Issue 447: A7 Embedding as a Physical Feature like entity

ID: 
447
Starting Date: 
2019-10-23
Working Group: 
3
Status: 
Open
Background: 

In the 45th joint meeting of the CIDOC CRM SIG and SO/TC46/SC4/WG9; 38th FRBR – CIDOC CRM Harmonization meeting and in the context of discussing issue 283 (superproperties to CRMarchaeo properties), and according to the decision reached in the 43rd meeting of the CIDOC-CRM sig, A7 Embedding should not be declared a temporal entity (previously adopted semantics for A7: isA E3 Condition State) but some sort of feature-like entity (like for instance S20 Rigid Physical Feature).

This decision has implications regarding the semantics of A7 as well as the properties linking it to other CRM classes (plus their superproperties).

Affected properties are:

  • AP17 is found by (found) [D: A7 Embedding; R: S19 Encounter Event]
  • AP18 is embedding of (is embedded) [D: A7 Embedding; R: E18 Physical Thing]
  • AP19 is embedding in (contains embedding) [[D: A7 Embedding; R: A2 Stratigraphic Volume Unit]
  • AP20 is embedding at (contains) [D: A7 Embedding; R: E53 Place]
  • AP21 contains (is contained in) [D: A2 Stratigraphic Volume Unit, R: E18 Physical Thing]
    [this one is a shortcut from the fully developed path:
    A2 Stratigraphic Volume Unit -- AP19i contains embedding --> A7 Embedding -- AP18 is embedding of --> E18 Physical Thing
    hence any decision reached for A7 and affected properties AP18/19 will also affect AP21]

HW: CEO to edit A7 Embedding accordingly and also check properties AP17 through AP21 (check for consistency with newly postulated semantics for A7 and also determine their superproperties). This HW was not explicitly assigned over the 45th CRM sig meeting, but it was a pending decision from the 43rd meeting, and the HW had previously been assigned to CEO.

Heraklion, October 2019

Current Proposal: 

Posted by Martin on 26/02/2021

Dear All,

In the October SIG Meeting, the following arguments were made:

"Gerald Hiebel would prefer A7 Embedding as a S20. One point in the argumentation may be, that the surrounding matter of an Embedding has a specific condition is often used to determine parameters like the time of deposition"

I support this.

"What we loose when we define it as S20:

“It further allows for specifying temporal bounds for which a particular embedding has existed, as specified by the evidence.
Maybe we could create a property e.g. was embedded for (instead of AP20?) that relates to a time span to be able to state temporal bounds of the embedding.
If we define A7 as S20 we would not really need a property AP20 embedded at as it is already a E53 Place that we could attach spatial information to."

Counterargument: A Rigid Physical Feature has a genesis event, that allows for specifying the time of embedding. No extra property needed.

The property AP20 is necessary, because "This property identifies the E53 Place that is documented as the E53 Place of the A7 Embedding. This place must be at rest relative to the instance of A2 Stratigraphic Volume Unit that contains the A7 Embedding."

I believe the relation to the stratigraphic unit would be quite cumbersome to make otherwise.

Posted by Christian Emil on 5/03/2021

Dear all,

In the I October SIG Meeting, the following arguments were made:
"Gerald Hiebel would prefer A7 Embedding as a S20. One point in the argumentation may be, that the surrounding matter of an Embedding has a specific condition is often used to determine parameters like the time of deposition"

MD supports this, and in my opinion this is a correct decision. I was asked to have a look at this "CEO to edit A7 Embedding accordingly and also check properties AP17 through AP21 (check for consistency with newly postulated semantics for A7 and also determine their superproperties)."

If A7 Embedding becomes a subclass of S20 Rigid Physical Feature the temporal aspect disappear and  an instance of E7 will be a physical feature surrounding an instance(s) of  E18 Physical Thing and a place for this feature. Isn't this another way to say that for an instance x of E18 Physical Thing , y of E53 Place, z of S20 S20 Rigid Physical Feature

x P53 has former or current location (is former or current location of) y P121 overlaps with/P89 falls within (contains)​ z?

 

Posted by Martin on 5/03/2021

Dear Christian-Emil,

I think so! More precisely:

x P156 occupies y P89 falls within (contains)​ z
...as long as the thing was completely covered.

x P156 occupies y P121 overlaps with z
if parts were above surface, or less constraint for complete coverage.

We may include partial embedding or not.

Both y and z are P157 at rest relative to z

Correct?

Posted by Christian Emil on 5/03/2021

I agree, and thnak you for reminding me of the existence of  P156 occupies​.

Posted by Achille on 9/03/2021

Dear Christian-Emil, Martin,

I cannot help but continue to think that an embedding is something more immaterial than an S20, for which E3 seems more adequate to me, especially in cases where it is necessary to indicate the position, orientation and other modalities of placement of an object within an archaeological layer.

Gerald's consideration is of great importance, and it is true that the surrounding matter of an Embedding can be used to make inferences about the modalities or times of the deposition. However, the “surroundings" of an embedding are not the embedding itself.

I imagine that archaeologists would prefer to make such inferences by means of stratigraphy, i.e. identifying the relationships between the surrounding stratigraphic volumes and interfaces. Embedding would be useful more for considerations regarding the state of the deposit over time.