Linked Data Design Patterns

CIDOC Linked Data Patterns: initial statements of problem

Introduction

This document attempts to formalize the areas noted by the Documentation Standards WG as a set of problems. Each problem should, ideally, be solved by a specific Linked Data Design Pattern.

The areas and individual problems listed below are a first attempt, based broadly on the discussions at the CIDOC Documentation Standards Working Group meeting in Sibiu, September 2011. The first task is to review and develop them; for example at present there are no issues recorded about recording personal information. Once we have agreement that certain individual problems are worth addressing, the next stage is to work up an entry for that problem which addresses it in the format proposed by Dodds and Davis. This will involve finding examples in existing data, discussing them, and finally agreeing on the best way to address the problem.

Museum objects and their identity

1. Who should generate identities for museum objects?

2. What relationship should hold between the number physically marked on a museum object and its Linked Data identity?

3. Ditto, for the object number(s) recorded in the museum's internal documentation system

4. To what degree should museum object Linked Data identifiers contain “meaning”?

5. How should the institution responsible for an identifier be identified?

6. How should LD identifiers be assigned for multi-part objects?

7. How should LD identifiers be assigned for non-separable parts and constituents of a museum object, for example those parts which are made of a specific material?

8. How should part-whole relationships between objects and object groups be expressed in a LD context?

Dates

1. How should different calendar systems be represented in a LD context?

2. How should date ranges be expressed?

3. How should uncertainty about a single date be expressed?

4. How should named periods be recorded?

5. How should the varying geographical extent over time of a named period be expressed?

Places

1. How should the varying extent of a named place over time be expressed?

Conservation

1. Can textual treatment descriptions be recorded as Linked Data?

2. If so, how?

3. How should treatments be recorded so that they are searchable Linked Data?

4. How should you indicate that a treatment was applied to specific aspects of, or materials within, an object?

Text

1. How should different textual descriptions of an object, each designed for a specified audience, be expressed?

2. How should the perspective (cultural, personal) of a textual statement be expressed?

3. How do you indicate the reason, type, author and date of a text block?

4. Where a controlled term is used, how do you indicate that it is a controlled term?

Other

1. How should the authority for a statement be expressed?

2. How should uncertainty about a statement be expressed?

3. How do you make a commentary on, or justification for, a statement?

4. How should the geo-temporal range of an artistic style or movement be expressed?

5. Should properties such as materials be modelled as “short-cuts”, or as the result of a materials-identifying activity?

Richard Light

11 October 2011