FactGrid:Remove NA: A LGBTIQ*-Knowledge Graph

▴

About the project

The project "Remove NA" is about creating a knowledge graph on queer history. The project is funded by the German Federal Ministry of Education and Research and the Prototype Fund.

For a long time, queer history was rather ignored by large archives and hardly recorded in a structured way, especially not outside the narrative of perversion or criminalization. The self-organized archive Forum Queeres Archiv München e.V. has been documenting LGBTIQ* history for more than 20 years, providing queer context on people, organizations, events, places, points in time - so far mainly in analog space.

In the digital sphere, these contexts are still missing. I have therefore named my project Remove NA. The English abbreviation NA stands for "not available" and is used when information is not available.

Data on the history of non-heterosexual people is often NA and what Nigerian artist Mimi Onuoha calls "missing data": "Missing implies both a lack and a should: something is missing but should exist."

Data, information, and knowledge about queer history should exist. Only then can queer history become visible in the more complex, data-based algorithms that are increasingly part of our everyday lives. These technologies rely directly on sources such as Wikidata. If queer history is missing there, it will be missing in the applications that build on it and that we use.

Remove NA on Factgrid

A collection of SPARQL queries:

• All items created within this Factgrid research project
• Inns in Munich with an explicit LGBTIQ* target group (mostly gay men)
• available gender identities
• LGBT choires
• Persons in Remove NA dataset that migrated from Nazi Germany
• Books: Publishing years over time
• Timeline of LGBTIQ* events

More ressources

• Remove NA on Prototypefund.de
• Blog posts
• Github Repository
• Technical Documentation

Queries to get a grip on SPARQL

<syntaxhighlight>

1. Find all items containing "CSD" in the label
2. Uses FILTER and CONTAINS https://www.wikidata.org/wiki/Wikidata:SPARQL_tutorial#FILTER

SELECT DISTINCT ?item ?label WHERE {

 ?item wdt:P131 wd:Q400012;
   rdfs:label ?label.
 FILTER((LANG(?label)) = "[AUTO_LANGUAGE]") # auto language, specify with "en", "de", etc.
 FILTER(CONTAINS(?label, "CSD")) # search string, can be extended to a regex see: https://www.wikidata.org/wiki/Wikidata:SPARQL_tutorial#FILTER

} </syntaxhighlight>