W3C home > Mailing lists > Public > www-rdf-interest@w3.org > December 2002

RE: cwm/n3 and naming blank nodes?

From: Seaborne, Andy <Andy_Seaborne@hplb.hpl.hp.com>
Date: Sun, 1 Dec 2002 19:59:52 -0000
Message-ID: <5E13A1874524D411A876006008CD059F0586CB32@0-mail-1.hpl.hp.com>
To: "'www-rdf-interest@w3.org'" <www-rdf-interest@w3.org>


Given RDF's open world assumption, the absence of the "p:died" does not mean
that no such statement has been made somewhere or will be made.  You might
merge in some more RDF which did contain the statement "ab:jdoe p:died

There may well be other, better solutions, but one approach would be to give
end dates to all people - and make it a bNode where it is as yet unknown
((are literals resources at the moment?)).  Appointments also always have
end dates, although it might be unknown.  Your application can choose to
interpret "[] db:end []" as unknown and mean repeat forever (i.e. repeat
until known to have passed). 

There is now only one rule that triggers on each contact record so I think
you only get one appointment per contact.


-----Original Message-----
From: Norman Walsh [mailto:Norman.Walsh@Sun.COM] 
Sent: 1 December 2002 19:15
To: www-rdf-interest@w3.org
Subject: cwm/n3 and naming blank nodes?

Hash: SHA1

Hello world,

Maybe there's already an easy way to do this, so let me just describe the
problem and see what the world says. I'm hacking about with my PIM in RDF.

Let's say I have a contact like this:

ab:jdoe a ab:Contact;
  ab:firstname "Jane";
  ab:surname "Doe";
  p:born "1970-01-01" .

- From such a record, I want to imply a yearly repeating appointment:

this log:forAll :p, :s, :o, :t, :u, :l, :k, :m.

{ :p a ab:Contact .
  :p p:born :o } log:implies { [ a db:Appointment ;
                                   db:begin-date :o ;
                                   db:repeat [
                                      rdf:type db:Repeat ;
                                      db:frequency "1";
                                      db:type "Yearly" ] ] } .

That's all well and good, but now suppose I have

ab:jdoe2 a ab:Contact;
  ab:firstname "John";
  ab:surname "Doe";
  p:born "1970-01-01" ;
  p:died "1999-12-31" .

For such a record, I want to imply a yearly repeating appointment with an
end date:

{ :p a ab:Contact .
  :p p:born :o .
  :p p:died :s } log:implies { [ a db:Appointment ;
                                   db:begin-date :o ;
                                   db:repeat [
                                      rdf:type db:Repeat ;
                                      db:end :s ;
                                      db:frequency "1";
                                      db:type "Yearly" ] ] } .

But if I have both of these rules, then I get two appointments, one with an
end and one without.

If I could provide an identifier for the birthday event, I think I could
just add the db:end property if there was a p:died property, but I can't
think of any way to give names to these newly constructed nodes.

Am I overlooking something? Is there some way to construct names
automatically? (For example, if ":p" in the expression is ab:jdoe, I'd be
quite content to imply a new "db:jdow-birthday" node that I could
subsequently refer to.)

                                        Be seeing you,

- -- 
Norman.Walsh@Sun.COM    | We look back on our life as a thing of broken
XML Standards Architect | pieces, because our mistakes and failures are Web
Tech. and Standards | always the first to strike us, and outweigh Sun
Microsystems, Inc.  | in our imagination what we have accomplished
                        | and attained.--Goethe
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.7 <http://mailcrypt.sourceforge.net/>

Received on Sunday, 1 December 2002 15:00:06 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:07:43 UTC