This post was written by Nathan Yarasavage, a Digital Projects Specialist in the Library’s Serial and Government Publications Division.
Chronicling America* users can now browse the collection’s thousands of digitized historical newspapers using an interactive map and timeline recently launched by the Library of Congress. The new “Exploring Chronicling America Newspapers” application dynamically maps publication locations of over 3,000 digitized newspapers currently available in the Chronicling America online collection. Users can also interact with a timeline of publication dates for digitized newspapers available in Chronicling America, currently covering years between 1777-1963. Powered by the Esri ArcGIS Instant Apps platform, the map and timeline are updated weekly to include the latest additions to the collection. Users can also download the currently updated dataset underlying the new features to create their own custom data visualizations or analyses.
Background
Chronicling America provides access to millions of historic American newspaper pages digitized through the National Digital Newspaper Program (NDNP). Program partners select and contribute digitized newspapers published in their states or territories, creating the national collection. All of the newspapers are in the public domain and have no known copyright restrictions. To facilitate a wide range of potential uses of the newspaper data, in addition to providing the ability to search and browse historic newspaper pages on the web, Chronicling America offers a well-documented application programming interface (API). For over a decade, researchers and scholars have used the API to interface with Chronicling America data leading to a variety of projects based on digitized historical newspapers. In 2019, the NDNP team at the Library of Congress released their first set of interactive data visualizations, designed to better inform researchers of the scope and coverage of the newspapers available in Chronicling America. These included several different types of data visualizations describing the newspapers’ locations, dates, subjects, languages, and quantities. In addition, since 2018, under a program spearheaded by the Geography and Map Division, collection specialists from across the Library of Congress have produced Story Maps about the hidden and not-so-hidden collections of the Library. Also created within the Esri Geographic Information Systems (GIS)-based software platform, Story Maps combines text, images, multimedia, and interactive maps featuring Library collections to create immersive online experiences, placing the collections in context around a central theme. The Library is continuing to explore new ArcGIS tools, such as Instant Apps and Dashboards, to publish and visualize its massive collections.
Using the Map and Timeline
The new interactive map and timeline uses the ArcGIS Instant Apps web mapping platform to build upon the foundation of the earlier Chronicling America data visualizations and the experience of the Library’s use of Story Maps. The “Exploring Chronicling America Newspapers” application is particularly useful to researchers who know the approximate date range(s) and geographic area(s) they want to research but don’t know the extent of coverage (if any) available for digitized newspapers available in Chronicling America. To answer the common “What do you have from…” question, users can interactively explore newspaper publisher locations with the ArcGIS dynamic mapping platform that powers the application. The map’s zoom controls will reveal additional details such as county and city names. Users can click on a dot to see the details about a newspaper and follow the links to view digitized issues in Chronicling America. In specific areas with multiple newspaper publications, dots will typically appear clustered around the city center.
Researchers can access the timeline feature by clicking the clock icon in the lower left-hand corner of the map. The dates in this time slider represent publication dates for digitized newspapers available in Chronicling America. Users adjust the slider to narrow the date range of newspaper publications included on the currently displayed dynamic map. It is important to note that while Chronicling America provides access to over 3,000 newspaper titles, this is a representative sample of American historical newspapers. In many cases, not all issues of specific titles are digitized. Meanwhile, Chronicling America is continuing to grow, and new content is added to the database every week.
Application Development Process
The new map and timeline application uses efficient automation procedures to remain up-to-date with the ongoing additions to Chronicling America on a weekly basis. The first step of the process involves a simple python script that interacts with the Chronicling America API to collect basic metadata about all available digitized newspapers. The metadata retrieved in this step includes the newspaper names, publication locations, dates, and quantities of digitized issues available in the system. A second script adds geo-coordinates corresponding to the publication locations to complete the updated dataset. The full dataset is loaded into the ArcGIS Instant Apps tools to create the interactive map and timeline.
While most of this process is automated, there are rare occurrences when staff must intervene. Because the newspapers in this collection are historical and most data sources for geo-visualization are based on present day place names, occasional location mismatches are found. Chronicling America includes a small number of newspapers published in ghost towns or towns that no longer exist on modern maps. Additionally, in some areas of the United States, such as Alaska and Louisiana, the concept of a county as an administrative or political subdivision of a state varies or such data may be missing from the newspaper metadata altogether. These types of data anomalies, although rare, must be corrected on a case by case basis.
For users who wish to directly access the metadata used in the visualization application, a CSV file is available for download.
Looking Forward with ArcGIS Instant Apps
The “Exploring Chronicling America Newspapers” map and timeline is the first application to use the ArcGIS Instant Apps platform at the Library of Congress. The app was developed by Ting Dai from the IT Design & Development group in close collaboration with the User Experience Design team, Serial and Government Publications Division, Geography and Map Division, and the Geospatial Hosting Environment group, all at the Library of Congress. The processes described above represent the initial phase of development for this application. Further exploration on refining the process and adding features is ongoing and future releases of the map are planned.
According to Tim St. Onge, Cartographer and GIS analyst in the Geography and Map Division, “Instant Apps is an exciting platform that allows the Library of Congress to publish web maps that explore our collections and work. Building off our successes in using Story Maps over the last few years, we think Instant Apps offers a great opportunity to broaden our understanding of our collections and their stories through the lens of geography. Starting with this excellent web map of newspapers in the Chronicling America collection, we hope to expand the use of Instant Apps and awareness of the benefits of web mapping across the Library.”
For more information, you can view additional Chronicling America data visualizations or contact National Digital Newspaper Program (NDNP) staff at the Library of Congress. Visit the Library of Congress Data Exploration Github repository for more scripts and Jupyter notebooks using openly available Library of Congress digital collections.
*The Chronicling America Historic American Newspapers online collection is a product of the National Digital Newspaper Program and jointly sponsored by the Library and the National Endowment for the Humanities.
Follow Chronicling America on Twitter @ChronAmLOC.
Click here to subscribe to Headlines & Heroes–it’s free!