/swn#> .
  :Sean :homepage <http://purl.org/net/sbp/> .

Charles McCathieNevile    http://www.w3.org/People/Charles  phone: +61 409 134 136
W3C Web Accessibility Initiative     http://www.w3.org/WAI  fax: +33 4 92 38 78 22
Location: 21 Mitchell street FOOTSCRAY Vic 3011, Australia
(or W3C INRIA, Route des Lucioles, BP 93, 06902 Sophia Antipolis Cedex, France)
Received on Saturday, 22 June 2002 06:14:33 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:01:33 UTC