[Fwd: minor technical on 10.3 Constructing an output graph]

Comments on this text have come up a few times now.

SPARQL does define what happens in the case of an unbound variable in a 
CONSTRUCT template (no triple generated).  As we have no means for conveying 
warnings (and I'm not sure what a client can do with them except print them 
which is unfriendly for a library inside some application), it's really all 
left to the local implementation. I propose we delete the text from rq23.

	Andy

-------- Original Message --------
Subject: minor technical on 10.3 Constructing an output graph
Resent-Date: Thu, 12 Jan 2006 17:35:50 +0000
Resent-From: public-rdf-dawg-comments@w3.org
Date: Thu, 12 Jan 2006 09:35:36 -0800
From: Fred Zemke <fred.zemke@oracle.com>
To: public-rdf-dawg-comments@w3.org


10.3 Constructing an output graph
It says that under certain circumstances "a warning may be generated".
Two issues:
1. The document does not contain the usual definintion of "may".
2. It would be preferable to nail this behavior down, ie, it would
be preferable to simply say "a warning is given".  Though the mechanism
for delivering the warning is not specified...

Fred Zemke

Received on Tuesday, 17 January 2006 11:33:09 UTC