search

eProcurement-everis / ePO

2 stars 0 forks source link

OWL-TTL code. Suggestion-not to publish yet #22

Closed paulakeen closed 6 years ago

paulakeen commented 6 years ago

Dear OP,

We suggest not to publish the OWL-TTL code yet on the WG Github.

Reason:

The current version is misaligned with the Conceptual Data Model. We're still trying to agree with you about the Conceptual Data Model.

muricna commented 6 years ago

We agree before the changes to the conceptual model we spotted the following problems.
between the ontology schema (image) and the ttl concrete artifact:

1) epo:Lot a. In the picture the following holds: "epo:ProcurementProcedure" -- "epo:hasLots" -- > "epo:Lot" b. In the ttl model the following holds: "epo:Lot" -- "rdfs:subClass" -- > "epo:ProcurementProcedure" 2) "epo:hasLegalBasis" in the picture appears as "epo:hasLegislation" in the ttl 3) "epo:appliesTo" in the picture appears as "epo:appliesToLot" and "epo:appliesToGroupOfLots" in ttl (maybe this change is made on purpose but the picture has to be aligned) 4) "epo:PropertyGroup" does not appear in the ttl 5) "epo:hasDocumentReference" and "epo:DocumentReference" do not appear in the ttl (probably there is a reason since in the picture they are marked differently from the other classes and properties)

Can you please let us know when the ttl matches the conceptual model or a clear part of it without mistakes elsewhere

paulakeen commented 6 years ago

Indeed, the objective at this stage is to come up with an agreed and stable Conceptual Data Model. That has been the objective for a while for this very next WGM.

The OWL-TTL should be reviewed in two phases, at least:

  1. While preparing unitary tests for the discussions about the design proposals (see a very preliminary example for Lots), and
  2. During the development of the proof of concept;

If this revision is to be done by Enrico, before publishing the ttl on the WG Github, I'd prefer he waits for our go before starting, otherwise he'll waste his time. The ideal time for him to start having a first revision would be not before we finish the analysis of the glossary and its transposition into the Conceptual Data Model, which should take us a week, max two, yet.

About the example attached: beware that these are unitary test examples that may vary (will vary for sure) au-fure-et-à-mesure that the Conceptual Data Model Changes. But should help to understand the design principles proposed, and which are to be discussed and agreed with the technical sub-groups and ultimately with the WG members.

epoLots.zip

KR

sembu-eproc commented 6 years ago

Can this issue be closed?