Make rico:wasUsedFromDate and rico:wasUsedToDate sub-properties of rico:hasBeginningDate and rico:hasEndDate respectively

[williamsonrichard]

Unless I'm missing something, I see no reason not to do this, and no contradiction that would result :-).

[florenceclavaud]

Well finally IMHO these two properties should not be made subproperties of hasBeginningDate or hasEndDate. They are about the dates when an Appellation was used, which is different from the dates of existence (of any thing). An Appellation (e.g. a Name) may be used during several periods and therefore have other dates of existence than its dates of use. So I would vote for closing this issue without doing anything. Any thoughts @ivozandhuis and @jbkrause?

[williamsonrichard]

Apologies for the rather short issue description, which I should have elaborated upon! I think my reasoning was that a period of use of an appellation, or to put it another way an attachment of an appellation to some entity, too is a thing that begins and ends. I.e. that wasUsedFromDate can be understood as 'date upon which its use began'. I completely agree that this should not exclude repeated use, i.e. that wasUsedFromDate should be able to be used more than once, but this is as I understand it is also the case for hasBeginningDate: there is nothing currently in the ontology which ensures that hasBeginningDate can only be used once. In other words, in open-world semantics, I think that it is reasonable to interpret N hasBeginningDate D as: 'D is a date upon which N came into being in some context', which as I see it encompasses N starting being used as an appellation in some context.

I do fully agree that there is a difference between hasBeginningDate and wasUsedFromDate and that there is a need to distinguish between them, but making wasUsedFromDate a sub-property of hasBeginningDate does not change this. E.g. if one wishes to refer to the actual date of the creation of the name per se independently of any usage of it, then hasBeginningDate and not wasUsedFromDate would certainly be correct, but I think this is likely to be a niche case in comparison with wasUsedFromDate (i.e. I think it is somewhat rare that the date of an actual creation of a name qua entity-in-itself, independently of any context, is likely to be an independent event that needs to be described, though of course it might sometimes be the case).

A benefit of making wasUsedFromDate a sub-property of hasBeginningDate would be that it allows one to apply general reasoning around starts and ends, i.e. maybe one builds some general logic which applies to all entities with a start and end, which one might well define as the existence of hasBeginningDate and hasEndDate respectively. Or, on the other side of the coin, it avoids people using hasBeginningDate when wasUsedFromDate would really be more appropriate because they wish to make use of the fact that hasBeginningDate indicates 'start' in the logic they have built up.

[florenceclavaud]

I think that it is reasonable to interpret N hasBeginningDate D as: 'D is a date upon which N came into being in some context', which as I see it encompasses N starting being used as an appellation in some context.

Hi @williamsonrichard, that's actually on which I tended not to agree.... but I think I'm pretty convinced now. I probably was “splitting hairs in four” as we say in French! We will discuss it within the RiC-O team. At least it will only be a very small modification to make in the ocean of what remains to be done before releasing RiC-0 1.0 :-)

[williamsonrichard]

Hehe, great! Yes, this is definitely only a relatively minor point, and as ever I of course completely understand if it is decided not to make the change :-).

[florenceclavaud]

OK I am closing this issue as solved by #77