inheres in at some time URI re-used with different meaning

[GoogleCodeExporter]

We had been using BFO_0000052 for inheres_in (atemporal) and BFO_0000053 for 
bearer_of. These are declared inverses in RO.

BFO2 has re-used these URIs for relations with different semantics - 
inheres-in-at-ALL-times and bearer-of-at-SOME-time.

Complicating the picture, these are *not* declared as inverses, but they 
probably *should* be:
http://code.google.com/p/bfo/issues/detail?id=144

If this axiom is added to BFO2, then at least the atemporal and temporalized 
forms sharing the same URI will be the same in terms of their basic OWL axioms.

However, they are *not* the same at the FOL level, and we must be *very* 
careful about overloading the IRIs. The differences will probably leak into the 
OWL.

E.g.

'human with parkinsons' EquivalentTo human and BFO_0000052 some 'PD'

This is obviously *not* correct with the BFO2 at-some-times reading (in fact 
it's not clear that this is allowed in BFO2 since the LHS is non-rigid). 
However, we make statements like this all the time with the atemporal relation, 
intending a RO2005-style reading (see critique for details).

I think it's very dangerous to re-use the same URIs. It would be safer if BFO2 
were to use different URIs than the ones already in use.

Or perhaps this notion of URI sharing will not work. Maybe RO should just mint 
RO IDs for these relations if we can't have a guarantee that the same URI won't 
be reused with a different meaning.

See also:
http://code.google.com/p/bfo/issues/detail?id=163

Original issue reported on code.google.com by [email protected] on 24 Apr 2013 at 7:15

• Blocked on: Is bearer-of-at-some-time inverse relation at the instance level of inheres-in-at-all-times? #144