Re: Proposal for ISSUE-40 Skolemization

From: Richard Cyganiak <richard@cyganiak.de>
Subject: Re: Proposal for ISSUE-40 Skolemization
Date: Wed, 18 May 2011 14:09:05 -0500

> Peter,
> 
> On 18 May 2011, at 17:19, Peter Frederick Patel-Schneider wrote:
>> 3.2 URI-based Vocabulary and Node Identification
> 
> All changes fine with me as proposed.
> 
>> [[ADD:
>> 6.6.1 Replacing blank nodes with IRIs
>> 
>> Particular IRIs and literals occuring in RDF graphs can each be easily
>> identified in both the RDF abstract syntax and in RDF concrete syntaxes.
>> However, a particular blank node occuring in an RDF graph cannot be so
>> identified,
> 
> 1. I'd prefer not to mention literals here, as it's about replacing
> blank nodes with IRIs. 
> 
> 2. This says “IRIs can be easily identified”, which is a bit tautological.
> 
> Perhaps a different approach:
> 
> [[ADD: Blank nodes do not have identifiers in the RDF abstract
> syntax. The blank node identifiers introduced by some concrete
> syntaxes have only local scope and are purely an artifact of the
> serialization. In situations where stronger identification is needed,
> …]] 

This would be acceptable. 

>> … systems MAY systematically transform some or all of the blank nodes
> in an RDF graph into IRIs.  Systems wishing to do this SHOULD mint a
> new, globally unique IRI (a Skolem IRI) for each blank node so
> transformed. 
> 
> 
> Ok
> 
>> Implementors should realize that this transformation changes the meaning
>> of an RDF graph (but this change is generally not harmful).
> 
> That sounds a bit scary. Perhaps:
> 
> [[ADD: This transformation does not change the meaning of an RDF
> graph, except “using up” the Skolem IRI.]] 

But this isn't true.

> or perhaps best not mention it at all. It's hard to explain what
> exactly the “possible harm” would be ... 

Perhaps.

> Best,
> Richard


peter

Received on Wednesday, 18 May 2011 19:23:13 UTC