participates and has_participant relation with different domains/ranges

[GoogleCodeExporter]

Hi,

I am wondering why participates has as domain independent continuant whereas 
has participant has as range continuant. I would have expected that the 
relations are completely inverse.

Thanks,
Mathias

Original issue reported on code.google.com by MBrochhausen@gmail.com on 4 Apr 2012 at 8:33

[GoogleCodeExporter]

This is really a BFO issue, but it looks like a RO one since RO imports BFO.

I agree they should be reciprocal. A reasoner will infer the inverse D/R 
constraints rendering the weaker axiom redundant, but it should still be 
removed as it's confusing.

What is the appropriate level? Material entity? IC? C? Does a GDC like a .pdf 
participate in its being copied?

I favor IC or even C, but this is for the BFO folks to decide.

Original comment by cmung...@gmail.com on 4 Apr 2012 at 10:02

[GoogleCodeExporter]

Thanks for answering, Chris.
So I checked back with the (sadly still not official and publicly available) 
BFO 2 Reference document and it seems this affirms that IC, GDC and SDC can be 
participants in a process, spatial regions are explicitly excluded.

Furthermore, the reference says that whenever a dependent entity participates 
its bearer participates, too.
I am not 100% sure with respect to the practical implications of this.
Given that g is a GDC and p is a process would from the remark above follow 
that it is invalid to say that p has_participant g without affirming that p 
has_participant bearer of concretization of g.

But, basically, my question is answered. It would be great if RO could be 
amended to already fit the reference.

Original comment by MBrochhausen@gmail.com on 5 Apr 2012 at 1:50

[GoogleCodeExporter]

In the current version of RO (core) these inverse relations have the proper 
inverted domain/range:

- RO_0000056 "participates in": domain "continuant", range "occurrent"
- RO_0000057 "has participant": domain "occurrent", range "continuant"

Since this addresses the question of the expected inverse domain/range, I'm 
closing this as fixed.

This doesn't address larger BFO questions, but these are RO relations. If 
there's still a question about what the domain/range should be, please open a 
new tracker item.

Original comment by ja...@overton.ca on 21 Nov 2014 at 3:37

• Changed state: Fixed