review http:http://www.w3.org/2009/sparql/docs/update-1.1/Overview.xml
CVS version 02/01/2010

Change summary:

Compared to the FPWD mainly the MODIFY clause has been removed in the generic
DELETE/INSERT construct, and shortcuts for DELETE and INSERT have been defined in case the WHERE part is identical to the DELETE or INSERT clause. 
In DELETE and INSERT clauses, as opposed to FPWD, not only BGPs are allowed, but also GRAPH graph patterns in order to model insertion/deletion into/from different graphs in a single DELETE/INSERT construct. The respective examples in Section 3 have been updated.

Detailed Comments:

* Section 5.1.3:

the modify_template grammar misses something in the production for graph_template, that is VarOrIRIref is missing after GRAPH. For publication, that should at least be fixed and at least an todo/editor's note should be added that all the grammar productions or grammar snippets should be formatted the same.
 
Concretely (not critical to do this all now, as long as the note is added), I suggest to just reuse the productions from the Query grammar, i.e.

a) replace:

  modify_template :: = template | graph_template
  graph_template ::= GRAPH { template }

with:

  modify_template :: = ConstructTriples | graph_template
  graph_template ::= 'GRAPH' VarOrIRIref '{' ConstructTriples '}'


b) replace anywhere in grammars in the document:
 WHERE { pattern }
with 
 WHERE GroupGraphPattern

 triples
with 
 TriplesBlock
?

c) Also I suggest the keywords should be single quoted in syntax definitions that are supposed to be gramamr snippets, i.e. in the blocks of


* Section 4.2: 

 I am unsure to be honest what we shall do for the ISSUES here... some are closed, but actually some recent mails [1,2] indicate that they should be left open, e.g. ISSUE-21...

1. http://lists.w3.org/Archives/Public/public-rdf-dawg/2009OctDec/0628.html
2. http://lists.w3.org/Archives/Public/public-rdf-dawg/2009OctDec/0674.html

If time, I would like to go through the issues mentioned in these mails in the next TelCo to decide what is really to be closed and what noe.

* Finally, Appendix B should at least have an editor's note/todo that it doesn't yet reflect the new syntax in Section 5.1.3 before publication.