Re: Review of SPARQL 1.1 Protocol

Hi Luke,
Thanks for the comments! My responses are below.

David Charboneau
dcharbon@us.ibm.com

public-rdf-dawg-request@w3.org wrote on 01/11/2010 06:33:44 PM:

> [image removed] 
> 
> Review of SPARQL 1.1 Protocol
> 
> Luke Wilson-Mawer 
> 
> to:
> 
> SPARQL Working Group
> 
> 01/11/2010 06:35 PM
> 
> Sent by:
> 
> public-rdf-dawg-request@w3.org
> 
> Hi all,
> 
> Here are my comments on SPARQL 1.1 Protocol, which I thought was 
> particularly clear and well written.
> 
> Most of my comments are superficial and minor, but there are a couple of 

> technical ones in there.  I'm not really familiar with WSDL, but I hope 
> my comments are of some use nonetheless.
> 
> Kind regards,
> 
> Luke
> 
> *1) Introduction*
> 
>     * Documents are named inconsistently in the first paragraph (I think
>       the update document is named correctly, but not query): "SPARQL
>       Query Language 1.1" and "SPARQL 1.1 Update".
> 
I fixed this, query is now referred to as SPARQL 1.1 Query Language

>     * The Update and Query documents are named here, but should other
>       documents which are also affected by the protocol, such as service
>       description and the rest interface, be mentioned here too? 
> 
I think that they should be mentioned, but I wasn't sure how to introduce
them yet. I decided that perhaps it would be better to wait for the next
round to address that.

> *2.1) SparqlProtocol Interface*
> 
>     * I know they've already been mentioned, but it would be good to
>       have a definition of the update faults somewhere.  Perhaps they
>       should be in the update document.
> 
I agree. Perhaps I guess we'll need to coordinate and address this in
the next round, too.

> *2.1.1.1.4) Determining The Base IRI*
> 
>     * It is mentioned that the Base IRI may come from, for example, a
>       SOAP envelope, but it doesn't seem clear (to me at least) whether
>       it should take precedence over a BASE IRI specified using the
>       query language.
> 
I don't know the answer here either. I agree that we should resolve this 
in the
next draft.

> *2.1.2.1) XXUPDATEXX in Message*
> 
>     * It would be nice to have an XML fragment here showing an
>       update-request, like the one in section 2.1.1.1.
> 

>     * INTO and FROM are mentioned, even though I think they have been
>       removed them from the update spec.  WITH is present in the update
>       spec but not here.
> 
I deleted the relevant sentence. It only made sense for the old syntax. 
WHERE
specifies a pattern, rather than specific graphs or graph variables like
INTO and FROM had, so I don't think that more needs to be said here.

>     * If the INTO in the older update syntax of INSERT INTO <uri> {} is
>       replaced by the user of GRAPH in the new syntax of INSERT {GRAPH
>       <uri> {}}, does this mean that GRAPH is now taking the role of
>       specifying the dataset?  Does this matter, and should it be
>       included here?  Perhaps I've missed something in the dataset
>       conversation.
> 
My reading would be that there aren't datasets for update. I've removed 
the text
until the issue of datasets in update is resolved.

> *2.1.3) Fault Messages*
> 
>     * 'MUST', 'MUST NOT' etc. are used in capitals here, which is
>       inconsistent with the rest of the document.
>
Fixed this. 

>     * In my opinion, it would be nicer to include the second paragraph
>       of each of 2.1.3.1, 2.1.3.2, 2.1.3.3, 2.1.3.4 at the top of the
>       section, otherwise the user has to read the same paragraph 4 
times.
>
I agree. I changed this for both the query and the update faults so that 
the text
repeats only twice, rather than six times; this is now specified in for 
the operations
in sections 2.1.1.3 Query Fault Messages and 2.1.2.3 Update Fault Messages
 
> *2.2) HTTP Bindings *
> 
>     * There is an XX missing off XXUPDATE.
> 
All XXX forms have been replaced by proper names.

>     * Perhaps "A conformant SPARQL Protocol service" ought to include a 
1.1.
> 
Fixed.
`
> *General (all none technical)*
> 
>     * It's stating the obvious, I know, but there are still a couple of
>       @@sec@@ parts in there.
Fixed.

>     * Excerpts could have more descriptive titles than "XML Schema
>       Fragment".
Added a "showing <insert-type-name-here>" suffix to each to better 
describe the
contents of the excerpt

> Ok    * Items in square brackets don't always point to a reference.
> 
> 
> 
> 
>