Presenting at DCMI 2022

Last week we participated in the twentieth International Conference on Dublin Core and Metadata Applications, DCMI 2022 – Metadata Innovation: Inclusivity, Intelligence, and Interoperability, which was held virtually this year.

Both panels where we participated were organized by Shigeo Sugimoto, who was also the moderator for the panels. The first one was titled Metadata for Visual Media Arts – Manga/Comics, Game, Animation. As the abstract for the panel explains metadata for visual media comes with a unique set of considerations:

This panel is intended to discuss metadata for visual artworks such as Manga/Comics, Video Games, and Animations (Anime). These visual artworks have significantly different natures from those cultural objects collected by memory institutions, e.g., paintings, sculptures, books, and so forth. Some simple questions, for example, are ‘are video games a tangible object or an interactive visual function?’, ‘How can we link Manga, Anime, and Games that are created as a multimedia franchise? ‘is user experience an important metadata to characterize video games?’ and so forth. Therefore, we need to know what features need to be described by metadata and how we can use the descriptions in the networked information environment.

First, Jin Ha Lee used the example of the BTS media mix to highlight the many difficulties inherent in trying to properly archive and model a complex and dynamically shifting (with time travelling band members effecting permanent changes in already published materials) media landscape in her presentation Impact of Transmedia Storytelling on Information Organization. Furthermore, the preservation of the many contributions from the fan community pose a further challenge for knowledge organization in relation to the ever growing field of popular visual media.

Next up, in Unified Data Model for Media Arts Database Ikki Omukai introduced the Media Arts Database (MADB), an initiative of the Agency for Cultural Affairs to collect metadata on anime, manga, video games and media art in Japan. Although the four domains that the MADB collects data on each have their own characteristics, the database – built on RDF technology – aims to be able to combine all four areas under a unified data model. Beyond the challenges posed by the demands of applying this data model to the various works already within the scope of the MADB the problem of how to include “born digital” (e.g. streaming videos, webtoons, game apps) and “unclassifiable” works in the database was also raised in the presentation. The MADB is an open database and available for both download and use through the online search interface and a SPARQL endpoint.


[Source of the above PDF: Media Arts Database web pamphlet]

This was followed by our introduction of the JVMG project by Magnus Pfeffer in Enthusiast models of the Japanese visual media domain focusing on the specificities and shared features of the data models implemented by the enthusiast communities whose data we work with in the framework of the project.


In the last presentation of the panel, Graphic Medicine and LOD: Semantic Enrichment of Mental Health Comics Metadata Sean Petiya introduced the LOD-MentalHealth project that aims to automate the connecting of comics metadata with relevant “medical subject headings or descriptions of the narrative content” to enable the better “discoverability of relevant comics material and content for patients, caregivers, and healthcare providers” (

The next day the Core Cultural Metadata Model (CCMM) Workshop panel also featured four presentations to help start off the discussion. To once again cite the panel abstract:

The Core Cultural Metadata Model interest group (CCMM-ig) was proposed to develop a core model which help understand organization of metadata to describe various types of cultural entities and digital archiving of those entities in the networked information environment, including tangible and intangible entities, perpetual and ephemeral entities, digital and non-digital entities, etc. The CCMM group was inspired by the significant developments of digital archiving of cultural entities supported by Linked Open Data (LOD) resources, e.g., Europeana.

In the first presentation titled What should the CCMM Core cover? from viewpoints of Cultural Heritage Description and Standards Marcia Zeng offered a detailed overview of the many different nomenclatures and standards that come into play when we discuss describing cultural artifacts – from the many levels distinguished by the CDWA (such as item, group, collection, series, etc.) to the various standards for data structures that exist alongside each other – as well as pointing out the differences between the three different perspectives (production, content and audience based) on metadata creation and use.

The next presentation, Perspectives from intangible cultural Heritage by L.P. Coladangelo provided a perfect next step in setting up for the discussion to follow by introducing a set of similar questions this time in relation to intangible cultural assets (touching on CIDOC CRM, FRBRoo, the Sampo Model, CHDE, CEDA, and domain specific knowledge organization systems), as well as highlighting one of the central problems of the domain, namely the preservation of intangible cultural heritage for which mere documentation is not enough, but rather also requires the establishing of infrastructures for the sharing and transmission of these knowledge and practices.

Third up, Magnus Pfeffer introduced some of the key points from our work on data quality and the legal integration of the various databases incorporated in the JVMG knowledge graph in his presentation Enthusiast websites as a source for data on visual media.


The discussion starting series of talks was completed with Shigeo Sugimoto‘s problem statement Models for digital archiving of cultural entities highlighting some of the different challenges facing archivists when dealing with physical objects which can be archived, as opposed to events, phenomena and the like of which only recordings may be saved, but not the actual entity itself as it transpired. This final presentation was followed by a lively discussion picking up on most of the topics set out by the panelists.

We would like to kindly thank all the organizers of DCMI 2022 for the great conference, and especially Shigeo Sugimoto for organizing both of the above panels on cultural metadata. Very much looking forward to meeting with everyone and continuing our discussions again next year!