workPerformed add-on for Event structured data template

[dlh28]

While consulting the Artsdata Instructions – Schema Event properties spreadsheet, I noticed that although we offer guidance for modelling the workPerformed property, this property is not actually present in the Artsdata event structured data template.

Several participants in this year's and last year's Digital Discoverability Program expressed an interest in modelling workPerformed in their event structured data – particularly classical music presenters, whose audiences often search the web for live performances of works by specific composers. As a result, I started offering my own version of a workPerformed add-on, as requested. @fjjulien agreed with my initial proposal for adding a formalized version to the template and proposed expanding the scope of the workPerformed add-on to describe either a performance works or a dramatic work.

Here is the newly modified version of my original workPerformed add-on. I also proposed a revision to the existing workPerformed instructions in the spreadsheet above, providing additional clarification on the use of sameAs URIs and the creator subproperty.

Once we are satisfied with the add-on, I can submit a pull request to the Artsdata data model.

@saumier @GaryMan1968

[fjjulien]

@dlh28 I think you included the essential properties for describing a performance work (or musical work, dramatic work, or musico-dramatic work). I would only add a couple array brackets for the following values:

• The CreativeWork itself: in a classical music concert, it's common to have 2, 3 or 4 different compositions performed over the course of a concert program.
• Language of the work: some theatre works are mainly in English (or French) and also include significant parts in another, secondary language. I would make this an array and I would describe the second value as a "secondary language, if applicable".

The creator object could also be made into an array, but it's probably best to keep it simple.

[saumier]

@dlh28 I took a look at your newly modified version and I would like to point out a technicality that the order of a list in JSON-LD is NOT preserved. Because of that I don't see how to tell the difference between a primary work and a secondary work. If it is important to tell them apart, then some additional property needs to be added. Otherwise it looks fine.

[fjjulien]

If an organization was really keen about stating that a work is the primary work, they could nest the work within a schema:Role object, and give the role a human recognizable name.

Otherwise, for digital discoverability purpose, the role, rank or order of nested CreativeWork entities does indeed not matter much. Content-based recommender system will consider whether a given work or works or a given composer are present or not. They may not give much, if any, value to any role or rank in the data.

[dlh28]

So, should I add schema:Role as an optional property in the template? Or should we just leave it as it is @fjjulien

[fjjulien]

@dlh28 In my opinion, it would be preferable leave it out of the template. It would add a layer of complexity on an already fairly elaborate template.

[saumier]

@dlh28 Is this issue finished? If so then please close. Thx.

[fjjulien]

Let's keep this issue open. With the new ado:LivePerformanceWork class, we actually have an opportunity to publish an add-on to the structured data templates.