Observations concerning Persontypes

I cannot understand, why in UniversalVocabularyMt the
following is asserted:

(isa PersonTypeByActivity
AtemporalNecessarilyEssentialCollectionType),
(isa PersonTypeByOccupation
AtemporalNecessarilyEssentialCollectionType),
(isa PersonTpeByPositionInOrg
AtemporalNecessarilyEssentialCollectionType),
(isa PersonTpeByCulture
AtemporalNecessarilyEssentialCollectionType),

while (I agree here:)

(not (isa MedicalSpecialtyType
AtemporalNecessarilyEssentialCollectionType)),
(not (isa Nationality
AtemporalNecessarilyEssentialCollectionType)),
(not (isa PersonTypeByEthnicity
AtemporalNecessarilyEssentialCollectionType))

I find that in UniversalVocabularyMt
Nationality is only a Collection, but in
OrganizationGVocalurayMt a CollectionType,
the same for MedicalSpecialtyType, which is a
CollectionType in HumanSocialLifeMt.

Furthermore, I wonder whether the following singular rule
is exemplary:
Mt : HumanSocialLifeMt
Direction : Forward
(implies (ist BaseKB (genls ?X Artist)) (isa ?X
PersonTypeByActivity))
The binding to BaseKB cannot really prevent that
special subconcepts of Artist
will then belong to PersonTypeByActivity, - subconcepts
which should not
belong to that metaconcept, e.g. "An artist who dies in
poverty".
By the way, the rule behaved as a backward rule, not as
forward rule (by
storing and showing the assertion).

Dietrich