RE: [limit per resource] implementation experience and/or concerns?



> -----Original Message-----
> From: public-rdf-dawg-request@w3.org [mailto:public-rdf-dawg-
> request@w3.org] On Behalf Of Steve Harris
> Sent: 17 April 2009 13:15
> To: Kjetil Kjernsmo
> Cc: public-rdf-dawg@w3.org
> Subject: Re: [limit per resource] implementation experience and/or
> concerns?
> 
> On 17 Apr 2009, at 09:51, Kjetil Kjernsmo wrote:
> 
> > On Friday 17 April 2009 06:34:13 Lee Feigenbaum wrote:
> >> (just looking to get a bit of background discussion in advance of
> >> tuesday, to save time on the teleconference)
> >>
> >> http://www.w3.org/2009/sparql/wiki/Feature:LimitPerResource

> >
> > So, this is our main feature. If I had only one vote this would be
> > it. :-)
> >
> >> LIMIT, in SPARQL, normally just limits the total number of results
> >> returned. Combined with DISTINCT, it limits the total number of
> >> distinct
> >> tuples returned. Combined with an aggregate/grouping extension, it
> >> can
> >> limit the total number of groups returned.
> >
> > Yes, but that's only for SELECTs, while we work mainly with DESCRIBE
> > queries,
> > so it has wider applicability than that.
> 
> Hm, interesting. The wiki page gives SELECT examples, and it's a
> little hard (for me at least) to imagine what it would look like in
> DESCRIBE land. Is it something you could do in CONSTRUCT? Or do you
> have a lot of per-schema smarts in your DESCRIBE implementation?

If feature:subselect
then
   DESCRIBE ?x { SELECT ?x { ... } }

This cleanly split the design of creating the table of ?x's from applying DESCRIBE or CONSTRUCT.

 Andy

> 
> >> 2/ does anyone implement this elsewhere (e.g. in SQL)?
> >
> > It isn't very relevant to SQL, since SQL databases typically deal
> > with fixed
> > schemas and uniform data. Since one of the main reasons why people
> > go to RDF
> > and SPARQL is its usefulness in dealing with heterogenous data, this
> > is a
> > feature that would distinguishing it from the rest, so in that sense
> > I don't
> > see this question as very relevant. :-)
> 
> Agreed, but there are parallels. Subqueries + aggregates in SQL can do
> something like this.
> 
> - Steve
> 
> --
> Steve Harris
> Garlik Limited, 2 Sheen Road, Richmond, TW9 1AE, UK
> +44(0)20 8973 2465  http://www.garlik.com/

> Registered in England and Wales 535 7233 VAT # 849 0517 11
> Registered office: Thames House, Portsmouth Road, Esher, Surrey, KT10
> 9AD
>

Received on Friday, 17 April 2009 12:24:19 UTC