Copy DASH content to the new ontology

[bergos]

The new ontology will be based on DASH (talking about the content, not the name). There are two options for how this can be done:

1. Copy an initial set and use PRs afterward to adapt it to our needs.
2. Merge feature by feature (classes and properties; also bundled) via PR into an empty document.

Please vote with 🎉 for 1. or 🚀 for 2.

Update:
Changed "Copy everything" to "Copy an initial set"

[HolgerKnublauch]

In case it's useful, I created a subset of dash as a PR. This would go into a folder called input where we could also collect other relevant input vocabularies. The real main vocab should of course be placed elsewhere.

#13

[bergos]

Thanks a lot @HolgerKnublauch, and don't forget to vote: 🎉

I voted for 1. because it will speed up our work. Starting from scratch is better if there is a need to explore different ideas, but multiple parties already adapted DASH, so I don't see a risk from that perspective.

[HolgerKnublauch]

I don't know how to vote now, because the submission from my branch is basically the starting point where I think the group could start with, then evolve it, even beyond recognition if needed. So I wouldn't copy everything from DASH but exactly the provided subset and go from there.

[bergos]

@HolgerKnublauch OK, I see. I didn't want to get lost in details with the first commit. It's now "initial set", with the idea that it's more than a single feature, and separate PRs are used to discuss features in detail.

[silas-joekel]

I think we could also have a mix between the two options.

When looking at the PR @HolgerKnublauch created I identified these topics:

• editors/viewers + vocabulary
• shape lists (dash:DateOrDateTime, dash:HTMLOrStringOrLangString)
• property roles (dash:LabelRole, dash:IconRole)
• constraint components (dash:IndexedConstraintComponent)
• node shapes (dash:ListNodeShape, dash:ListShape)
• reification

I'd propose we start with the editors/viewers vocabulary and adapt it to our needs (workflow 1). For the other topics we should have separate discussions (workflow 2).

[HolgerKnublauch]

Yes, the DASH widgets are the main contribution. And there should be a third widget type to support entering filters for search forms.

The shape lists cover the common scenarios that properties need something like sh:datatype xsd:string or rdf:HTML or rdf:langString, so that not every ontology has to repeat that large blank node structure.

[tfrancart]

I'd propose we start with the editors/viewers vocabulary and adapt it to our needs (workflow 1). For the other topics we should have separate discussions (workflow 2).

Agree with this