Introduction

Building on previous workshops organised by the DHNB DHLAM Working Group, this event aims to bring together DH researchers and the LAM practitioners. This year’s edition explores intangible and non-textual digital cultural heritage through the lens of “Collections as Data” and long-term preservation.

The session is structured around three interlinked pillars:

• The content: with a special focus on non-textual digital objects.
• The tools: practical applications, reproducible workflows, and new approaches, including AI-driven methods.
• Ethics & preservation: Implementing FAIR/CARE principles, the GLAM Labs Collections as Data checklist (glamlabs.io/checklist) and strategies for long-term preservation of digital objects and project outputs.

Discussion points

Theoretical

• What are the pros and cons of treating collections (i.e., heritage artifacts) as data? What are the consequences of it? What is the mindset behind it? (e.g., “colonization” of the humanities (Allington et al., 2016, and others))
• What is the key difference between digitized and born-digital collections as data, and how does it impact the way we handle and interact with those collections?
• Who steers the collections-as-data discourse? Whose opinions are heard?
• What could be an alternative/complementary discourse?

Practical

• Thought experiment: consider a dataset of some kind, either one that you actually work with or one you would like to work with, and consider both close reading (of one or a few items) and distant reading. What can we get/how can we use the data in each exercise?
• Example: analyzing one recorded oral history interview vs. a collection of them – what data can you gather in each case (e.g., data vs. metadata), what is missing from the data you gather?
• Challenges: What is the primary source of your frustration when preparing a non-textual collection for publication or working with one? Is it the tools (AI/software limitations), the data (poor quality/intangible nature), or the process (lack of time/staff)?
• The “productive” failure: Share a specific instance where things did not go as planned. How has a past failure directly improved your workflows and policies? What do you do differently now to prevent that same headache?
• Workflow & automation: Walk us through your typical pipeline for preparing collections for publication. To what extent is this process documented and reproducible? If AI is integrated into your workflow, where does it sit and what specific problems does it solve?
• Preservation & the feedback loop: How are the outputs of your projects (datasets, tools, or models) preserved for the long term? Are these results reproducible by others, and do the findings ever lead you to change how the LAM institutions prepare the “source” data for future reuse?

Format

A 60-minute contextual introduction with invited lightning talks, followed by a 90-minute interactive breakout session for use-case sharing and brainstorming about best practices, and a 30-minute synthesis of next steps.

Outcome

Participants will contribute to a collaborative map of emerging practices in the field. Beyond peer-to-peer learning and documenting project successes and lessons learned from failures, the workshop will establish a practical knowledge base for the community. The collective insights gathered during the event will directly define the DHLAM Working Group’s strategic roadmap and activity priorities for the upcoming year.

Literature as a basis for discussion

Allington, D., Brouillette, S., Golumbia, D. (2016). Neoliberal Tools (and Archives): A Political History of Digital Humanities. May 1. LA Review of Books. https://lareviewofbooks.org/article/neoliberal-tools-archives-political-history-digital-humanities

Boyd, d., & Crawford, K. (2012). Critical Questions for Big Data: Provocations for a Cultural, Technological, and Scholarly Phenomenon. Information, Communication & Society, 15(5), 662-679.

Mordell, D. (2019). Critical questions for archives as (big) data. Archivaria, 87, 140-161.

Moss, M., Thomas, D., & Gollins, T. (2018). The reconfiguration of the archive as data to be mined. Archivaria, 86, 118-151.

Park (2019). Modeling Performing Arts Archives in South Korea Based on FRBRoo

 

Organizers

Camilla Holm Soelseth, Oslo Metropolitan University, Mo von Bychelberg, Uppsala University, & Olga Holownia, IIPC/CLIR

Co-chairs of the DHLAM WG. Read more about the working group here.