W3C home > Mailing lists > Public > public-wai-ert@w3.org > September 2006

Re: earl location pointer methods

From: Jim Ley <jim@jibbering.com>
Date: Sun, 3 Sep 2006 18:56:41 +0100
Message-ID: <00c401c6cf82$4faec630$0302a8c0@Sniff>
To: <public-wai-ert@w3.org>

>Thanks for working on this Jim. I have some questions for todays 

>1. Are both properties "earl:instance-of-pointer" and
>"earl:pointerCollection" needed? Since the class
>earl:PointerCollection is a sub-class of earl:Pointer, could something
>like this work:

> <rdf:Property 
> rdf:about="http://www.w3.org/WAI/ER/EARL/nmg-strawman#subpointer">
>    <rdfs:label xml:lang="en">sub pointer</rdfs:label>
>    <rdfs:comment xml:lang="en">An instance of a pointer in another 
> pointer.</rdfs:comment>
>    <rdfs:domain 
> rdf:resource="http://www.w3.org/WAI/ER/EARL/nmg-strawman#Pointer"/>
>    <rdfs:range 
> rdf:resource="http://www.w3.org/WAI/ER/EARL/nmg-strawman#Pointer"/>
>  </rdf:Property>

It could, but I actually held back from it because it seemed like the 
subPointer and the pointer pointed to points of the same individual error 
(perhaps start and end of a bad element) rather than seperate instances of 
the same error, I am happy with either approach though.

>2. How about having separate earl:byteSnippet and earl:charSnippet
>classes? I agree the schema would be slightly more verbose but also
>more comprehensible and less ambiguous (for example the usage of the
>properties is not apparent from the schema). The schema for the
>proposed earl:Snippets is attached.

This is fine, and makes sense to me.

>3. I'm not sure I understand the use for the earl:Offset class. It has
>earl:charOffset and earl:byteOffset properties but is not used by

You're right, it shouldn't exist, it was a left over from when I had another 
level of indirection isn't the snippet class. Sorry.


Received on Sunday, 3 September 2006 17:57:13 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:55:54 UTC