Add seed concepts for the Interaction and the Regulation extensions ?

[maximelefrancois86]

The HyperAgents proposal reads:

T2.1 (M0-M39) Ontological modelling of Hypermedia MAS (INRIA): this task defines a conceptual model and core ontology for representing Hypermedia MAS on the Web. The core ontology will support minimalistic descriptions of agents, interactions, regulations, any other entities in Hypermedia MAS and relations among them. The core ontology will be extended with modules for describing interactions (WP3) and regulations (WP4). The conceptual model and core ontology will be refined in each of the main project phases.

It would be nice that the interaction ontology: for example :interactsWith between (artefacts or agents) and (artefacts or agents)

Same for the core ontology contains at least one concept that the regulation ontology can extend, for example :Organization .

[FabienGandon]

Yes I agree we should create issues for each seed we want to include in the core ontologies v1 and close this issue hmas:Organization, hmas:Group and other candidates should be extracted from the initial doc and from the scenarios. hmas:interactsWith seems really strange to me as it describes an ephemeral situation or a trace and I don't remember disucssing it before. Nevertheless the meta-issue here is identify such seeds and create the corresponding issues. Then close this one.

[DrLeturc]

hmas:isInteractingWith would make more sense ?

[oboissier]

I agree with Fabien, hmas:interactsWith or hmas:isInteractingWith are ephemeral situations.

[maximelefrancois86]

I deleted :isInteractingWith from the ontology. It seems to me we should try to gain stronger consensus among parties before making significant changes to the ontology, deleting issues, or merging pull requests.

[FabienGandon]

if the idea is to trace you could have "hint:interactedWith" in the interaction extension of the ontology but we may have to consider the alignment for instance with the notion of association in an activity in Prov-O : https://www.w3.org/TR/prov-o/#Association

[DrLeturc]

I think such property would be relevant for the part concerning the regulation.

Such property is relevant to represent when an agent is using an artifact or communicating with other agents. This is what we would like to regulate (in a close future).

[maximelefrancois86]

Our conclusion (MINES Saint-Étienne):

Nothing related to organizations is necessary in the core ontology. We don't need organizations to have a MAS system. The Regulation ontology will define these concepts

Any MAS system requires interactions though. However, interactions are dynamic and the core ontology tends to model just static information. The Interaction ontology should introduce concepts for modeling dynamic interactions

[DrLeturc]

Inria Sophia: we agree !

[maximelefrancois86]

Closing this issue as it seems we agree