Re: ISSUE-89

Mark Birbeck wrote:
> Hi Ivan,
> 
>> You mean: the skip flag is not set if @property is around, right?
> 
> Right. One quick way to do it would be to change the wording at the
> end of step 4, from this:
> 
>   "otherwise, if [parent object] is present, [new subject] is set to
> that and the
>   [skip element] flag is set to 'true';"
> 
> to this:
> 
>   "otherwise, if [parent object] is present, [new subject] is set to
> that, and if
>   @property is not present, the [skip element] flag is set to 'true';"
>

Yep!

> This works because we're already testing for @rel and @rev at the top.
> 
> If we did this we'd be completing triples multiple times, so the next
> task would be to try to solve that. I think the only way that we can
> do that would be to set some kind of flag in the evaluation context
> that can be tested for in various settings, but it's made quite
> complex by the fact that you need to know how you arrived at the
> subject.
> 
> Anyway, that would move this out of the realm of being a purely
> editorial change, so I guess it would depend on whether there are
> other changes that are needed. If not, then we might just accept that
> we get these duplicate triples for now, since they are harmless.
> 

In my view: this is an implementation issue. It is all right to put a 
note/warning into the text to that effect, but I do not believe avoiding 
this is part of the spec. RDF is a set, so specification-wise it is 
absolutely kosher to add the same triple several times. Adding new flags 
and mechanisms into the text makes it very difficult to read, manage, 
edit, for no really good reason...

My two pence:-)

Ivan

> Regards,
> 
> Mark
> 

-- 

Ivan Herman, W3C Semantic Web Activity Lead
Home: http://www.w3.org/People/Ivan/
PGP Key: http://www.ivan-herman.net/pgpkey.html
FOAF: http://www.ivan-herman.net/foaf.rdf