You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@jena.apache.org by "McGibbney, Lewis John" <Le...@gcu.ac.uk> on 2011/05/04 14:59:47 UTC
Constructing a SPARQL query store
Hi list,
Out of curiosity I am wondering if anyone can discuss any implementations or practice of building a query store so that SPARQL queries are not constructed in an ad-hoc one-off fashion. I am thinking about the inherent difficulties presented to new (and unfamiliar) users by any HTTP engine which supports SPARQL querying. I we consider some store for SPARQL queries which can be picked and used instead of the requirement for system users to construct then execute their own queries; this would remove the high complexity of structured queries, instead enabling easy usage of powerful SPARQL queries stored within a central repository.
As I have only begun to learn SPARQL I think it would be good practice for me to begin storing queries and undertake benchmarking over the next few months to evaluate how my knowledge of the language is improving. This will also enable me to evaluate if I am making full use of the RDF dataset's which I am working with.
Thanks for any comments
Lewis
Glasgow Caledonian University is a registered Scottish charity, number SC021474
Winner: Times Higher Education’s Widening Participation Initiative of the Year 2009 and Herald Society’s Education Initiative of the Year 2009.
http://www.gcu.ac.uk/newsevents/news/bycategory/theuniversity/1/name,6219,en.html
Winner: Times Higher Education’s Outstanding Support for Early Career Researchers of the Year 2010, GCU as a lead with Universities Scotland partners.
http://www.gcu.ac.uk/newsevents/news/bycategory/theuniversity/1/name,15691,en.html
RE: Constructing a SPARQL query store
Posted by "McGibbney, Lewis John" <Le...@gcu.ac.uk>.
Thank you Bill
This has given me a lot to think about...
Lewis
________________________________________
From: Bill Roberts [billroberts1966@gmail.com] On Behalf Of Bill Roberts [bill@swirrl.com]
Sent: 10 May 2011 20:24
To: jena-users@incubator.apache.org
Subject: Re: Constructing a SPARQL query store
Hi Lewis
You make a good point, that it is difficult for someone new to SPARQL to start writing queries, and some kind of system that could help new users would definitely make RDF datasets easier to use and hence more valuable.
Another related problem is that someone coming to a new RDF dataset for the first time, even if they are experienced with SPARQL, needs to work out how the data is structured, what vocabularies are used etc, before they can write sensible queries.
A big challenge to writing some kind of user-friendly query builder is the flexibility of a general-purpose query language like SPARQL. You can write a query to ask almost anything, so it's impossible to make a comprehensive list of pre-canned queries that a user can pick and choose from and very difficult to design a user interface that lets people construct queries. Typically attempts to do this end up being harder to use than just writing the query 'longhand'.
I think it is definitely useful and good practice for the curators of a dataset to publish some example queries that are relevant to a particular dataset - that serves the double purpose of helping people understand what's in the dataset and also giving them examples of working SPARQL queries they can adapt.
The Linked Data API approach (see http://code.google.com/p/linked-data-api/ and http://data.gov.uk/blog/guest-post-developers-guide-linked-data-apis-jeni-tennison) maps URL patterns to some of the more common SPARQL queries that users of the data are likely to want - and on http://data.gov.uk/linked-data they also show the details of the query that was autogenerated to retrieve the data displayed on the page.
Good examples are definitely very powerful in helping people to learn how to use SPARQL and your suggestion of a repository of queries could be a useful educational tool as well as a useful time saving system. (This presentation is very good by the way http://www.cambridgesemantics.com/2008/09/sparql-by-example). But any solution will need to trade off ease of use and flexibility. It's very difficult to have both.
Cheers
Bill Roberts
On 10 May 2011, at 21:02, McGibbney, Lewis John wrote:
> Hi list,
>
> Wondering if anyone had any ideas of the post below... Understandably the silence implies that no-one is implementing something of this nature however I am keen to get feedback whether something of this nature has been considered.
>
> Thank you
>
> Lewis
> ________________________________________
> From: McGibbney, Lewis John [Lewis.McGibbney@gcu.ac.uk]
> Sent: 04 May 2011 13:59
> To: jena-users@incubator.apache.org
> Subject: Constructing a SPARQL query store
>
> Hi list,
>
> Out of curiosity I am wondering if anyone can discuss any implementations or practice of building a query store so that SPARQL queries are not constructed in an ad-hoc one-off fashion. I am thinking about the inherent difficulties presented to new (and unfamiliar) users by any HTTP engine which supports SPARQL querying. I we consider some store for SPARQL queries which can be picked and used instead of the requirement for system users to construct then execute their own queries; this would remove the high complexity of structured queries, instead enabling easy usage of powerful SPARQL queries stored within a central repository.
>
> As I have only begun to learn SPARQL I think it would be good practice for me to begin storing queries and undertake benchmarking over the next few months to evaluate how my knowledge of the language is improving. This will also enable me to evaluate if I am making full use of the RDF dataset's which I am working with.
>
> Thanks for any comments
>
> Lewis
>
> Glasgow Caledonian University is a registered Scottish charity, number SC021474
>
> Winner: Times Higher Education’s Widening Participation Initiative of the Year 2009 and Herald Society’s Education Initiative of the Year 2009.
> http://www.gcu.ac.uk/newsevents/news/bycategory/theuniversity/1/name,6219,en.html
>
> Winner: Times Higher Education’s Outstanding Support for Early Career Researchers of the Year 2010, GCU as a lead with Universities Scotland partners.
> http://www.gcu.ac.uk/newsevents/news/bycategory/theuniversity/1/name,15691,en.html
>
> Email has been scanned for viruses by Altman Technologies' email management service - www.altman.co.uk/emailsystems
>
> Glasgow Caledonian University is a registered Scottish charity, number SC021474
>
> Winner: Times Higher Education’s Widening Participation Initiative of the Year 2009 and Herald Society’s Education Initiative of the Year 2009.
> http://www.gcu.ac.uk/newsevents/news/bycategory/theuniversity/1/name,6219,en.html
>
> Winner: Times Higher Education’s Outstanding Support for Early Career Researchers of the Year 2010, GCU as a lead with Universities Scotland partners.
> http://www.gcu.ac.uk/newsevents/news/bycategory/theuniversity/1/name,15691,en.html
Email has been scanned for viruses by Altman Technologies' email management service - www.altman.co.uk/emailsystems
Glasgow Caledonian University is a registered Scottish charity, number SC021474
Winner: Times Higher Education’s Widening Participation Initiative of the Year 2009 and Herald Society’s Education Initiative of the Year 2009.
http://www.gcu.ac.uk/newsevents/news/bycategory/theuniversity/1/name,6219,en.html
Winner: Times Higher Education’s Outstanding Support for Early Career Researchers of the Year 2010, GCU as a lead with Universities Scotland partners.
http://www.gcu.ac.uk/newsevents/news/bycategory/theuniversity/1/name,15691,en.html
Re: Constructing a SPARQL query store
Posted by Bill Roberts <bi...@swirrl.com>.
Hi Lewis
You make a good point, that it is difficult for someone new to SPARQL to start writing queries, and some kind of system that could help new users would definitely make RDF datasets easier to use and hence more valuable.
Another related problem is that someone coming to a new RDF dataset for the first time, even if they are experienced with SPARQL, needs to work out how the data is structured, what vocabularies are used etc, before they can write sensible queries.
A big challenge to writing some kind of user-friendly query builder is the flexibility of a general-purpose query language like SPARQL. You can write a query to ask almost anything, so it's impossible to make a comprehensive list of pre-canned queries that a user can pick and choose from and very difficult to design a user interface that lets people construct queries. Typically attempts to do this end up being harder to use than just writing the query 'longhand'.
I think it is definitely useful and good practice for the curators of a dataset to publish some example queries that are relevant to a particular dataset - that serves the double purpose of helping people understand what's in the dataset and also giving them examples of working SPARQL queries they can adapt.
The Linked Data API approach (see http://code.google.com/p/linked-data-api/ and http://data.gov.uk/blog/guest-post-developers-guide-linked-data-apis-jeni-tennison) maps URL patterns to some of the more common SPARQL queries that users of the data are likely to want - and on http://data.gov.uk/linked-data they also show the details of the query that was autogenerated to retrieve the data displayed on the page.
Good examples are definitely very powerful in helping people to learn how to use SPARQL and your suggestion of a repository of queries could be a useful educational tool as well as a useful time saving system. (This presentation is very good by the way http://www.cambridgesemantics.com/2008/09/sparql-by-example). But any solution will need to trade off ease of use and flexibility. It's very difficult to have both.
Cheers
Bill Roberts
On 10 May 2011, at 21:02, McGibbney, Lewis John wrote:
> Hi list,
>
> Wondering if anyone had any ideas of the post below... Understandably the silence implies that no-one is implementing something of this nature however I am keen to get feedback whether something of this nature has been considered.
>
> Thank you
>
> Lewis
> ________________________________________
> From: McGibbney, Lewis John [Lewis.McGibbney@gcu.ac.uk]
> Sent: 04 May 2011 13:59
> To: jena-users@incubator.apache.org
> Subject: Constructing a SPARQL query store
>
> Hi list,
>
> Out of curiosity I am wondering if anyone can discuss any implementations or practice of building a query store so that SPARQL queries are not constructed in an ad-hoc one-off fashion. I am thinking about the inherent difficulties presented to new (and unfamiliar) users by any HTTP engine which supports SPARQL querying. I we consider some store for SPARQL queries which can be picked and used instead of the requirement for system users to construct then execute their own queries; this would remove the high complexity of structured queries, instead enabling easy usage of powerful SPARQL queries stored within a central repository.
>
> As I have only begun to learn SPARQL I think it would be good practice for me to begin storing queries and undertake benchmarking over the next few months to evaluate how my knowledge of the language is improving. This will also enable me to evaluate if I am making full use of the RDF dataset's which I am working with.
>
> Thanks for any comments
>
> Lewis
>
> Glasgow Caledonian University is a registered Scottish charity, number SC021474
>
> Winner: Times Higher Education’s Widening Participation Initiative of the Year 2009 and Herald Society’s Education Initiative of the Year 2009.
> http://www.gcu.ac.uk/newsevents/news/bycategory/theuniversity/1/name,6219,en.html
>
> Winner: Times Higher Education’s Outstanding Support for Early Career Researchers of the Year 2010, GCU as a lead with Universities Scotland partners.
> http://www.gcu.ac.uk/newsevents/news/bycategory/theuniversity/1/name,15691,en.html
>
> Email has been scanned for viruses by Altman Technologies' email management service - www.altman.co.uk/emailsystems
>
> Glasgow Caledonian University is a registered Scottish charity, number SC021474
>
> Winner: Times Higher Education’s Widening Participation Initiative of the Year 2009 and Herald Society’s Education Initiative of the Year 2009.
> http://www.gcu.ac.uk/newsevents/news/bycategory/theuniversity/1/name,6219,en.html
>
> Winner: Times Higher Education’s Outstanding Support for Early Career Researchers of the Year 2010, GCU as a lead with Universities Scotland partners.
> http://www.gcu.ac.uk/newsevents/news/bycategory/theuniversity/1/name,15691,en.html
FW: Constructing a SPARQL query store
Posted by "McGibbney, Lewis John" <Le...@gcu.ac.uk>.
Hi list,
Wondering if anyone had any ideas of the post below... Understandably the silence implies that no-one is implementing something of this nature however I am keen to get feedback whether something of this nature has been considered.
Thank you
Lewis
________________________________________
From: McGibbney, Lewis John [Lewis.McGibbney@gcu.ac.uk]
Sent: 04 May 2011 13:59
To: jena-users@incubator.apache.org
Subject: Constructing a SPARQL query store
Hi list,
Out of curiosity I am wondering if anyone can discuss any implementations or practice of building a query store so that SPARQL queries are not constructed in an ad-hoc one-off fashion. I am thinking about the inherent difficulties presented to new (and unfamiliar) users by any HTTP engine which supports SPARQL querying. I we consider some store for SPARQL queries which can be picked and used instead of the requirement for system users to construct then execute their own queries; this would remove the high complexity of structured queries, instead enabling easy usage of powerful SPARQL queries stored within a central repository.
As I have only begun to learn SPARQL I think it would be good practice for me to begin storing queries and undertake benchmarking over the next few months to evaluate how my knowledge of the language is improving. This will also enable me to evaluate if I am making full use of the RDF dataset's which I am working with.
Thanks for any comments
Lewis
Glasgow Caledonian University is a registered Scottish charity, number SC021474
Winner: Times Higher Education’s Widening Participation Initiative of the Year 2009 and Herald Society’s Education Initiative of the Year 2009.
http://www.gcu.ac.uk/newsevents/news/bycategory/theuniversity/1/name,6219,en.html
Winner: Times Higher Education’s Outstanding Support for Early Career Researchers of the Year 2010, GCU as a lead with Universities Scotland partners.
http://www.gcu.ac.uk/newsevents/news/bycategory/theuniversity/1/name,15691,en.html
Email has been scanned for viruses by Altman Technologies' email management service - www.altman.co.uk/emailsystems
Glasgow Caledonian University is a registered Scottish charity, number SC021474
Winner: Times Higher Education’s Widening Participation Initiative of the Year 2009 and Herald Society’s Education Initiative of the Year 2009.
http://www.gcu.ac.uk/newsevents/news/bycategory/theuniversity/1/name,6219,en.html
Winner: Times Higher Education’s Outstanding Support for Early Career Researchers of the Year 2010, GCU as a lead with Universities Scotland partners.
http://www.gcu.ac.uk/newsevents/news/bycategory/theuniversity/1/name,15691,en.html