review of SWC section 2 (ACTION 518)

Reviewing "Symbols in RIF Versus RDF/OWL" in SWC [1], it looks good.
The one thing I'd like to see would be a little more explanation of when
and how blank nodes are observably different from constants.  Right

> However, in contrast to blank nodes, which are essentially
> existentially quantified variables, RIF local symbols are constant
> symbols. 

maybe add something like this (which may be wrong)...

     In many applications and deployment scenarios, this difference is
     inconsequential, so RDF blank nodes can be considered to be
     equivalent to RIF local symbols.  However the difference can
     produce different results when an RDF graph is used in a
     non-assertional context, such as in a query pattern.

If this subject is covered elsewhere in a document, a link would be
fine, but I couldn't find it.

I did find this bit in Section 3:

>_ :x ex:hasName "John" . 
> saying that there is some blank node that has the name "John",

which should, I think, be:

     saying that there is something, denoted here by a blank node, which
     has the name "John" 

(it's not the *node* which has the name John.)

Also, a minor editorial point, I noticed the links to specific Use Cases
point to the TR version of the document.   I believe they should point
to the Wiki version, so they'll end up (when published) pointing to the
concurrently published version of UCR instead of the old one.

    -- Sandro

(This complete ACTION-518 [2])


Received on Tuesday, 24 June 2008 13:55:46 UTC