The Spark of Linked Data and Libraries: OCLC Passage Project White Paper Translation Report

• Main Page
• Event Record

2025/12/26

In a quiet corner of Daan District, just a few blocks from National Taiwan University, sits LightBox Library—a small non-profit space dedicated to recording, preserving, and cultivating Taiwan’s photographic culture. From this secluded spot in Taipei, LightBox has become one of the newer partners of Wikidata Taiwan. Like many of our partners, they show strong potential and bring a unique perspective to the movement.

As with most of our events this year, the discussion was driven by our translation project of the OCLC Passage white paper. Due to the nature of LightBox Library, its audience aligns closely with the paper’s original target audiences. Their specialization in photography also allowed us to dive deeper into the case studies presented in the paper and ground the discussion in real-world practice.

During the presentation, we focused on explaining how the technology works in practice and how precise information can be modeled using relatively simple modules, an approach that is often more intuitive than traditional library cataloging workflows. At the same time, the limitations were clear. Image collections that lack sufficient contextual information cannot be fully modeled in Wikidata. In such cases, catalogers must step into a more research-oriented role, uncovering additional details and information so the material can be integrated into the knowledge graph. While this task is essential in the long run by enriching the data, it is also way more resource-intensive than current free-text–based workflows.

We concluded the presentation by reflecting on our broader work within the GLAM communities and using our learnings to complement the findings of the OCLC Passage project. One of the reocurring challenge is the “black box” nature of the knowledge graph, where the impact of the work is often diluted by the sheer amount of data in the graph and shadows the result of our work. This uncertainty can discourage adoption, yet our experience suggests that once a critical amount of data is reached, the effects begin to surface. Even with only a few hundred entries, we have already observed our data being picked up by Google’s search engine, and seeping into the public sight.

The event closed with an open discussion on our work as a whole. Once again, AI emerged as a central theme. The concern of AI and the expectation of how it can shape the world is both frightening and fascinating. We shared our concerns, ongoing experiments of integration of technologies, and reflections on how we might coexist with these tools as they become an inevitable part of our daily lives. And most importantly, how culture workers should position ourselves in this post-truth society, where knowledge becomes easier to access yet more and more difficult to verify.