![Sample our Information Science journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5940/TOC-Subject-Sample-2021-Communication-Studies.jpg"})
Abstract
As institutional archives digitize their historical map collections and make them publicly available online, researchers and members of the public can pursue new methods of engagement with the materials. Certain transformations, like “georeferencing” wherein a scanned map image is turned into a geospatial raster dataset, can be time-consuming and difficult to automate, but do lend themselves to crowdsourcing, a well-accepted means by which archival processing and public engagement can be carried out in concert. Inspired by the concept of an archival commons, this article recounts the design, creation, and public participation period of a new web-based, crowdsourced georeferencing platform—something like a georeferencing commons. This work took place as a masters thesis project through 2021 and 2022, and the result continues to grow and evolve as OldInsuranceMaps.net, a public space for georeferencing historical fire insurance maps from the Library of Congress digital Sanborn Map collection. While the initial iteration of this project was built from GeoNode and GeoServer, and focused on maps of Louisiana, this article concludes with significant updates on the platform, and how it is being used by institutions and individuals well outside of Louisiana today.
Acknowledgments
The author would like to thank his advisor Andrew Sluyter and thesis committee Ed Benoit and Craig Colten at LSU for their support as this project developed. Additionally, many thanks to the following people for collaboration or advice along the way: Julie Stoner (Library of Congress), Eliot Jordan (GeoBlacklight, IIIF Maps), Bert Spaan (Allmaps), Jeff Meyer (OpenHistoricalMap), and Elizabeth Williams (Midlo Center for New Orleans Studies, HistoryForge). More broadly, this work would not have been possible without the software developers behind GeoNode, GeoServer, Django, GDAL, Svelte, OpenLayers, and TiTiler, all open source projects with strong individuals and communities behind them. The author would also like to acknowledge the work performed by Historical Information Gatherers to digitize the Library of Congress Sanborn Map collection, and the LOC for maintaining free access to it. Finally, a huge thank you to all of the georeferencing contributors to OldInsuranceMaps.net, especially András “WallyKitty” Nagy, for his unexpected and truly indispensable dedication to the project.
Disclosure Statement
Through the development and study period described in this article, the author self-funded OldInsuranceMaps.net, and he would like to thank the handful of individuals who made contributions to the project after that period. Institutional funding has also come from the Midlo Center for New Orleans Studies at the University of New Orleans and Ohio State University Libraries. Currently, OldInsuranceMaps.net is funded in part by the National Institutes of Health (National Institute on Aging: R01AG080401) through a partnership with University of Michigan Institute for Social Research, University of Richmond Digital Scholarship Lab, and the National Community Reinvestment Coalition.
Notes
1 More recent scholarship in this vein has pointed out that the term digital archive should be considered a “misnomer”—less an archive than a new environment for recombinant, born digital materials (Hodder and Beckingham Citation2022, 1301). Though not pursued further here, this concept of recombination is especially relevant to a system designed, ultimately, with the purpose of virtually breaking down, transforming, and mosaicking bound historical map volumes.
2 Items in the LOC collection are most commonly single map editions—a multi-page atlas of a single city in a single year. However, some items are single volumes from multi-volume atlas of large cities. This makes it difficult to apply a single hierarchical term to all items in the collection. Imprecision notwithstanding, the term “volume” came to be used in this project.
3 To this author’s knowledge, no critical analysis has been made with regard to which settlements the Sanborn Map Company chose to survey. Swab (Citation2020, 33) suggests the personal papers of surveyor Daniel Beard would do much to illuminate this subject.
4 See https://www.loc.gov/apis/json-and-yaml/. The LOC also provides a IIIF endpoint for the collection content that was not heavily utilized in this project, primarily due to ignorance on the part of this author.
5 See Cox Citation2022 for a full illustration of the relational data model used to store “canonical” GCPs (62), as well examples of cut-lines, GCPs, and masks stored in JSON at the session level (120-22).
6 See also “A Note on Map Warper”, https://geoservices.leventhalmap.org/cartinal/guides/lmec-dc-geo.html. Leventhal Map & Education Center - BPL. October 21, 2020.
7 Virtuelles Kartenforum, https://kartenforum.slub-dresden.de, last accessed December 1st, 2023.
9 The platform was (and still is) deployed on a single Ubuntu 22 server hosted on Linode (now Akamai).
10 See Cox Citation2022 for a more detailed analysis of survey results (84-85), as well as a full example of the questionnaire (117-19).
11 A key technological component of this interface is the Crop filter from ol-ext, an extended set of tools built for OpenLayers by Jean-Marc Viglino. https://viglino.github.io/ol-ext/.
14 A fantastic blog post “GeoTIFF Compression for Dummies” by software developer Paul Ramsey was the inspiration behind this upgrade, see http://blog.cleverelephant.ca/2015/02/geotiff-compression-for-dummies.html. Last accessed February 17th, 2024.
18 When the switch from GeoServer to TiTiler was made, the latter did not have WMS capabilities. However, a WMS extension was fortuitously added by an open source contributor to that project a few months later, making this distribution method available to us when we needed it for HistoryForge.
Additional information
Notes on contributors
Adam Cox
Adam Cox is a geospatial software developer with an interest in crowdsourcing and historical maps. This article describes work carried out through dual MLIS and MS Geography degrees at Louisiana State University in Baton Rouge, LA. Currently, he lives in New Orleans and works as a research software engineer in the Healthy Regions & Policies Lab at the University of Illinois, Urbana-Champaign.