Dear All,
Here my homework for CRMarcheo.
http://cidoc-crm.org/Issue/ID-480-ap14-justified-is-justification-of
<http://cidoc-crm.org/Issue/ID-480-ap14-justified-is-justification-of>
Here the graphical representation in CRMarcheao:
I observe that is is not necessary to connect AP14 to AP13.1 and AP11.1:
The particular instance X of AP13 is determined already for the selected
instances of A4, whatever the type U assigned by AP13.1 is. The
analogous holds for the particular instance Y of AP11 and the type W
assigned by AP11.1. Hence, from X follows U, and from Y follows W. If X
is justified by Y, it follows that W was used to infer U in this case
(only!).
I maintain that the CRM must NOT produce a model which fixes the
physical relation types to specific corresponding stratigraphic relation
types, because I regard these correspondances as an open-ended research
questions. For instance, that being "on top" justifies "after", holds
only for certain undisturbed contexts and microscopic interface
features. I can imagine that progress of science may even discard
completely the "on top" as argument for certain contexts, such as
sedimentation, and refer solely to microscopic interface features in the
future.
E.g., the dinosaur footprints in Portugal in a stratigraphic interface,
currently on an inclined slope, prove the temporal sequence by being
*negative* imprints of the dinosaur feet, whereas the removed stratum
constituted *positive* representations of the feet, by filling in the
dried footprints with another layer of mud, even if the were inlined
upside down. If however the positive representations were petrified
feet, we would argue that the same shape on the one side justifies the
opposite temporal relation! This exemplifies the potential complexity of
involved background knowledge.
Hence, *I propose* that AP14 should be redefined as
*AP13.2 justified (us justification of)*,
domain: AP13
range: AP11
I would assume that the cardinality should be (0,1:0,1), not to confuse
multiple associations. Otherwise, we would need complex rules about how
to combine multiple physical relations to justify one stratigraphic one,
or one physical relation to justify more than one stratigraphic one.
We may require that *AP13.1* and *AP11.1* are *necessary*, even though I
wonder if more complex interface considerations are adequately
represented by a type. Nevertheless, it may be beneficial to require a
clear specification, so that one *can always query* what sort of
equivalence was used. I would not restrict the domain cardinality to
one, because one may combine multiple types , e.g., "On Top" AND
"Negative Imprints Below".
Note, that the use of the type W is relative to the direction of Y,
instance of AP11, and the use of the type U is relative to the direction
of X, instance of AP13, and the use of AP13.2 must follow this order.
It may in practice be necessary, to have a "twisted" version of AP13.2,
which relates AP13i to AP11....😛
..........................................................................................
We can interpret AP13.2 as a shortcut of
I2 <about:blank?compose#_S2_Sample_Taking>Belief("AP14 is TRUE"). J2i
<#_J2_concluded_that> was concluded by: I5 Inference Making. J1
<#_J1_used_as>used as premise: I2 <#_S2_Sample_Taking>Belief("AP11 is
TRUE") <about:blank?compose#_S2_Sample_Taking>
But I would *not *regard this as replacing the need for AP13.2 in CRMarcheo.
All the best,
Martin
--
------------------------------------
Dr. Martin Doerr
Honorary Head of the
Center for Cultural Informatics
Information Systems Laboratory
Institute of Computer Science
Foundation for Research and Technology - Hellas (FORTH)
N.Plastira 100, Vassilika Vouton,
GR70013 Heraklion,Crete,Greece
Vox:+30(2810)391625
Email: [email protected]
Web-site: http://www.ics.forth.gr/isl
_______________________________________________
Crm-sig mailing list
[email protected]
http://lists.ics.forth.gr/mailman/listinfo/crm-sig
