For the people: How we make online LAM collections more democratized

Abstract

The article discusses how digitization in libraries, museums, and archives (LAM) can become more democratic. Digitization within LAM scholarship has been seen historically as a democratic act because it provides universal access to cultural heritage content, breaks down authoritative narratives, and enables participation from users. The article critiques the misconception that online collections democratize artifact information for public consumption and explores the ways in which LAM institutions fall short of living up to their democratic ideals when it comes to digital collections projects. Inspired by others with similar critiques, the authors discuss how LAM institutions can better fulfill the ideal of accessible and equitable access to their collections. The article emphasizes the importance of five areas of digital collections projects: system design, metadata practices, digitization selection and prioritization, labor, and user participation and engagement. The widespread misconception that digitization and digital collections are democratizing is a result of institutional biases that have masked undemocratic processes and systems which the authors strive to expose.

Keywords:

• Digital collections
• metadata practices
• labor

Notes

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 While we do not adopt a single critical lens, we build on the work of numerous predecessors, including drawing inspiration from the “slow” theoretical framework (e.g. Christen & Anderson, 2019; Farkas, 2021; Petrini et al., 2007). We also lean on previous works by feminist scholars, like bell hooks (2000), Michelle Caswell (2019), Catherine D’Ignazio and Lauren Klein (2020), and Baharak Yousefi (2017), among others. We acknowledge that any feminism that is not grounded in intersectionality is exclusionary and non-liberatory. For that reason, our work also draws on the work of anti-racism practice (e.g., Berry, 2021; Coleman, 2020) and queer cataloging practices (e.g., Billey et al., 2014; Drabinski, 2013). We acknowledge that, as two cishet white women, we have privilege in spaces others do not by virtue of our whiteness, and our positionalities have directly affected this article.

2 In addition to the sources cited above, see also Parry, 2010, p. 464 and Schweibenz, 2019, pp. 16–17 for discussion of attitudes towards computing in museum scholarship.

3 One exception is Bennett (1990) who uses the word democracy when applied to online collections of museums as a critique towards a better outcome, which is more similar to its usage in archives scholarship as opposed to other usages in scholarship around libraries and museums.

4 The authors collaborated on a grant project to bring together digitized cultural heritage materials into a single platform called Marble. At that time, Bertoldi worked at the Snite Museum of Art, and Narlock worked at Hesburgh Libraries, both located at the University of Notre Dame.

5 An additional example is the Ohio State University Library, which includes their Library’s online collections on their Diversity, Equity, Inclusion and Accessibility (DEIA) webpage as evidence of their efforts without any additional context beyond basic metadata from the source system (The Ohio State University Libraries, accessed May 26, 2023). Similarly, Yale University’s recently-launched digital collections platform, LUX, which contains holdings from Yale’s museums, archives, and libraries, includes a paragraph near the bottom of the webpage, which states that the platform “exposes longstanding and deep bias within our acquisition activities, collection systems, and descriptive practices,” with the warning that some of the content and descriptions are problematic (Yale, 2023). These examples are indicative of a trend in LAM to use online collections as a proxy for diversity.

6 We are using this term inclusively to talk about LAM systems, acknowledging that some are content management systems in which a component is for cataloging (e.g., ArchivesSpace).

7 The authors use the term “cataloger” to encompass all practitioners involved in describing or creating metadata for collections.

8 For a case study on metadata extraction, harmonization, and optimization for machine-readability, see Bertoldi et al., 2023.

9 Many online collections use Dublin Core, or a variant thereof, to make their content accessible online. While there is a field for “Additional Description,” where such contextual information could be added, this is not used consistently.

10 Even terms like “hidden collections” or “undiscovered gems” absolve institutions of blame: were the collections hidden, or were they obscured in favor of other collections, perhaps due to inappropriate description? There is also a distinction between hidden and absent. As Dorothy Berry notes, when archives are missing collections, “that is not a silence, it’s not an absence: it’s a loud, resounding clarity” (Berry et al., 2022). Archives are not silent or incomplete; “[t]hey are very loud about what is important to them and they are very complete about their choices” (Berry et al., 2022). Similarly, collections aren’t hidden–they have been pushed aside, forgotten, or rewritten.

11 There are many reasons an item or collection may be prioritized for digitization, including the need to capture and preserve at-risk media on legacy formats.

12 Walsh observed that museums are so shy about getting feedback, they are unprepared when they receive it (1997, p. 73). Without a healthy feedback loop, something that other companies rely on to stay in business, museums stay out of touch with their audiences. But, engagement is more than “Contact us” forms or “Submit feedback” requests because this request is only a passive invitation (e.g., Ahmed, 2021, pp. 137, 171).

13 Increased online engagement could take different shapes depending on the cultural institution as well as the needs of the community, but there are a few examples to look to that provide some suggestions for future efforts. There are projects that center on transcription (e.g., Moyle et al., 2011; Owens, 2013; Spindler, 2014); there are sites that encourage and celebrate citizen scientists (e.g., U.S. General Services Administration, 2023; Society for Science & the Public, 2023); there are tools for allowing visitors to annotate institutional content (e.g., Hypothesis Inc., 2023); there are an increasing number of digital collection sites that allow users to create their own collections of materials (e.g., Harvard Library, 2023) with a few even allowing users to add their own descriptive text in said collection (e.g., University of Notre Dame, 2023). All of these online cultural memory sites prioritize and celebrate the knowledge of communities alongside the institutional voice.

14 There are those in the community that can and will make spaces unsafe for the most vulnerable, e.g. an in-person event close to Narlock (Mazurek, 2022). Additional ethical considerations, including content moderation and facilitation, lack of access, and unequal benefit distribution are explored in detail in Liew and Cheetham (2016). We don’t have an answer to this problem, and recognize that any solution for the needs of specific institutions interacting with specific communities will need to be localized (Berry et al., 2022). Instead, we want to make it clear that web pages with static content do not build community, but alienate communities.