[rdfweb-dev] Another relationships proposal

David Menendez zednenem at psualum.com
Fri Mar 19 20:58:57 UTC 2004


Ian Davis writes:

> Hi David,
> 
> On Friday, March 19, 2004, 7:06:11 PM, David Menendez wrote:
> > For kinship relations, I suggest this heirarchy of properties:
>
<snip> 
> This is really good.

Thanks!

> I feel however, that different cultures will treat ancestry in
> different ways. Some societies are matrilineal, tracing ancestry
> solely through the female line, others are patrilineal. Some cultures
> associate males with the father's male line and females with the
> mother's female line. There are many other lineage systems.

True, but the vocabulary is merely asserting the existence of a
relationship (and again, { A ancestor B } means "B is an ancestor of A
according to the rules A uses"). I think that even the most patrilineal
society would accept that people *have* maternal grandmothers, even if
they aren't important in terms of inheritance.

That is to say, if I say { A ancestor B }, I'm claiming that there is
*some* line of descent from B to A.


> > Note that "biologicalParent" is entirely unconnected with "parent".
> > It may be useful to define convenience terms that are subproperties
> > of "biologicalParent" and "mother" or "father" for brevity, but I
> > can't think of any good names.

> I think these would be genitor (male) and genetrix (female).

I don't think those quite capture the meaning I was thinking of.
Basically, it would be convenient to say { A father B; biologicalParent
B. } in one triple. { A genitor B. } implies { A biologicalParent B. },
but not { A father B }.

> > Marriage is more complicated, because it has a time aspect. People
> > are married after a certain date or during a certain period. I
> > suggest something like this:
> 
> >     [ a Marriage
> >     ; between A, B
> >     ; during
> >       [ a Period
> >       ; began "2004-01-01"
> >       ]
> >     ].
> 
> Is this making the assumption that it is between two people?

No. For that, you could define a subclass of marriage that restricted
the cardinality of "between". (Or, you could say that Marriage is a
subclass of PersonalUnion or something.)

(It occurs to me that not everyone may be familiar with Turtle or N3, so
here's the above code given in straight triples (_:x and _:y are blank
nodes):

  _:x rdf:type Marriage.
  _:x between A.
  _:x between B.
  _:x during _:y.
  _:y rdf:type Period.
  _:y began "2004-01-01".
  
Looking at this way, I think it would be clearer to say "involving"
rather than "between".)

> There is a distinction between the relationship of Marriage (which I
> would model as a subClass of rel:Relationship) and the Marriage event
> itself (bio:Marriage is subclass of bio:Event). The event involves
> more participants than the relationship. The relationship lasts longer
> (one hopes) than the event. Once the event has happened it exists for
> all time whereas the relationship will necessarily not be in existence
> at some time in the future.

I agree. (It sounds like the event you're describing might be better
described as "Wedding".)

This is the sense I was trying to get across: Kinship relationships are
pretty much timeless. Either you are my brother or you aren't; no one
says "he was my brother between August, 1987, and October, 2001". In
contrast, being someone's spouse has a temporal component. She wasn't
his wife before the wedding, but now she is. (And after the divorce, she
won't be any more.)

Thus, saying { A wife B. } is only semi-useful, because we'd have to
constrain it to mean either "B is A's wife right now" or "B was A's wife
at some time". Worse, if we say { A wife B, C. } it isn't clear whether
A is a bigamist or merely remarried.


(Speculating further...)

If person A lived a while ago, he might take a second wife. This is
probably best modelled like so:

  # the first marriage
  [ a Marriage
  ; involving A, B
  ; during [ a Period; began "-566"; ended "-600" ]
  ].
  
  # the second marriage
  [ a Marriage
  ; involving A, C
  ; during [ a Period; began "-580"; ended "-600" ]
  ].

Note that B and C are only related by their overlapping marriages to A.
I suspect this properly captures the sense of their relationship, but I
could be wrong.

Polyamory, unlike polygamy, bills itself as a more egalitarian
relationship, so we might describe such a group like so:

  [ a Marriage
  ; involving D, E, F
  ; during [...]
  ].

Alternately, we can use subproperties of "involving", if it's deemed
important to capture the roles played in a marriage:

  [ a Marriage
  ; husband G
  ; wife H
  ; during [...]
  ].
-- 
David Menendez <zednenem at psualum.com> <http://www.eyrie.org/~zednenem/>



More information about the foaf-dev mailing list