[Buildingdata] Representing a person's role
Christopher Gutteridge
cjg at ecs.soton.ac.uk
Sun Mar 4 13:01:19 GMT 2012
I've been thinking lots about how often I need to talk about people
based on their role, not as a person.
Some systems I've seen model job roles as a group where zero or more
people belong, but I'm more interested in roles which are an explicit
relationship between a person and an organisation.
The driver for this is normally in a programme of events we list a
person as the speaker in a session, then associate them with their
affiliations.
http://data.dev8d.org/2012/programme/?person=AlexanderDutton&output=rdf
This is a bit ugly, and I want to re-think it. I currently feel that
what would make more sense is that if you want someone's affiliation to
be considered relevant to a talk, then they, the foaf:Person, are not
the speaker, but rather them as a x:PersonRole which is a named
relationship between a person and another thing. Generally an
organization, but it could also be an event (eg a conference...)
It would be generalisation of work Nick Gibbins, Marcus Cobden, Alastair
Cummings, and Colin Williams did for ECS many years back;
http://id.ecs.soton.ac.uk/docs/
PersonRole would be a subclass of foaf:Agent so it can be a speaker at
event, member of a group etc. Many but not all properties of a
PersonRole can be considered also properties of the Person who holds
that role, (eg phone number, group membership, hat size) but a few
properties may explicitly describe a PersonRole, not a person.
A PersonRole being a role of an organisation does not imply the person
is a member of that organisation-- some roles could be 'advisor' or
'guest'.
It's very possible there's an existing scheme that does this, but I've
not run across it.
Any idea where it should live if we coin it ourselves? openorg
namespace, I guess.
-- Christopher Gutteridge -- http://id.ecs.soton.ac.uk/person/1248 You
should read the ECS Web Team blog: http://blogs.ecs.soton.ac.uk/webteam/
More information about the Buildingdata
mailing list