On 6/23/15 5:24 PM, Young,Jeff (OR)
[log in to unmask]"
The problem might be clearer by asking what the subject and
object of bf:hasHoldingOrganization statement are in this
The subject is apparently a blank node bf:Organization and
the object is something with "organizations/dlc" as an
identifier. That's odd. Presumably the item (which IS
identified) should be the subject instead of the blank node.
I agree that the concept of "Authority" shouldn't be dragged
into this particular example. Statements involving the
bf:hasHoldingOrganization property should be understandable
trying this out as triples, because that's where the proof of the
pudding always is:
X bf:heldBy _:b1 . //the subject X is held some by thing
identified by _:b1 .
_:b1 a bf:Organization . //_:b1 is of type "Organization" .
_:b1 rdfs:label "DLC" . //the display label for _:b1 is "DLC"
_:b1 bf:hasHoldingOrganization <http://id.loc.gov/vocabulary/organizations/dlc>
. //_:b1, which is of type Organization, hasHoldingOrganization
Putting this again in words:
Some thing (X) is held by some other thing (_:b1). The thing _:b1 is
an organization as defined by BIBFRAME. The display label for the
organization is "DLC". The organization "_:b1" has a holding
organization with an LOC IRI.
This is what Jeff said, I just find it easier to see it when
triple-ized. If the domain of bf:hasHoldingOrganization is bf:Item,
then this has declared the bf:Organization "_:b1" to also be an
item. In any case, it makes no sense that the organization has a
holding organization. It looks like the correct set of triples would
be more like:
X bf:heldBy _:b1 .
_:b1 a bf:Organization .
_:b1 rdfs:label "DLC" .
X bf:hasHoldingOrganization <http://id.loc.gov/vocabulary/organizations/dlc>
Leaving us, of course, still with the question of organizations vs.
authorities for organizations.
[log in to unmask] http://kcoyle.net