Visual Correspondence

Reviewed by: Leila Markosian

Review start: March 2, 2024

Site Link: https://www.correspondence.ie/index.php

Archive Link: https://archive.ph/Qusbg

Data Sources

The data for this project is derived from 165,327 letters from 56 sources. The sources for the letters are primarily university libraries and national archives, but some of the letters come from other digital humanities projects or were published in unique volumes.

Processes

In order to create data visualizations about correspondence, the project creator isolated metadata about each letter’s sender, recipient, place of origin, destination, and date.

Presentation

The “Visual Correspondence: Analysing Letters Through Data Visualization” site is designed with a “Home” page, and then a sequence of tabs that walks the site visitor through the project’s facets: “Data Visualisations”, “Letters”, “Where & When”, “People”, “Crossovers”, “About”, “Contact”, and “Niall O’Leary Services”. The homepage offers a rotating banner of images, depicting different letter collections (ie “Bess of Hardwick”) and offering context about who wrote the letters and how their data might be analyzed. The “Data Visualizations” tab allows users to select letter collections to analyze, and then submit analysis parameters based on their specific research needs. For example, user can select specific people’s letter writing activity to trace; or base their analysis on a time period or sending origin within the letter collection’s dataset. Users can also choose which kind of data visualization suits their needs best: options include map, scatterplot, pie chart, zoomable sunburst, and more. The “Letters”, “Where & When”, and “People” tabs isolate the corresponding metadata from the letters in the collection and present it as a list of letter collections, a dated map, and a list of all people noted in the collections (with the number of letters they’ve sent and received). The “Crossovers” tab displays names that have appeared in more than one collection. The entire website is searchable. The content of each letter can be accessed through hyperlinks to the letter’s original source: the “Visual Correspondence” site only serves to analyze data, and does not contain any information about a letter’s content.

Digital Tools Used

To create data visualizations about correspondence, the project creator used open-source programs including Leaflet, Exhibit, JQuery, D3, Dimple, and Sigmajs.

Languages

The site is available in English.

Review

The goal of “Visual Correspondence” is to put historical letters into context. By doing so, the project aims to “to make sense of a person's life through their correspondence. Who they wrote to, who wrote to them, when and where - these flashes of detail unveil a rich narrative about people and our past through images”. There is no funder or host institution indicated on the project site. Despite the project’s ambitious scope — offering analysis for 165,327 letters — the letters themselves are distinctly focused on Irish politics, the 19th-20th centuries, modernist literature, the socialist movement, and the Modern Era in Europe and the United States. On the “Where & When” tab, the interactive map is automatically centered on Dublin. These features indicate that the project is likely created or funded by an Irish-focused institution or individual. Each component of the project — from its detailed source guides, which discuss and link to where each collection of letters came from; to the customizable data visualizations, which account for metadata on time, location, and authorship — suits the larger goal of “Visual Correspondence” to tell a story about letters based on “the set of circumstances surrounding a letter's creation, transmission and reception”. The information available about the letter dataset is comprehensive, and the varied and customizable data visualizations suit a wide range of researchers’ needs. The project is perfect for a scholar with clear research goals in mind, as well as a distinct focus on material that is reflected in the project’s database. For the general public, the site is not particularly useful: the collection of letters is too limited, and the amount of data and visualization tools available is too much. However, the seemingly narrow scope of the project also helps to keep it focused on its goal of presenting data visualizations. At no point on the site can the user view or read the text of any letters: instead, if a researcher wishes to include letter content in their analysis, they are directed to external links that bring the user to a letter’s source website. Not only does this feature emphasize the intended use of the “Visual Correspondence” project — to present data visualizations about letters — but it also places the project in dialogue with other digital humanities and archive projects. The latter means that the “Visual Correspondence” project fits its niche without being redundant in the context of DH scholarship. Overall, “Visual Correspondence” is an effective project for scholars working with letters in the context of European modernist history.

How does this project address information?
In order to gather data for the “Visual Correspondence” project, the project’s creator scraped information about letters’ senders, origins and destinations, and time periods from multiple databases. In doing so, the project does not create new information; rather, it collates similar kinds of information from separate sources to create an interactive and searchable data visualizer. However, it is not always apparent what good can be made of the project’s information. If nothing else, “Visual Correspondence” serves as a model for the kinds of digital data visualizations that may be useful to humanities scholars.

How well does this project handle information?
“Visual Correspondence” acts as a helpful model for how information may be presented to humanities scholars, but the site itself does not encourage rigorous engagement with the information presented. Although the project demonstrates how data about primary source materials might be interpreted through different kinds of maps and charts, the information that the site relies on is largely unavailable to the site users. With this in mind, the project is guilty of creating more information — in the form of interactive and customizable data visualizations — without arguing for the meaningfulness of this information in the pursuit of knowledge.