The Otletosphere is a research tool developed by HyperOtlet, a research program dedicated to the study of documentation, in particular through the work of Belgian jurist and bibliographer Paul Otlet (1868-1944). Otlet’s ambition was to develop documentation as a scientific and multi-faceted practice—collaborative, transmedia and universal in its approach. The Otletosphere and HyperOtlet in general work towards that same ambition.
The Otletosphere allows us to explore the intellectual and professional network surrounding Paul Otlet. It can be used as a starting point for reflecting on Otlet’s inspirations, the genesis of his work in several areas (especially classification, transhumanism and freedom of thought), but also on his relations with his peers and opponents. As a data visualization project, it helps contextualize the information by displaying it as a network, in the hypertextual fashion of which Paul Otlet was a precursor.
In addition, the Otletosphere is a reusable program that can be adapted to other research topics. As a web interface with various functionalities, it can be an effective tool for navigating a complex set of entities (people, organizations, works and events). Descriptive elements are intentionally kept brief and point to external sources: the focus of the Otletosphere is the characterization of the relationships between entities, and their visualization in the form of an interactive graph.
In this blog post, we explain the features of the Otletosphere, on the occasion of its 2.0 release.
Guillaume Brioudes, Arthur Perret, Olivier Le Deuff, & Clément Borel. (2020). Otletosphere (Version 2.0.0). Zenodo. http://doi.org/10.5281/zenodo.3981189
Version 2.0 of the Otletosphere follows a prototype developed by Jean David in which the main components were established: an interactive graph with a sidebar containing some information about the selected entity; a list of entities in alphabetical order; a description of the project on a separate page. The database is built from a collaborative Google Sheets spreadsheet and regularly exported to update the website.
The key aspect of the new version is fluidity. Our objective was to develop an open tool with fluid use, maintenance and updating. This applies to updating the data, changing the way it is processed, designing the interface as well as translating it and navigating it.
At the center of the graph lies a central node, which represents Paul Otlet. It is connected to most of the other entities, which are essentially the people and institutions related to Otlet. Each link and node has a color, which is associated with a type of relation to Paul Otlet (family, colleague, rival…).
We improved the readability of the graph when compared to the first version, notably by allocating more space to it. We also modified the side panel: below the description of the selected entity, we added a list of the entities to which it is linked, with colored dots representing each entity’s type of relationship to Paul Otlet. With this change, the side panel now actively helps us understand the graph.
The amount of links in our dataset has increased, so we worked on their readability as well. Hovering over an entity highlights it, as well as its relationships. This effect is accentuated by selecting the entity: the node and its links thicken, which makes it easier to keep them in focus while hovering over other entities.
As in the prototype, we can filter which entities are displayed by selecting the appropriate buttons on the navigation bar. This affects three components: the graph, the list of entities but also the search engine’s suggestions. The search engine is one of the biggest additions of this 2.0 version. It allows us to quickly find entities and zoom in on the corresponding node.
We have also strengthened the interconnection of the three main elements of the interface—the graph, the list of entities and the side panel. Selecting an entity from one of these three areas updates the description displayed in the panel, and also zooms in on the corresponding node. Selecting the language from the navigation bar immediately switches the language of all these elements.
The Otletosphere is now closer to a web application, designed in synergy with the possibilities offered by browsers. Each entity has its own URL: this allows us to share a link to that entity and thus load the Otletosphere directly to its corresponding location, with the description panel displayed. This also means that each selection of an entity is saved in the browser history, which makes it possible to go back and forth by using the backward and forward buttons of the browser, without reloading the page.
Reusing the code
Updated on March 15th, 2021
The Otletosphere can be used as a visualization interface in other research contexts. The source code is open and freely reusable (under the MIT license). Version 2.0 has been deposited on Zenodo. Active development has moved to Opensphere, a reusable template for Otletosphere-like projects. An online documentation is provided.
One possible reuse, which does not require any particular experience in software development, simply consists in replacing the data files (nodes and links) with another set. The documentation gives some advice on how to generate these files.
– Website : http://hyperotlet.huma-num.fr/otletosphere/
– User Manual : https://hyperotlet.github.io/otletosphere/
– Software documentation: https://hyperotlet.github.io/opensphere/
Olivier Le Deuff, Arthur Perret, Clément Borel, Jean David, “Surfer dans l’Otletosphère. Des outils pour visualiser et interroger le réseau de Paul Otlet”, 2019, https://hal.archives-ouvertes.fr/sic_02480515