Modelling with the wizard

The modelling wizard is the fastest way to learn RiC modelling: it turns “how do I model this?” into a guided, branching walkthrough that creates the entities as you go.

How it works

For each step the wizard gives you:

1. A prompt - a real decision (“How do you represent the tape as a whole?”).
2. Choices - the candidate entities, each with why it fits / why it doesn’t.
3. A capture - the exact API call it produces, every field editable, run live.
4. Branching - your answer can change the path (a single letter → Record; a whole exchange → Record Set).

It ends in an assembled, citable model.

Worked example: a magnetic tape

The flagship scenario models an ethnographic tape with mixed-provenance tracks:

1. The tape’s content → Record (POST /records)
2. Each track → Record Part belonging to the tape (POST /record-parts, parent_id = the tape)
3. Each track’s own creator → Agent + has_creator relation
4. The physical reel → Instantiation

Because per-track creators are relations on each Record Part, the tape carries genuinely mixed provenance.

Describe your own material (AI-assisted)

The wizard has a “describe your material” box. Type something like “a parish baptism register, 1881-1899, with hundreds of entries” and it proposes a RiC model - every entity code and relation validated against RiC-CM 1.0 before it’s shown, so it never suggests an invalid model. It proposes; you review and create. (Expect ~30s while the model thinks.)

Sample capture (what a step sends)

POST /api/ric/v1/record-parts
Content-Type: application/json

{ "title": "Track A - wedding songs, Epirus", "parent_id": 914120 }

The library

There are 18 scenarios - magnetic tape, photographs, maps, oral history, registers, newspapers, family papers, email, web archives, and more. Pick one from the scenario dropdown.

Next

• The RiC entities & relations
• Creating data - the API behind the wizard’s captures