Comment on Formal Semantics: transformation processor knowledge

Comment on http://www.w3.org/TR/2008/WD-powder-formal-20080709/, Formal 
Semantics

I am sorry but the whole of 3.2 is a bit unclear to me. It reads as if 
the transformation processor had extra knowledge about the targets of 
references, ie, that it had access to several files at the same time. 
How else can I read:

[[[
A typical case might be where the referred to descriptor set represents 
compliance with an established code or standard, see the Pre-Defined 
Descriptors section of the Description resources document [DR]. If these 
conditions are satisfied, then referring to an external descriptor set 
using src can be equivalent to the following example in which the 
rdf:resource construct is used. Here, a processor SHOULD assume that the 
referred to resource is an OWL class, even if it cannot access or 
process it.
]]]

is this a correct supposition? I am not really sure. I wonder whether 
more explicit knowledge should be present in the POWDER-XML level, in 
the hands of the author, to disambiguate here. (Eg, the default case 
would be to refer to rdf:resource, and let the author use another 
attribute for the rdfs:seeAlso case; something like that.)

I may got it wrong, though, in which case editorial help in the text 
would, well, help:-)

Cheers

Ivan

-- 

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