W3C home > Mailing lists > Public > public-rdf-dawg@w3.org > January to March 2011

Re: SPARQL Query review

From: Andy Seaborne <andy.seaborne@epimorphics.com>
Date: Mon, 21 Mar 2011 10:32:30 +0000
Message-ID: <4D87293E.9070207@epimorphics.com>
To: "Polleres, Axel" <axel.polleres@deri.org>
CC: SPARQL Working Group <public-rdf-dawg@w3.org>
Comments processing other than those for sections 11, 12, 18.4, 18.5

> * Abstract
>
>   "The results of SPARQL queries can be results sets or RDF graphs."
>
>    s/results sets/result sets/

Done.

>
> * Status of this Document
>
>   "The new features are:"
>
>   I'd add here to make it clear
>
>   "The new features in SPARQL 1.1 Query are:"

Done.

> * Section 1.1
>
>   "Section 17 defines SPARQL's extensible value testing framework"
>
> suggestion:
>
>   "Section 17 defines SPARQL's extensible function library for creating values and value testing"

Changed to:

Section 17 defines SPARQL's extensible value testing and expression
framework. It presents the functions and operators that can be used to 
constrain the values that appear in a query's results and also calculate 
new values to be returned by a query.

> * Section 2.3.1
>
>   not sure why, but the table for the second solution doesn't seem to have a right border in my browser.

For me, the tables appear erratic, esp borders both internal and 
external, if viewed at anything below 100%.  Browser-ism.

> * Section 2.5
>
>   shall "expressions in the SELECT clause" link to Section 16.1.2 ?

Link added, text changed, example with BIND added.

> * Section 4
>
>   "The full grammar  is given in appendix A."
>
> should be
>
>   "The full grammar  is given in section 19."

Done.

> * Section 9.1
>
>   In the table, may I suggest not to use italic font for '|' in order to avoid possible confusion with "/"

Done - only elt and iri are <i> throughout the table.

> * Section 9.3
>
>
>   "whereas if the query were written out to include the intermediate variable (?a),
> 	no rows in the results are duplicates:"
>
>   is not a sentence, suggest just to write:
>
>   "If the query were written out to include the intermediate variable (?a), no rows in the results are duplicates:"

> * Section 14
>
>   Should this section mention something about the status of this feature,
>   i.e. mention that it is optional? (cf. closed ISSUE-1, i.e. by leaving the grammar in, we haven't really
>   specified it in two separate documents, but have the syntax fully in SPARQL1.1 Query, whereas we've only "outsourced"
>   the semantics of the feature.

Could do but the current text has the advantage of being neutral as to 
the status of federated query and so its nature is entirely in its own 
document, as are evaluation issues etc.  Left as is.

I'd hope that the parsing of SERVICE is done by all engines and a useful 
error issues, not "parse error - unrecognized - SERVICE".

> * Section 16.2.2
>
> "where app:customDate identified an extension function to turn the data format into an xsd:dateTime RDF term."
> -suggestion->
> "where app:customDate identified an extension function to turn an application-specific custom date format into an xsd:dateTime
> RDF term."

This is SPARQL 1.0 so lets' leave it.

Did fix:
   data -> date
and tided the example query.

> * A.2 Other References
>
>   "[TURTLE] Turtle - Terse RDF Triple Language, Dave Beckett."
>
>    as we refer to the Team submission, we should add Tim Berners-Lee as author.

Done.

	Andy
Received on Monday, 21 March 2011 10:33:07 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:15:45 GMT