Re: Is it a redundancy? Indetected inconsistency?

On Mar 10, 2015, at 21:56, "Leila Bayoudhi" <bayoudhileila@yahoo.fr<mailto:bayoudhileila@yahoo.fr>> wrote:



Le Mardi 10 mars 2015 22h40, Bijan Parsia <bijan.parsia@manchester.ac.uk<mailto:bijan.parsia@manchester.ac.uk>> a écrit :


On Mar 10, 2015, at 21:31, "Leila Bayoudhi" <bayoudhileila@yahoo.fr<mailto:bayoudhileila@yahoo.fr>> wrote:

Hi,
thanks for your answer.
So, the proposed pattern cannot be taken as a  solution for preserving ontology consistency?

Since you go from inconsistent/incoherent to consistent/coherent, you aren't preserving the consistency state.

How?

You go from incoherent to coherent.  So you change the coherence status. Thus you don't preserve it.

What I wanted to say in that case is going from incoherent ontology to coherent one(since animal-plant is unsatisfiable)

Sure. You want to repair the ontology.

can we have another solution so as to adapt the new change to the semantics of ontology without any redundancy?

I think the proposal works. So does dropping the disjointness. So does dropping the problematic subsumption and other remodelings.

We don' want loose some information

You *have* to lose information.

which son important in the modeleld domain(we haven't to remove disjointness in some cases such as between man and woman).

But why?

I don't understand why this is a pattern at all. If I have an obvious modelling flaw like this I want to fix the modelling not "work around it".
 the framework of this change is a user may introduce a subClassOf( animal-plant plant) while forgetting that there is already another one between animal-plant and animal and disjointClassesAxiom(animal plant).

Right. So they should fix it.

Fixing it by saying "well here's a 'pattern'" makes no sense. You have no idea if that pattern is the right way from a domain perspective.

Perhaps what the incoherence is telling you is that the disjointness is wrong!

Note: perhaps, Jedidi wanted tto apply the change, while keeping the semantics of ontology at any cost(adapting the new change to the semantics of ontology is well than cancelling the change even it causes redudancy). have we to apply her solution if we maintain her same priciple?
thx in advance for your hints.

I guess the vacuous extension could be a hint that you wanted both of then but didn't know how to resolve it.  But I'd rather write an annotation.
How?

By putting an axiom annotation. Or a class annotation.