NPGObjProvenance

[VladimirAlexiev]

(@si-npg, @rhao: Split off from #10)

Look at NPGObjProvenance_2.xlsx.

• @workergnome do you plan to try your provenance parser on this? Doesn't look promising.
• if not, I propose to map it to

<object/{ObjectID}> crm:P24i_changed_ownership_through <object/{ObjectID}/provenance>.
<object/{ObjectID}/provenance> a crm:E8_Acquisition; crm:P3_has_note "{Provenance}".

or alternatively, if the events only involved new custodian (caretaker) but not new owner:

<object/{ObjectID}> crm:P30i_custody_transferred_through <object/{ObjectID}/provenance>.
<object/{ObjectID}/provenance> a crm:E10_Transfer_of_Custody; crm:P3_has_note "{Provenance}".

[workergnome]

I'm not sure that I'd consider the complete provenance as acquisition: It contains records of acquisitions, but I think that a series of acquisitions are not an acquisition in and of themselves. I'd call it an E7_Activity.

Coming out a long set of discussions with @edgartdata and others, my current suggestion is to model it as such:

############################################################################### 
# The provenance text itself is documented as a E31 Document.  
#
# This is the traditional, written text of a provenance statement, 
# typically written following a museum's standard form.  
# 
###############################################################################
freer_object:10173_provenance a crm:E31_Document ;

  # We use `P2 has type` to distinguish this text from any other
  # document about the object.
  crm:P2_has_type aat:300055863;

  # We use `P70 documents` to refer to period covered by the documented history
  # of ownership for the object.  This is what the content of provenance is.
  crm:P70_documents freer_object:10173_provenance_period ;

  # We use `P129 is about` to refer to the object that is the subject of the
  # the provenance.  This differs from `P70 documents` in that it is what the
  # provenance refers to, not what it is documenting.  
  crm:P129_is_about freer_object:10173 ;

  # We use `P3_has_note` to document the actual text of the provenance.
  crm:P3_has_note "To ?\nMrs. Serunian. [1]\n \nTo 1937\nDr. H. H. Serunian, Worcester, Massachusetts; by inheritance from his mother, Mrs. Serunian. [2]\n \nFrom 1937\nFreer Gallery of Art, purchased from Dr. H. H. Serunian, Worcester, Massachusetts. [3]\n \nNotes:\n \n[1] Excerpt from the Curatorial Remark 1 in the object record: “Dr. Serunian inherited the book from his mother who was an Assyrian and brought it with her to this country. Dr. Serunian says that the covers were removed from the book to make it more easily portable when his mother's relatives escaped from Kurdistan to Persia in 1916.” See Curatorial Remark 1 in the object record.\n \n[2] Curatorial Remark 1 in the object record.\n \n[3] See note 2." ;

  # We have a `rdfs:label` to provide information to a human user who wants to 
  # understand what this entity represents. 
  rdfs:label "The provenance of Freer|Sackler object 10173, 'The New Testament'." .

############################################################################### 
# The complete sequence of events that make up the provenance are documented
# as a `E7 Activity`.
#
# This event contains all of the events in the object's history
# that relate to the ownership and custody of the object.
#
# this period stretches from the moment the work is considered to exist
# until either the current moment or the documented distruction of the work.
# 
# At Level 1, there are no specifically modeled acquisition events within
# this period.
# 
###############################################################################
freer_object:10173_provenance_period a crm:E7_Activity ;
  crm:P2_has_type aat:300055863 ;
  crm:P16_used_specific_object freer:10173 ;
  crm:P4_has_time-span freer_object:10173_provenance_period_timespan;
  rdfs:label "The transfers of ownership and custody of Freer|Sackler object 10173, 'The New Testament'.".

[VladimirAlexiev]

a series of acquisitions are not an acquisition ... an E7_Activity

Depends on your point of view. You can consider it as a macro-event that transfers from the very first owner to the very last. If you concede that the series are components of E7_Activity, why not use for it the more specific class Acquisition? Other notes:

• Modeling {Provenance} as a Document is too roundabout IMHO. Why not put it as P3 on the event itself? IMHO Document should be used when there's an actual historic document (text) that you want to describe
• the distinction P129 is about vs P70 documents is IMHO wrong. These two have the same semantics (stronger than Refers to), only the domain of P70 is more specific.
• don't emit freer_object:10173_provenance_period_timespan if you have nothing to say about it

I take it you won't try to parse the free text into a series of events? Doesn't look promising.

[workergnome]

All excellent questions. I've got a bunch of thoughts, and I'll try and find time to write them up this evening—I have to, anyway, coming out of the Art Tracks meetings this week.