search

zhengj2007 / bfo-trunk

0 stars 0 forks source link

Suggested "atemporal" permanent generic relations (ro2005-ish) do not have natural inverses and need continuant/occurrent distinction #168

Open zhengj2007 opened 9 years ago

zhengj2007 commented 9 years ago

From alanruttenberg@gmail.com on May 07, 2013 13:19:10

The inverse of permanent generic part of is not permanent generic has part. This is easy to see in that, in general the left hand side of this relation is a single entity, whereas the right hand side in general is either a type, or at least a number of individuals.

That would follow the pattern that the RO class level relations are not inverses of each other.

Only the -at-a-time instance-of relation are (sort of) inverse.

Also note that permanent generic part hood doesn't make sense for occurrents.

Therefore the suggestion that the atemporal relations are simpler and don't require as many variants as the temporalized relations is false. There would need to be, at least, a distinction between occurrents and continuants is needed (because generic parthood only makes sense for continuants) and separate inverses are needed (since the part-of relation inverse is not has-part).

This would suggest that a strategy that defined "ro2005-ish" atemporal relations, would not seem to be a simpler solution than that which is offered with the temporal relations.

Original issue: http://code.google.com/p/bfo/issues/detail?id=169

zhengj2007 commented 9 years ago

From cmung...@gmail.com on May 07, 2013 10:25:21

I think there are a lot of assumptions hidden in here. The first assumption doesn't hold of the TQC interpretation.

zhengj2007 commented 9 years ago

From alanruttenberg@gmail.com on May 07, 2013 10:30:21

IIRC the TQC version I proposed used the temporalized relations to good effect. If the TQC interpretation means that during any TCQ there is only one rhs for each lhs, can you say why you need something different than those? I hadn't previously heard it offered that the atemporal has-part relation would operate only on TCQs in the case that there was permanent generic parthood. If so, how would you actually express permanent generic parthood?

zhengj2007 commented 9 years ago

From alanruttenberg@gmail.com on May 07, 2013 10:42:01

ps. Still hopefully waiting for the explicit formulation so I don't need to make assumptions any more...