Conference Agenda

Historic music rooms, hybrid spaces of art, architecture, and performance, played an important role in shaping modern European sociability and aesthetic experience. In Portugal, these environments are now primarily accessible through visual and textual records, such as illustrated magazines, postcards, fragmentary documentary sources, and some remaining artworks that still capture the interplay between domestic interiors, musical instruments, and cultural identity [1-3].

Developed as part of an ongoing doctoral research project in Heritage Studies - History of Art, this study explores some of the potential of digital humanities methods and tools to reconstruct and interpret historic music rooms as sensory and social environments. The focus of this presentation lies on Portuguese press coverage (see figure 1), specifically Illustração Portugueza [4], of comparable spaces within a selected chronological framework (1903-1924), to elucidate the relationship between residential architectures and the cultural and artistic practices in place at the time of the foundation of the first museological nucleus of musical instruments in Portugal.

The applied strategy consists of using open-source software and services, such as the Wikimedia ecosystem of tools and data infrastructures, which are capable of supporting complex, relational, and dynamic forms of knowledge representation [5-8]. To this end, a Wikibase cloud instance is created and run by CODA as a collaborative platform for art history and music research. Then, qualitative and quantitative data and metadata from approximately one thousand issues of Illustração Portugueza are cleaned and transformed into a semantically rich and structured dataset. Next, Wikidata, the world's largest and connected multilingual semantic database, is used to model the data and metadata, as well as to reconcile controlled vocabularies as linked open data. The Wikidata Query Service software package is used as the main SPARQL endpoint to query against the Wikidata dataset and to produce both question- and data-driven knowledge graphs, chronograms, and maps, among other visualisations, for further analysis and interpretations.

Our research is grounded in a strong commitment to Open Science, as advocated by the European Commission [9], and to the FAIR data principles (Findable, Accessible, Interoperable, and Reusable) [10]. Together, these frameworks support open access to research outputs, collaborative infrastructures, and citizen engagement, while also fostering transparent and reproducible research practices.

By integrating the results of this research into the forthcoming exhibition programme of the Museu Nacional da Música (National Music Museum, Mafra, Lisbon), the project extends beyond academia to create participatory interfaces for public dialogue and co-creation. Visitors will be invited to explore reconstructions of music rooms and to contribute with interpretive insights, fostering reciprocal knowledge exchange between researchers, curators, and the public.

This presentation explores how digitisation and data-driven interpretation can activate new infrastructures of engagement, reconnecting audiences with musical, visual and architectural heritage. Furthermore, it addresses challenges of linking knowledge, open science, accessibility, and sustainability; and argues that historically grounded, digitally mediated approaches can enhance civic participation and cultural awareness, positioning digital art history research as an active contributor to living heritage.

This contribution reflects on a long first year of existence of our new Working Group, highlights some first achieved milestones, and undertakes an outlook to the nearer future. As we declared in our foundational statement, the group set out to build bridges between DH researchers, the GLAM sector, and people active on and around MediaWiki-based software solutions, be it on Wikimedia’s public platforms, or on own instances. After one year of official existence as a group, and thanks to the funding obtained in the DARIAH Theme Call, we have come together to collaborate in several in-person events as well as continuous online activities.

Our group is as diverse as the profiles of the community WG DHwiki is targeting with its activity. Some have contributed to or use Wikidata in their research projects, some use their own Wikibase instances for working with Linked Open Data (LOD) catalogues or other types of datasets, some have developing and system admin expertise, some in data visualisation, etc.

By sharing our fields of expertise to each other, we are able to share it with the larger community - which was the idea behind the two panel events we organized last year (DARIAH Annual Event and ADHO conference), and also behind our first larger milestone: Creating our own Wikibase instance, which is now available at https://dhwiki.wikibase.cloud, currently including the following contents:

• Information about the group, group members, and group meetings;

• A collection of bibliographical items covering work related to Wikidata and Wikibase in research and GLAM contexts, collected by the group members in a Zotero group library, and presented here as LOD.

• Documentation pages, intended as a collection of guidelines for everybody planning to adopt Wikidata and Wikibase in their own projects.

• A collection of jupyter notebooks showcasing Wikibase content reuse.

The documentation pages are thematically focused and coordinated by different WG members. The addressed topics are the following:

• Structure and features of Wikibase for DH

• Wikibase and Wikidata, differences and round-tripping

• Wikidata and Wikibase as a hub for identifiers

• Examples of existing Wikibase instances in the field of DH and GLAM

• Requirements from a DH perspective

• Wikibase Lexical Data

• Resource directory and Bibliography

We are currently working on a dissemination of our output in different formats, such as an overview article in a journal, presentations and papers (see our Zenodo community) and contributions to other community platforms. We aim at constantly updating and enriching the documentation pages listed above.

Based on feedback collected until now and in the future, we will try to improve the role of the group as a consultancy service in DH research and GLAM data projects. Only based on the experience made by now, we may already confirm that there is a constant demand for project-tailored one-time or even continued consultancy in topics related to FAIR data, LOD software solutions, and related tools and workflows. We will explore the integration of the Wikibase Ecosystem with public data infrastructures such as ECHOES and the common European data space for cultural heritage.

More and more Cultural Heritage data are available as Linked Open Data
(LOD) in SPARQL endpoints. However, a recurring practical problem is how to
use such data easily for Digital Humanities (DH) research without deeper knowl-
edge about SPARQL querying and programming skills. This talk presents and
demonstrates how this challenge has been tackled in practice by using the Sampo-
UI framework.

In the evolving Semantic Web ecosystem, where centralized hubs such as Wikidata coexist with third-party databases, ensuring semantic consistency and data quality across platforms is both a technical necessity and a significant governance challenge. Data round-tripping—the bidirectional synchronization of data between Wikidata and external databases, including, more recently, Wikibase instances (Lindemann et al., 2026, submitted)—has gradually emerged as a critical practice. It is driven by the Wikimedia community’s vision and activities aimed at sustainable, participatory knowledge management (Pellizzari & Marchetti, 2025).

This paper analyzes the evolution of the concept of data round-tripping within the Wikimedia ecosystem, tracing its origins from early GLAM-oriented activities focused on Wikimedia Commons (Larsson & Zeinstra, 2019) to its full adoption by Wikidata users, via the creation of a dedicated community page in July 2022.

Within Wikidata, the term has been predominantly adopted by users active in the field of authority control (Pellizzari, 2023; Pellizzari, 2024). Their practices evolved from simple data exchange into a co-creative process in which technical workflows are deeply intertwined with community collaboration and institutional dialogue, as can be traced in the history of the community pages on Wikidata.

Volunteers and institutional staff of external databases collaborate to correct errors, reconcile divergent records, and negotiate data models, embedding metadata directly into the shared knowledge infrastructure (Pellizzari & Marchetti, 2025). Over time, such practices became increasingly visible, structured, and codified within the Wikidata community. Data round-tripping thus transformed from a niche activity focused on retropatrolling and data clean-up into a set of established practices and guidelines, reaching a milestone with the creation of a dedicated community project on Wikidata in October 2025.

This paper investigates the evolving practical meaning of the concept across diverse contexts of Wikidata volunteer activity, as well as its presence within scholarly discourse. The term “data round‑tripping” (including variants such as “round‑tripped”) appears in recent scientific literature related to Wikidata also within the field of cultural heritage (Meiners & Bulle, 2025), where the concept first emerged, as well in other domains (von Mering et al., 2023; Braisher & Fitchett, 2025).

Ultimately, this research provides a snapshot of the process’s current sociotechnical dimensions, including an evolving toolkit of community-developed solutions—such as community pages, queries, and shared files—which support complex synchronization workflows. It shows that data round-tripping has come to be recognized as more than a technical aspect within the Wikidata-Wikibase ecosystem. It constitutes a foundational practice for fostering ethically engaged and resilient digital communities, enhancing accountability, and strengthening durable connections among academic research, GLAM institutions, and volunteer networks.
The paper concludes by offering governance recommendations to third‑party institutions, inviting them to understand how the Wikidata-Wikibase community is shaping the concept of data round‑tripping. Ultimately, the successful development of federated knowledge ecosystems for public metadata depends also on building a shared stakeholder understanding of this practice.