When working with datasets that contain historical data, date values like "ca. 1820" or "1880?" are not uncommon. Here is an example that shows how such entries typically look like:

title: Watch and pair case
creator: Francis Stamper
mediaURL: <url to image>
date: ca. 1690

This entry is taken from the Europeana dataset and is about a pocketwatch created in "ca. 1690" by the British watchmaker Francis Stamper, where the media URL leads to an image of the watch.
Vague date values like the above "ca. 1690" are problematic because they are not precise. "ca." indicates that the watch was most likely created in 1690 but it could have been another year as well. Furthermore, the degree of deviation is unknown since different historians may have different definitions of "ca.": some may interpret it as a deviation of +/- 5 years, others as a deviation of +/- 15 years.

With my research, I want to solve this problem and enable historians to be on the same page when creating or analyzing such vague statements in the context of dates. I will extend the existing Linked Open Data ontology "LODE" which represents historical time(see: https://vocabs.acdh.oeaw.ac.at/date_entities/en/). I will create a new data model that is capable of representing different types of temporal vagueness or rather vague dates.

With this survey I want to find out how domain experts typically deal with dates and occurrences of vagueness within them in everyday work and where the main difficulties and problems lie when doing so. Based on this feedback I will prioritize the functionalities that should be implemented.

Thank you for the feedback in advance!