- From: <MDaconta@aol.com>
- Date: Tue, 10 Feb 2004 23:04:59 EST
- To: davemenendez@optonline.net, MDaconta@aol.com
- Cc: zednenem@psualum.com, mike@daconta.net, www-rdf-interest@w3.org
- Message-ID: <184.25b68e76.2d5b03eb@aol.com>
Hi Dave, Thanks for the detailed explanation. I see my mistake now. Appreciate the help, - Mike In a message dated 2/10/04 6:43:12 PM, davemenendez@optonline.net writes: > > In a message dated 2/7/2004 9:51:04 PM Eastern Standard Time, > > David Menendez <zednenem@psualum.com> writes: > > >If you want to say that <#cell> has a <#part> relationship with > > ><#cell_wall> and <#cytoplasm>, then what you want is: > > > > > > <owl:Class rdf:ID='cell'> > > > <part rdf:resource='#cell_wall'/> > > > <part rdf:resource='#cytoplasm'/> > > > </owl:Class> > > > > That is exactly what I wanted to do and tried to do by > > declaring the Object Properties with the domain of #cell; > > however, it failed loading in protege with the error message > > that "part" was a duplicate. Of course, this could just be > > a bug in protege, or, my syntax could have been wrong. > > I see. If you want to say that <#part> is a relationship between one class > and another, you could declare it like so: > > <owl:ObjectProperty rdf:ID='part'> > <rdfs:domain rdf:about='&owl;Class'/> > <rdfs:range rdf:about='&owl;Class'/> > </owl:ObjectProperty> > > (I'm using &owl; so I don't have to type the full URI. It isn't required.) > > What you had tried before was to declare <#part> to be a relationship > between a resource of type <#cell> and a resource that is both a <#cell-wall> and a > <#cytoplasm>. You are allowed to do that, so Protege is probably guessing > that you really meant to do something else. Or else it's a bug. > > > Do you have a particular OWL validator that you run > > your syntax against? > > There's one on the web, but I don't have its URI handy. I think you can get > to it via the OWL home page. Your problem was more at the RDF/RDFS level, > though. You might want to read through the RDF primer if you haven't already. > > ------------------------------- Michael C. Daconta Chief Scientist, APG, McDonald Bradley, Inc. www.daconta.net
Received on Tuesday, 10 February 2004 23:05:31 UTC