Investigating original order with cybernetics and community detection algorithms

Abstract

This article is an exploration into the principle of original order and begins with the presentation of a case study, which involved subjecting the records of a single department to a series of community detection algorithms. The results of the algorithms were visualized and presented to department staff members during qualitative interviews, who were asked to decide which community detection algorithm best represented their department. The case study results, along with ideas from the field of cybernetics, placed the work of Muller, Feith, and Fruin in a new light. More specifically, it became clear that the principle of original order and the principle respect des fonds crucially result in aggregates comprised of mutually relevant records. This mutual relevance is a necessary feature of an archival aggregate, as it makes it possible for researchers to corroborate pieces of data, which in turn enables a reconstruction of the past. Ultimately, this article delves into the various ways that record aggregates become mutually relevant, investigating physical original order, administration-based original order, and functional classification.