Computational access to library’s digital collections

Born-digital or digitized resources enable researchers to apply computational methods to various research topics in the field of digital humanities (Klingenstein, Hitchcock, & DeDeo, 2014; Nanni, Dietz, & Ponzetto, 2017). There have been tutorials and workshops on how to apply them with multiple tools like R or Python targeted for DH researchers (Unsworth, 2009; Mäkelä, 2019; Mullen, 2018; The Programming Historian, n.d.; The Digital Humanities Summer Institute: Technologies East 2020, n.d.). In addition, there is an initiative, Always Already Computational: Collections as Data to develop a strategic direction and guide libraries and cultural heritage institutions to provide collections as data for researchers so that they can leverage computational methods (Padilla, Allen, Frost, Potvin, Russey Roke, & Varner, 2019).In this lightning talk, I will share my journey to making the NAHO (National Aboriginal Health Organization) WARC (web archive format) file computationally accessible. The University of Ottawa Library initiated a web archiving project to preserve entire NAHO web content which is currently accessible only via Wayback Machine. There is no easy way to extract data from the NAHO collection which prevents researchers to apply computational methods in research tools like R, Python, or Archives Unleashed toolkit. I will also talk about processes, challenges, and resources in order to provide computational access to library collections with an example of the NAHO WARC file.This lightning talk is targeted for DH or librarians, but open for all and they can learn the importance and processes of how to make digital collections computationally accessible.