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

Re: Minor error in SPARQL Update 4.2.5

From: Axel Polleres <axel.polleres@deri.org>
Date: Tue, 15 Nov 2011 09:51:23 +0100
Cc: "SPARQL Working Group" <public-rdf-dawg@w3.org>
Message-Id: <B19E5297-8C19-4D11-A255-FB553AA0605C@deri.org>
To: Andy Seaborne <andy.seaborne@epimorphics.com>
This appears to be still unanswered and I think I now fixed those all...

On 17 Oct 2011, at 23:37, Andy Seaborne wrote:

> 
> 
> On 17/10/11 19:57, Axel Polleres wrote:
> > Thanks Andy for spotting this.
> > Fixed in the editor's draft on CVS, see http://www.w3.org/2009/sparql/docs/update-1.1/Overview.xml
> >
> > As for the stray '{' that actually was missing the closing '}' and meant to denote the Dataset-UNION over a set (of datasets).
> > I added a clarifying remark for this slight abuse of notation now after the definition of Dataset-UNION:
> >
> > --------
> > <p>Note that, in the following, when we write Dataset-UNION(<em>X</em>  ) where<em>X</em>  = {DS<sub>1</sub>,DS<sub>2</sub>,... ,DS<sub>n</sub>} is a set of datasets, we understand this as a shorthand for Dataset-UNION(DS<sub>1</sub>, Dataset-UNION(DS<sub>2</sub>( ... , Dataset-UNION(DS<sub>n</sub>),{}) ... ).
> > </p>
> > --------
> 
> 4.2.4  Dataset
> has a typesetting problem.  4th bullet.

done.

> 
> 4.3.1 Insert Data Operation
> has another notation.  Dataset-UNION(*GS*, ...)
> 
> Ditto 4.4.1 and other places
> 
> (yes, it's clear informally what it means, but ti's not what it says)
> 


This is IMO covered by the note following definition 4.1.1 :

"Note:

We will use GS for the graph store, but sometimes also - synonymously - for the RDF dataset corresponding to the current graph store content in subsequent definitions." 

> 
> This could do with tidying up editorially:
> 
> 4.1.1 says a Graph Store is
> 
> GS = {DG, (iri1, G1),  ... , (irin, Gn) }
> 
> but it's not written like this later which is a bit confusing.
> 
> 4.2.5 writes a dataset as:
> 
>   DS={DG} union {(irii, Gi) | 1 ≤ i ≤ n}

Mathematically, they both mean the same, but for my personal taste, the latter notation 
comes handy e.g. in Definition 4.4.2 (DropOperation). 
I added a note explaining the alternative notation now directly after Definition 4.1.1.
If that's not sufficient, I'd prefer to just stick with the latter notation.

> (typo: "... } be and RDF Dataset")

fixed.

Thanks for the comments again,

Axel

> 
> 
> 4.3.5, 4.4.1, 4.4.2 write:
> 
> Let GS = {DG} union {(irii, Gi) | 1 ≤ i ≤ n} and graphNames(GS) = { irii
> | 1 ≤ i ≤ n}
> 
> Using the same way helps the reader pick up the sameness.  The
> definition form in 4.1.1 is better.
> 
>         Andy
> >
> > Axel
> >
> >
> > On 14 Oct 2011, at 20:37, Andy Seaborne wrote:
> >
> >> The definition in 4.2.5 Dataset( QuadPattern, P, DS, GS )
> >>
> >> uses:
> >>
> >> Dataset(QuadPattern, μ)
> >>
> >> but should be
> >>
> >> Dataset(QuadPattern, μ, DS, GS)
> >>
> >> and there is a stray {
> >>
> >> I think is should be:
> >>
> >> Dataset(QuadPattern, P, DS, GS ) =
> >>         Dataset-UNION( Dataset(QuadPattern, μ, DS, GS)
> >>                        | μ in eval'(DS(DG),P) )
> >>
> >>          Andy
> >>
> >>
> >
> 
Received on Tuesday, 15 November 2011 08:51:54 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:15:47 GMT