Locate the Past: Geospacial metadata in Archives

This project is part of the GLAMHack26, the Swiss Open Cultural Data Hackathon.

Baseline

TL;DR Project scope is to enrich a dataset – containing 23 338 places – with geospatial metadata to gain new insights and visualization possibilities.

The Burgerbibliothek Bern uses descriptors (index terms) for the description of their descriptive units. The advantage of assigning descriptors is that archive units relating to, for example, the same place can be found across different archive holdings. One of the commonly applied descriptor types is “places” of which there are currently 23 338 recorded.

The dataset provided contains these place descriptors and their related person descriptors, and descriptive units. For a more detailed explanation of the dataset, see the folder data. As of right now the place descriptors lack geospatial metadata.

The application of geospatial metadata in archives facilitates the possibility of new visualizations and the gain of new or the confirmation of insights. The description of places in archives is more complex than simple point geocoding, because the relevant geography is often temporal, and contextual (e.g. addresses and structures cease to exist, and/or are renamed, borders change; a rare but interesting case: In Aarau (AG) an entire building was moved 54 meters).

The primary goal is to acquire geospatial data for the place descriptors, which the Burgerbibliothek Bern can incorporate in their archival information system. The assignment of geospatial metadata ensures that our data is unambiguous and enables it to be linked to external resources automatically or semi-automatically. Furthermore, the data forms the basis for further presentation options, such as display on maps. All of this helps to provide better search options for our users.

Pains

• The place descriptors lack geospatial data.
• The place descriptors are not differentiated, sometimes a place descriptor describes a point (e. g. Bernastrasse 15) sometimes a zone (e. g. Bern).
• A challenge remains how to describe places in a historical manner. In other words, how does one depict the various stages in the expansion of a changing territory?
• Most place descriptors have no persistent ID, which makes it not ideal for interoperability (Which criteria has to be fulfilled in order that the promise behind a persistent id can be held?).

Goals

• Add geospatial data to the existing descriptors (e.g. by from wikidata.org or openstreetmap.org).
• Assure the quality of acquired data.
• Documentation of process and challenges.

Further possible goals and/or outcomes

• Create a data model in which geospatial data can be recorded in a historized manner.
• Use enhanced data for visualizations (e.g. on a map).
• Use the place descriptors in combination with the person descriptors to visualize or gain insights (e.g. who died the farthest from their birthplace?).
• Your idea!

We need your help!

• The challenge owners have a background in History and Web Development and know the dataset very well.
• Sought after skills: GIS-Specialists, Metadata-Geeks, Wiki(data)pedians, OpenStreetMappers, UX Experts, Knowledge Managers, …

Organization of repository

This repository is organized as a monorepo:

• The data folder contains the dataset and some scripts for data wrangling.
• The docs folder contains the documentation of the project (methodology, insights, etc.)

Challenge owners