fullname property in Person vocabulary

23/05/2012

We're starting to use the Person Core Vocabulary v1.00 and noticed that the UML diagram mentions a "fullname" property, but it's not detailed in the spec. Instead, there's the foaf:name property.

Should we assume that person:fullname does not exist and use foaf:name ?

[EDIT] : Oooops, this should have been post in Person CV Issues list, not here...

Dominique Guardiola
http://quinode.fr/
 

Component

Documentation

Category

bug

Comments

Fri, 25/05/2012 - 07:46

Thanks for the comment Dominique. Naturally I'm delighted to hear that you're finding the vocabularies useful - I look forward to hearing more about what you have done with it, what works and what doesn't. Implemenatation experience is critical so do please share it.

The short answer to your question is yes: use foaf:name for a person's full name.

In more detail:

The conceptual model - as expressed in the UML diagram - is, if you like, "what the WG wants to see" - it's an expression of the properties considered important for describing a natural person (which can be extended in any given context but these are the core properties). It is a separate step then to create a schema that can be used in implementations. Best practice is always to re-use existing vocabularies wherever possible, particularly if those existing vocabularies are already in widespread use as FOAF already was even before Facebook started using it. Re-using existing vocabularies:

  • maximises interoperablity;
  • gives confidence in the vocabulary (these guys know what's already out there and are not re-invening the wheel).

We use other FOAF terms and those from schema.org an elsewhere to create the schema for the person vocabulary. Only where no existing vocabulary has a term that does the job we wanted have we had to create a new term. Where the textual terms match in the conceptual model and the existing schemas, so much the better, but it's not essential. 

The same thinking applies to your comment on the location vocabulary. Clearly "goegraphic identifier" and "see also" are not the same words - but in this context they do convey the same thing.

A geographic identifier is one that can be used in multiple data sets and is the key that links them. If I publish data about, say, Athens (where I happen to be as I write this), then if I use an existing geographic identifier, like http://sws.geonames.org/264371/, I'm saying "this is an existing identifier for the place I'm talking about and if you deference that URI you will get more information about it." Hence the rdfs:seeAlso property - again, a massively used and well understood property name - actually has the same meaning even if, clearly, the tedxtual term is different.

Hope this answers your questions?

Phil.

Fri, 25/05/2012 - 07:47

Just closing the issue

Fri, 25/05/2012 - 15:39

Thanks for this lengthy answer Phil

I like very much your approach about vocabularies, only creating new terms when needed.
I started using the Organization ontology before the joinup initiative and was really happy to see it re-used now, and extended with the legal and locn namespaces.

On our project, it's still too early, but we work on bridging semantic web and social and solidarity sector with different french regions. Many organizations we work with and the project itself is EU-funded so it was logic to "joinup". I'll post somehting on the group when we'll have online public data.

We started a vocabulary to host some extensions here : http://ns.economie-solidaire.fr/

I have another question, but I'll go on the forum to ask it :)

The content of this field is kept private and will not be shown publicly.