NoteBook Query

[mrchristian]

Is this workflow possible for the workshop?

1. User builds a visual query in wikidata: See paintings, Bavarian Collections, 1590 - 1750 >> https://w.wiki/6KRQ
2. Move query to Graph builder https://angryloki.github.io/wikidata-graph-builder/ (d3.js) - test out - you can paste a sparkle query into it
3. Move query to a Notebook using wikidata-graph-builder d3.js or something else? View:Graph

See: Obviously very unsophisticated, but... simple paste

Notes: Baroque https://openartbrowser.org/en/movement/Q37853?tab=timeline

[SimonXIX]

The problem I'm coming across when testing this out is that SPARQL queries for data retrieval and SPARQL queries for visual graphs need to be written differently. So while you can paste a Query Service query into the Query Builder, it doesn't result in a visually informative or interesting graph. We'd need a firm idea of what data participants are graphing to ensure that it is graphable. For example, while you could graph all paintings in the collection (P195) of Bavarian State Paintings (Q812285), that doesn't make sense as a graph: https://angryloki.github.io/wikidata-graph-builder/?item=Q812285&property=P195&mode=reverse

I'll play around with using this Graph Builder code in a Notebook though to see if it's possible with one of their pre-formed example queries.

[SimonXIX]

The most immediate issue is that the Graph Builder is written in Typescript and Quarto focuses on Jupyter Notebooks with a Python or R kernel. It's not impossible that we could get a Notebook requiring a Typescript kernel to run but it may take more time than we have considering I'm on strike all but two days this week.

[mrchristian]

Just to note it: We'll shelve this for now