W3C home > Mailing lists > Public > public-rdf-dawg@w3.org > January to March 2004

RE: Use case: Find the email address of "John Smith"

From: Seaborne, Andy <andy.seaborne@hp.com>
Date: Wed, 24 Mar 2004 10:13:01 -0000
Message-ID: <E864E95CB35C1C46B72FEA0626A2E80801EA189F@0-mail-br1.hpl.hp.com>
To: Dan Connolly <connolly@w3.org>
Cc: RDF Data Access Working Group <public-rdf-dawg@w3.org>



-------- Original Message --------
> From: Dan Connolly <>
> Date: 23 March 2004 21:36
> 
> On Fri, 2004-03-19 at 09:03, Seaborne, Andy wrote:
> > == Task
> > 
> > A user wants to find the email address for a person whose name is
> > "John Smith".  The user has an address book in RDF that contains all
> > the information needed.  The address book uses the FOAF vocabulary
> > [1]. 
> 
> This looks fine to me. Since it got some support last week, I'm
> inclined to regard silence at this point as agreement from
> most members of the WG.
> 
> Anybody who doesn't like this will please let us know, preferably
> before I send out the agenda on Wednesday for Thursday's teleconference.
> 
> One or two more like this and it will be time to build an outline
> of the document and start to flesh it out, yes?

Agreed - it will be time to build an outline of a document.

> 
> Hmm... one shortcoming I can see is that it doesn't explictly
> identify a requirement. Hmm... another imperfection is: we already
> have LDAP; ideally our use cases should show why the existing
> technologies are not enough. Perhaps a query like "find the
> email address of the person in <picture23.jpg>" would be more
> compelling?

Adding requirement: no problem.  I have been avoiding doing this, assuming
that we will collect some use cases then extract requirements.  This allows
agreeing the UC and then discussing the requriments. But if this is seen as
generally stable, extracting requirements is fine.

LDAP: Assumes a planned deployment.  And my address book is an RDF file, not
a RDF-fronted LDAP server.

	Andy

> 
> 
> > == Importance of DAWG
> > 
> > A query is more efficient in terms of application programmer time than
> > writing custom code to a lower level interface.  The resulting
> > execution can be more efficient because an application might choose a
> > simple way of doing it where as a toolkit can invest in optimization.
> > 
> > If the query is issued to a remote data source then a way of
> > serialising one or more operations requests is needed.  A query
> > language is one way of providing this. 
> > 
> > Applications that use the DAWG rec. can be switched from a local
> > address book to an interface to an LDAP server (say) with no change
> > to the application other than pointing it to a different source.
> > 
> > Information sources that use the DAWG rec. can be accessed by a wider
> > set of applications with no extra work.
> > 
> > == Other
> > 
> > The query has two parts - locate the resource and extract some
> > information from that resource.  The presentation of the results is
> > the mailbox email address.  As people may have several mailboxes,
> > there can be multiple results even if the conceptual task is to find
> > one. 
> > 
> > The resource for the person could be a bnode or a labelled resource.
> > 
> > An example query:
> > 
> > SELECT ?mbox
> > WHERE
> >   (?x foaf:name "John Smith")
> >   (?x foaf:mbox ?mbox)
> > USING
> >   foaf  FOR <http://xmlns.com/foaf/0.1/>
> > 
> > -------
> > 
> > [1] http://xmlns.com/foaf/0.1/
> >     FOAF namespace document
Received on Wednesday, 24 March 2004 05:23:05 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:15:18 GMT