W3C home > Mailing lists > Public > public-rdf-dawg-comments@w3.org > March 2006

Re: Comments on Feb 20 Working Draft of SPARQL [OK?]

From: Seaborne, Andy <andy.seaborne@hp.com>
Date: Tue, 14 Mar 2006 16:48:41 +0000
Message-ID: <4416F3E9.4000703@hp.com>
To: Reto Krummenacher <reto.krummenacher@deri.org>
CC: public-rdf-dawg-comments@w3.org

On Fri, Feb 24, 2006 at 05:40:55PM +0100, Reto Krummenacher wrote:
 > Dear editors,
 > I have another set of minor comments on the SPARQL Query Language for RDF
 > working draft. Most of them are solely of editorial matter:
 > * first paragraph section 2: Combining tripleS gives a basic graph...

"Combining triple" => "Combining triple patterns"

 > * I would suggest to exchange 2.1.7 and 2.1.6. In my opinion
 > the query syntax is based on the data format used and not vice versa.

The previous sections (2.1.1-2.1.5) are about query syntax so it seem natural 
to conclude query syntax.  Also, the data syntax is just for this document so 
downplaying seems appropriate.

 > * 2.9 should probably start with "RDF defines A reification vocabulary..."

It reads right as it is to me, it does read as saying it is the only 
vocabulary.  But as it has caused confusion to you, I have changed it.

 > * 5.4 should probably say "...; or it passes all solutions without adding any
 > additional bindings." The 'any solutions' seems wrong to me.

Matching is defined per-solution so I changed it to: "the solution".

 > * Still 5.4 end: should OPTIONAL not be capitalized in the syntactic form 

Yes - fixed.

 > * 5.5 the second sentence of paragraph one is very confusing to read. Doesnt it
 > say exactly the same as the next sentence: "The outer optional graph
 > pattern..."? Could it may be make sense to mention that this basically means
 > that all varialbes in the outer graph patterns have to be bounded, doesnt it?

Deleted that sentence.

 > * End 5.5: the conclusion of the example is that the optional part is only
 > reached if there is a vcard:N predicate. Shouldnt it not also include a 
 > vcard:Given predicate, as it ?vc vcard:Given ?gname is also part of the outer
 > graph pattern?

Changed to:
By nesting the optional pattern involving vcard:Family, the query only matches 
these if there are appropriate vcard:N and vcard:Given triples in the data.

 > Small wording question: "the query only reaches these..." should IMHO be
 > "...reaches this...", as it refers to the optional graph pattern.

used "matches"

 > * Could it make sense to mention in 10.1.1 that projection is basically the
 > sequence modifier applied in all SELECT queries presented so far in the
 > document. Projection is basically the default modifier of SELECT, isnt it?

In 10.2 (SELECT) is does tie SELECT queries to projection.

 > * In 10.4.3 first sentence you mention that the output of DESCRIBE
 > is determined by the information publisher. Who or what is the
 > information provider? Is it the query service that provides the information 
 > to a user or rather the entity that actually published the information.
 > If I publish my foaf file and it is accessed over a SPARQL interface
 > I would expect that I am the publisher, however in my
 > opinion it is not clear how I could influence the DESCRIBE output
 > besides using the same blank node as subject of related
 > statements (cf. CBD)

The "information provider" is a term to express the whole collection of issues 
that go along with the provision and deployment of a SPARQL service.  It may 
well be that the people providing the data are different from the people 
running the service platform and they'll need to work together to get the 
service running but from the client's point of view it's what is avilable at 
the interface that matters.  Your example of your foaf file is a good one from 
the point of view of you making your foaf file available - it would take you 
and the service platform to do that.

 > * In 11: is there a reason why xsd:dateTime is in another font than the other
 > datatypes?

No. This has been fixed. Thank you.

 > * In the listing of 11.2 there is twice a redundant "will return" for 
logical or
 > and logical and

Also fixed.

 > * in 11.3 for example XPath is written as xpath.

Also fixed.

 >                                                  i also observed that sometimes
 > cannot is written in two words.

Both are, I believe, acceptable, but I have changed
complete structures can not be assumed in all RDF graphs
complete structures cannot be assumed in all RDF graphs

 > Thank you very much for reading, I hope it helps a bit to finalize the working
 > draft.

Indeed it does. Thank you kindly for your assistance.

Please respond indicating whether you are or are not satisfied with
this response. If you are, you can help our issue tracking system by
prefixing the subject of your response with [CLOSED] (where this
subject has [OK?]).

office: +81.466.49.1170 W3C, Keio Research Institute at SFC,
                         Shonan Fujisawa Campus, Keio University,
                         5322 Endo, Fujisawa, Kanagawa 252-8520
         +1.617.258.5741 NE43-344, MIT, Cambridge, MA 02144 USA
cell:   +81.90.6533.3882

Feel free to forward this message to any list for any purpose other than
email address distribution.
Received on Tuesday, 14 March 2006 16:49:04 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:52:07 UTC