wdrs:describedby edits

As highlighted on Friday, we need to correct an error in the 
specification of wdrs:describedby.

We're really sorry about this and grateful to the LOD community for 
pointing out the mistake.

The current namespace document [1] states that it has a range of 
wdrs:Document, i.e. a POWDER document. This is at odds with the 
intention and the wording of the Description Resources specification 
itself [2] (the one known as the DR Doc).

The plan is to:
- add text to the DR Doc's errata page [3], today, stating that we are 
correcting the error by removing the range restriction. Calls for 
comments on this may be made to this mailing list. New text will be 
proposed for the relevant section 4.1.4 [4]

- edit the namespace document to reflect this change.

The first of those is easy. The second is actually slightly more 
involved and requires the following changes:

1. The HTML representation needs to be amended so that the Document 
class is not shows as being in the range of wdrs:describedby

2. The RDF schema for wdrs:describedby currently says:

<rdf:Property rdf:about="#describedby">
   <rdfs:label xml:lang="en">
     described by
   </rdfs:label>
   <rdfs:comment xml:lang="en">
     An RDF property to link to a POWDER document. Provided
     for use in RDFa, ATOM etc.
   </rdfs:comment>
   <rdfs:isDefinedBy
      rdf:resource="http://www.w3.org/TR/powder-dr/#semlink"/>
   <rdfs:range rdf:resource="#Document"/>
</rdf:Property>

The comment needs to change to say something like:
"An RDF property to match the describedby relationship type introduced 
in http://www.w3.org/TR/powder-dr/#assoc-linking and formally defined in 
http://www.w3.org/TR/powder-dr/#appD, i.e. "The relationship A 
'describedby' B asserts that resource B provides a description of 
resource A. There are no constraints on the format or representation of 
either A or B, neither are there any further constraints on either 
resource."

3. The actual range restriction needs to be removed in both 
representations (HTML and RDF)

No further references to wdrs:describedby need amending. There is a 
reference to wdrs:describedby in the formal semantics document [5], but 
this is unaffected.

W3C process
===========
Notice that we are proposing to edit the namespace document immediately 
and add information about the error to the DR doc's errata page. Current 
W3C process [6] essentially requires that any change to a Recommendation 
that could affect implementations must be proposed by a working group. 
However:

- this is an error correction, not a change to the design of POWDER;
- the POWDER WG is no longer in operation;
- the change is fully backwards compatible;
- the only impact the correction could have would be on an 
implementation that inferred that the object of a wdrs:describedby 
predicate was a POWDER document. We think it's unlikely but if such an 
implementation exists, please let us know!
- Following review at the recent TPAC meeting, W3C is implementing a new 
community-based change-management process [7] that would enable the 
changes outlined here to be made without reforming the working group. 
The implementation of the new process will take a few months but no 
longer. In the meantime, the errata document will serve as an 
authoritative guide to the intention to edit the Recommendation.


[1] http://www.w3.org/2007/05/powder-s#describedby
[2] http://www.w3.org/TR/powder-dr/#assoc-linking
[3] http://www.w3.org/2007/powder/powder-errata
[4] http://www.w3.org/TR/powder-dr/#semlink
[5] http://www.w3.org/TR/powder-formal/#pp
[6] http://www.w3.org/2005/10/Process-20051014/tr.html#errata
[7] http://www.w3.org/2010/08/26-maintenance.html


-- 

Phil Archer
Former POWDER WG chair
http://philarcher.org/
@philarcher1

N.B. For POWDER-related issues neither my Talis nor my W3C Team 
affiliation is appropriate. Instead, I am allied to Institute of 
Informatics & Telecommunications (IIT), NCSR "Demokritos" for this.

Received on Monday, 8 November 2010 10:57:34 UTC