Towards a V3 ? What is the next milestone ??!

[DrLeturc]

Hi,

Now the milestone has been closed, what is our next milestone ?

What is the deadline ? The end of the next plenary meeting in Paris ?

Here is a proposition for the next milestone :

• finalize scripts i.e. :

• profile checking, consistency of the ontology, and warnings ;
• automatic deploying of the ontology ;
• test scripts ;
• script for the automatic generation of the documentation (?)

That could be great that we rework on meta.ttl I think that could be usefull for the automation process ?
(@FabienGandon @NicoRobertIn ??)

• ending the discussions on :

• interaction (what terms would we like to integrate now for the next milestone ?);
• organization (can it be finish for the next milestone?) ;
• integrating terms into organization for regulation .

Is there any possible new terms that could be integrated in core ?

I can change the priorities on issues and put high priority on the following issues :

• Property :hasBody ? #42
• Update test script #96
• Label of hmas:contains is currently "directly contains" #139
• Does a resource profile always describe a single entity? #157
• What is the link between ldp:container and hmas:contains ? #159
• Discussing about the disjointness between :Workspace :Artifact :HMASPlatform :ResourceProfile ? #160 (important one!)
• Resolve ambiguity of prov:used #161

What about ?

• A need to define a domain as a union between BehaviorSpecification and BehaviorExecution #143
• How do we make the link between a BehaviorExecution and the possible effects of such behavior on a workspace?
• SKOS Concept and SKOS metadata are not documented by PyLODE #151

The following issues could be closed :

• Should we use XDTesting for the automation process of the project ? #67

[FabienGandon]

Thanks for starting this thread.

From the top of my head I would like to add to this list:

• adding the ontologies to LOV https://lov.linkeddata.es/
• creating a dataset instantiating the newly released ontologies to describe scenario 1 for instance i.e. having a first dataset populating the ontologies
• finalizing and documenting the use of ACIMOV in Hyperagents and on GiHub in general and considering the creation of a repository template from our structure and scripts to support the work on (new) other ontologies

[NicoRobertIn]

I would like to contribute to the scenario 1 ABox since it could be useful for me in two ways:

• Know more precisely the 1000 ways to misuse and misdevelop an ontology for more accurate TBox and Abox tests
• Increase my knowledge of the ontology (how to properly use it in practice and have the full consciousness of what is missing or not)

About the TBox tests I invite any of you to discuss on #171 about new features

[DrLeturc]

It's precisely what's needed for the project. Testing the TBox on a concrete use case scenario.

I will create a new milestone next Tuesday so that we can discuss it during the next heartbeat meeting.

[andreiciortea]

@DrLeturc @FabienGandon sending a summary of our group's input for v3:

What is the deadline ? The end of the next plenary meeting in Paris ?

We suggest aiming for the end of the year: the plenary meeting in Paris might be too soon, but until then we should have made considerable progress — and beginning of December the release should already be in sight.

What we propose to consider for v3:

• formal alignment with other relevant ontologies (especially the WoT TD vocabulary); this is currently missing
• for both the core and interaction ontologies, we aim to cover several scenarios that will likely surface missing terms:
  • core ontology:
    • a scenario focused on integration with online social platforms (going towards socio-technical networks and hybrid communities); this is simply a must, the work on the Reddit use case started by @Antoine-Zimmermann and @Nesrine-Hafiene could be a good starting point
    • a scenario focused on integration with the Solid ecosystem
  • interaction ontology:
    • a scenario for representing and using enacted interactions; this relates to the transparency and accountability requirements
    • a scenario for signifiers of communicative actions following an interaction protocol known to agents (e.g., FIPA protocols)
    • a scenario for signifiers of communicative actions following an interaction protocol specification discovered at run time (e.g., a BSPL protocol)

(concise summary, but we can provide more details in the next heartbeat meeting)

[FabienGandon]

To prepare the plenary meeting session about the ontologies I extracted from that thread and the other issues the following list:

Open questions to be discussed:
#161 the relation between the action and the form should be more specific than what is provided in the PROV model
#157 Does a resource profile always describe a single entity?

Simple issues and issues to close:
#139 Label of hmas:contains is currently "directly contains"

Alignment issues:
#159 alignement with LDP
#187 alignement with WoT TD vocabulary?

Low priority issues :
#42 Property :hasBody ?
#160 disjointness between :Workspace :Artifact :HMASPlatform :ResourceProfile

Other actions:

• for both the core and interaction ontologies surface missing terms from new scenarios (in particular social platforms)
• FIPA OWL ontology update
• adding the ontologies to LOV https://lov.linkeddata.es/
• considering the creation of a repository template from our structure and scripts to support the work on (new) other ontologies
• Recommendations section within the scenario on the Discovery of Behavior Specification could be extended to explicitly state that the use of SHACL shapes in the scenario represents a specific implementation choice.

[danaivach]

The update of the Recommendations section withing the scenario on the Discovery of Behavior Specification regarding SHACL has been addressed in the merged PR #178 (commit: 83a4e1e).