ODIS: An Introduction

At the dawn of the new millennium, four major private cultural archives in the Flemish region of Belgium were looking for a system to store, interconnect, and disclose contextual data sets on their heritage collections, which consist primarily of archival, library, and audiovisual materials. These organisations – ADVN, archive of national movements in Antwerp; Amsab-Institute of Social History (Amsab-ISG) in Ghent; KADOC-KU Leuven, Documentation and Research Centre on Religion, Culture, and Society in Leuven; and Liberas, Heritage Centre for the History of the Freedom Ideal in Ghent (the former Liberal Archive) – joined forces. In 2000–2003, funded by a grant from the Max Wildiers Fund of the Research Foundation Flanders (FWO), the organisations were able to develop ODIS, a contextual online database where heritage organisations and researchers compile, build, and share knowledge on the history and heritage of civil society in Flanders, within its Belgian and international context (Heyrman & Weber, 2007; ODIS, n.d.-b).

After the launch of the ODIS database, other heritage and research organisations and projects joined the partnership, making ODIS one of the most widely used historical research instruments in Flanders and Brussels.^[1] As ODIS does not receive recurrent structural funding, its maintenance is made possible by the annual operational usage fees paid by the members of the partnership. In 2006, the management of the database was entrusted to a nonprofit association under Belgian law (asbl/vzw).^[2] The universities of Antwerp (UA), Brussels (VUB), Ghent (UGent), and Leuven (KU Leuven) are represented in its general assembly. The day-to-day coordination of ODIS is ensured by KADOC, which is an interfaculty centre of the Human Sciences Group of KU Leuven (University of Leuven), while LIBIS-KU Leuven is the main technical service provider.

From 2009 to 2014, a research infrastructure grant from the Hercules Foundation of the Flemish authorities enabled the development of a new ODIS database. The data model was expanded with new modules for the description of buildings, families, and events. At the same time, bilingualism (Dutch and English) was implemented, significantly increasing the database’s usability (De Maeyer, 2015). Since 2022, ODIS has been undergoing another renewal process, which we will explore more in detail in the final section of this paper.

Stimulating cross-fertilisation has always been a major goal of ODIS. As the database is used by a large network of partners, its content addresses many thematic fields at the international, national, regional, and local levels. Data sets on various social domains are interconnected, including social organisations, politics, art and culture, media, church and religion, education, care, migration, and local civil society. Furthermore, as both heritage and research organisations are involved, ODIS has a bridging function between the custodians of cultural heritage collections (mainly archivists and heritage librarians) and researchers. It offers them a joint platform where the contextual disclosure of heritage is a lever for innovative, interdisciplinary research into the history of civil society (Angelaki et al., 2019).

In January 2025, the database contained 318,026 records. They include comprehensive data sets on organisations, persons, families, buildings, and events, as well as records with basic information on the documentary heritage of civil society in the form of archives and publications (mainly periodicals). Almost half of the records (153,970; 48.4%) were published in the ODIS public catalogue (OPAC). They are available under a Creative Commons license (CC BY-NC-SA 4.0), encouraging their reuse in other noncommercial initiatives and projects. In 2024, the OPAC was visited 108,173 times by 72,720 unique visitors, resulting in 577,810 page views.^[3]

The Multifunctional Use of ODIS

The question arises as to how the members of the ODIS partnership make use of the database. Foremost, ODIS offers the members a safe and stable environment to store, supplement, update, and correlate contextual data sets and repertories. Data entry is done in two ways: manually and automatically. In 2023, about 75 people manually entered data. They included staff from the partner organisations, as well as a pool of volunteers. Manual data entry is mainly based on sources kept in the collections of the partners. Well-structured data sets that are available in digital form (e.g. an Excel file or an Access database) can be loaded into the database by the ODIS help desk, making time-consuming manual data entry unnecessary. These data conversions often involve data sets that were created in the context of research or heritage projects. By incorporating them into ODIS, they will be permanently preserved after the project ends, protecting them from dispersion and oblivion.

Second, ODIS is used to validate data – this means subjecting data to quality checks – and publish them on the World Wide Web, via the ODIS OPAC and specific OPACs on the websites of partners and projects. The partners and their data inputters autonomously decide which records are ripe for publication. However, care must be taken to protect personal data, following the European General Data Protection Regulation (GDPR) and the ensuing Belgian legislation. ODIS contains a lot of data that the GDPR considers as ‘special categories of personal data’, such as data on ‘political opinions, religious or philosophical beliefs, or trade union membership’ (Art. 9 GDPR). The general principle within the ODIS partnership is that records about living persons are never published unless the persons concerned grant permission or the records contain only information of a public nature.

End users of ODIS (outside or within the partner organisations) can approach the available data sets multifunctionally (Colla & Heyrman, 2019; Heyrman, 2025). First, ODIS functions as a digital encyclopaedia where people can look up validated basic and/or background information on organisations, persons, families, buildings, and events. The partners’ reading room services also make intensive use of the database in that way. They often answer content-related questions by referring readers to ODIS records. Sometimes, the consultation of the database even replaces the consultation of physical heritage. An example are the ODIS records on the members of religious orders and congregations, which incorporate information from the membership registers and/or personal files in the archives of those institutes, often kept in KADOC.^[4] Researchers looking for basic information on the individuals involved no longer have to come to the reading room to explore the archives.

ODIS can also be approached as a heuristic instrument. It is a joint gateway to source materials, offering basic information on (periodical) publications and archives, with (deep) links to primary catalogues and digital repositories. Many ODIS partners use the system to offer end users a repertory of the periodicals and archives in their own collection. However, the database is also suitable for the development of thematic source guides, which are not limited to the heritage collections of the partners. Examples are the Repertory of the Belgian Rural Press made up by the Interfaculty Centre for Agrarian History (KU Leuven) or the archival records that are part of the Survey Map of the Migratory Civil Society and its Cultural Heritage in Flanders, 1830–1990, a project carried out by Amsab-ISG and KADOC (CAG, 2021; Migrantenerfgoed, 2014).

Third, ODIS is used as an authority database. Links to ODIS records are included in the primary catalogues and digital repositories of the members of the ODIS partnership but sometimes also of external organisations. Establishing such links provides information seekers with validated background data on, for example, the author of a book, the organisation that published a particular brochure, or the building that is the subject of an article. Because of the collaborative nature of ODIS, partners do not need to collect those background data separately in their own systems and can benefit from each others’ data sets. A good example of the use of ODIS as an authority database are the links to records on the historical development of parish structures included in the digital repository Teneo, where KADOC’s digital/digitised collection of local parish journals is made available for consultation (Limo, n.d.).

Finally, ODIS also functions as a digital humanities research tool, enabling its partners and end users to analyse data sets using advanced search tools based on systematic data input done via thesauri and vocabularies (see further). The extensive biographical data sets in ODIS, for example, are well-suited for prosopographic research. When one has, for instance, a complete repertory of members of a particular organisation, one can easily query those data using a combination of various parameters, such as place of birth and death, education, occupations, honorary decorations, language skills, or political mandates. Often, data sets resulting from a research project are loaded into ODIS after the completion of the project to ensure their preservation and the reproducibility of the research conducted. However, some researchers use ODIS from the beginning of their project as a tool for data storage and analysis. An appropriate example is the research on book history at the Research Group Early Modern History of KU Leuven, where ODIS was the central research tool for the PhD projects Impressa Catholica Camercacensia (ICC); Impressae. Women Printers in Early Modern Antwerp, Leuven and Douai; and Manuale Lovaniense (Cammaerts, 2024; KU Leuven, 2023; Soetaert, 2019; Wyffels, 2021).

Key Features of the Database

The ODIS Data Model

The ODIS data model consists of seven interlinked main entities/modules in which organisations, persons, families, buildings, events, archives, and publications are described. Furthermore, an auxiliary module for the description of repositories is available. It is linked to the repository module of Archiefbank, the database of Archiefpunt, a cultural heritage organisation that aims to raise awareness for private archives in Flanders and Brussels (Archiefpunt, n.d.).

The structure of the ODIS modules is based on international standards wherever possible: ISAAR (CPF) for organisations, persons, and families; the DOCOMOMO Guidelines for buildings; ISAD(G) for archives; ISBD for publications; and ISDIAH for repositories (DOCOMOMO International, n.d.; International Council on Archives 2024a, 2024b, 2024c; IFLA, 2011). All modules share a parallel structure. Each of them features authority entries for identification data (e.g. names, titles, dates, and places of birth and death). Furthermore, data inputters can use both free text fields and repeatable fields/field groups. In the latter category, data are systematically inputted based on controlled thesauri and vocabularies, enabling advanced search queries. Finally, ODIS features relational field groups, a particular type of repeatable field group. These groups ensure the interconnection between the ODIS modules and enable the creation of clear, unambiguous links between records. By doing so, data are placed in a broader context and contextual information clusters are built, which – with some research effort – paint a picture of civil society and its key players in Flanders/Belgium. As only some authority entries are mandatory, the ODIS partners can decide how to build their data sets and which fields and field groups to prioritise in light of their objectives. As already mentioned above, the partners also independently determine if and when they would like to publish their data sets in the OPAC.

Multilingualism and Multicontextuality

A main goal of the ODIS Renewal Project 2009–2014 was the internationalisation of the database. In that respect, there are two pillars. The first one is multilingualism. Since 2013, ODIS has been available in both Dutch and English, and each record has a Dutch and an English version. Data inputters choose which language they want to use. As the Dutch and the English versions of a record only share some language-independent authority entries, both versions can be developed independently. Translation projects aimed at closing the gaps between the Dutch and English versions of records have already yielded great results. In parallel, strategies for the automatic exchange of more data between the language versions of a record are currently being investigated. In the future, we also hope to add additional language versions, moving from bilingualism to true multilingualism.

The second pillar is the so-called multicontextuality of the database. It means that some thesauri, which initially had a Flemish/Belgian scope, were broadened to other contexts/countries. Therefore, the thesaurus of legal forms in the organisational module and the thesauri of decorations, levels of education, political positions, political bodies, and political competences in the biographical module contain terms from 10 European contexts. The geographical thesaurus – which is hierarchically structured and based on present-day administrative geography but with historical terms also included – encompasses Europe as a whole, supplemented with some specific place names outside Europe. The multilingualism and multicontextuality of ODIS make it possible to put Flemish/Belgian data in an international perspective and allow the storage and publication of data from other European contexts, for instance within the framework of international networks and projects.

Interconnectivity

ODIS positions itself as an information node in a broad network of online data collections. Its interconnectivity takes shape in the many links to and from the database. Both ODIS partners and other institutions and projects, such as the Royal Library of Belgium (KBR) or Wikipedia, include links to ODIS records in their catalogues, databases, and digital repositories. The persistent URL of each record can easily be used for that purpose, but several catalogues of the partners also dispose of made-to-measure plug-ins that facilitate linking to ODIS. In the opposite direction, many links to external databases and catalogues are included in ODIS records. These links can be included ad hoc, but sometimes links are systematically established between ODIS and an external database. ODIS is also equipped with a database linking tool, which makes it possible to incorporate deep links to several library catalogues. Further, synchronisation procedures exist between ODIS and the archival catalogues of its partners, while the database itself is a provider of archival data for the above-mentioned Archiefbank. In collaboration with meemoo (Flemish Institute for Archives), the connection between ODIS and Wikidata is currently being strengthened on a project-by-project basis. In 2023, the authority data of about 31,000 biographical ODIS records were loaded into Wikidata; reciprocal links were added in both databases.

From ODIS 2.0 to ODIS 3.0

Striving for versatility, ODIS seeks to address and even anticipate the needs of its diverse partners and users, particularly in the domains of data visualisation, data analysis, and linked data use. Several features of the current database are somewhat outdated and no longer provide the expected functionalities. Therefore, nearly 10 years after the launch of ODIS 2.0 (2013), another comprehensive renewal of ODIS – which was recognised as a KU Leuven Core Facility in 2021 (KADOC-KU Leuven, 2024; KU Leuven, 2024) – became necessary. Propelled by a research infrastructure grant from FWO and benefiting from the technical collaboration with LIBIS, the database has embarked on a transformative journey from 2022 onwards.^[5] The core ambition of the ODIS Renewal Project 2022–2027 is the development of a new ODIS 3.0 database (backend) and user interface (frontend).

Regarding the database, the decision was made to switch from a relational Oracle database to a triple-store database developed in Virtuoso. This transition is a challenging procedure. It involves organising all fields and objects that form the ODIS data model in a Google Sheets spreadsheet. This organisation is done module by module because each module poses a specific set of challenges. The spreadsheet is then used to generate SHACL files. These files are, in turn, used to raise both the architecture of the triple-store database and a JSON API, which will allow interoperability between the database, the new frontend, and third-party users (Aernouts & Celik, n.d.; Celik, 2024; JSON:API, n.d.; W3C, 2017). The process provides the opportunity to reassess the current metadata model and update certain fields. As mentioned above, the model used in the current ODIS database comprises elements of ISAAR(CPF), ISAD(G), and ISDIAH. These standards have been merged by the International Council on Archives (ICA) into the new description standard Records in Contexts (RiC).^[6] Compliance with RiC and with the Open Standards for Linking Organisations (OSLO)^[7] of the Flemish authorities is an important aspect of the model’s redesign (International Council on Archives, 2024d; Vlaanderen, n.d.).

The new ODIS user interface will dispose of an interconnectivity and discovery layer, offering more and better connections with other catalogues, research instruments, platforms, and linked open data resources. It will be the cornerstone of a durable open access policy, based on the findable, accessible, interoperable, reusable (FAIR) principles (GO FAIR initiative, 2022). The interface will also facilitate the semantic enrichment of the ODIS content through interconnections with complementary data sets.

Further, the interface will be equipped with innovative tools to query, analyse, and visualise data sets, allowing ODIS users to respond to the challenges of present-day digital humanities research. The website will not only have a new layout but also an integrated geographical tool. This tool will make it possible to visualise all the geographical terms in a particular record on a map and spatially represent the results of advanced search queries with a geographical parameter. In that context, OpenStreetMap (OSM) is being used to refine and extend the existing geographical thesaurus (OpenStreetMap, 2024). At the time of writing, the old geodata are being mapped to the OSM model, and ways to incorporate the OSM data, either by ingesting and regularly updating them or through a live call, are being researched.

Network visualisation is also high on the agenda. Inspired by, for instance, the GND Explorer, a tool to search and represent the Integrated German Authority File (GND) managed by the German National Library, a network representation will be incorporated in ODIS records (GND-Explorer, n.d.). Users will be able to open the network of one record and, from there, further explore the multitude of relationships interconnecting records, potentially discovering network clusters at a higher level. In the academic year 2023–2024, an MSc Digital Humanities student at KU Leuven (supervised by Prof. Dr Katrien Verbert) worked on this topic. In her master’s thesis, she produced a first draft of what a network visualisation of ODIS data could look like (Jin, n.d., 2024). Her research was already very promising, and her use of Gephi to represent a select data set yielded some exciting results. Her findings will serve as a starting point for developing the visualisation tool in the frontend.

In addition to the technical developments outlined above, the ODIS Renewal Project 2022–2027 also seeks to update the mission and goals of ODIS, foster data input and quality care, implement a revamped communication strategy, and strengthen participation. The latter means, among other things, that ODIS is increasingly profiling itself as a device for citizen science. Currently, people who consult the database already pass on information by mailing the ODIS help desk. In the new public interface, each record will have a contact form, which will make it easier to submit additions, comments, and suggestions. Spreadsheet templates for supplying information will also be provided.

An interesting test case in citizen science is the crowdsourcing project Chapels in Flanders (Kapelletjes in Vlaanderen in Dutch). Between 1998 and 2007, two volunteers created an inventory with basic data on and pictures of nearly 13,700 chapels in Flanders. In 2022–2023, these data and pictures were loaded into ODIS and subsequently forwarded to the new Omeka S website Kapelletjes in Vlaanderen, developed by LIBIS on behalf of PARCUM (Museum and Expertise Centre for Religious Art and Culture) and KADOC and in collaboration with the National Geographic Institute of Belgium (NGI-IGN). The website offers all interested individuals the opportunity to complete the data sets on chapels, building on the texts already present. It is also possible to register new chapels and place them on the map. After a quality check, the information appears online. The collected data will eventually flow back to ODIS, where they will strengthen the contextual information clusters on local religious heritage (KADOC-KU Leuven & PARCUM, 2023).

Conclusion

Since its development in 2000–2003, ODIS has been used by a broad network of heritage and research organisations in Flanders and Brussels to compile, build, and share knowledge on the heritage and history of civil society. The functionalities of the device align with the mission of cultural heritage organisations such as archives and heritage libraries: to make historical sources and their context accessible and interpretable. ODIS is based on a cooperative model that creates bridges between heritage organisations and between the fields of heritage and research. It proves that cooperation is not only cost-saving and efficient but also generates significant added value.

The design and technical environment of ODIS have undergone changes over the years, with new functionalities being added. In 2022, a new transformation project was launched based on a research infrastructure grant from FWO. At the core of this project is the development of a new triple-store database and public interface. This aim stems from a desire to develop new ways to query, analyse, and visualise data sets and create an interoperability and discovery layer, strengthening the interconnectivity of ODIS. The journey to the new ODIS environment is an adventurous one. We are currently halfway through the process and on schedule to complete the project on time. We are convinced that the new features, tools, and functionalities will stimulate the innovative contextual disclosure of heritage collections and be a lever for impact-driven research initiatives, spanning boundaries and connecting various disciplines. We also hope that the investment programme will encourage organisations to join the ODIS partnership and incorporate their own data series, further strengthening the multifunctional potential of the database.

Acknowledgements

We would like to thank our colleagues Bert Aernouts (KADOC-KU Leuven), Mehmet Celik (LIBIS-KU Leuven), Peter Heyrman (KADOC-KU Leuven), Stephan Pauls (LIBIS-KU Leuven), Winand Van Meerbeek (KADOC-KU Leuven), and Roxanne Wyns (LIBIS-KU Leuven) for their valuable input.

References

Abstract

Established in 2000–2003, ODIS is a joint database on the history of civil society used by a broad network of heritage and research organisations in Flanders and Brussels. This paper provides a general introduction to ODIS and highlights how the members of the ODIS partnership deploy the system for the contextual disclosure of heritage by storing, supplementing, validating, and publishing data sets on organisations, persons, families, buildings, and events, as well as by providing basic information on documentary heritage. End users can approach ODIS as an encyclopaedia, a heuristic device, an authority database, and a digital humanities research tool. In the following section, we consider some key features of ODIS: its data model, its multilingual and multicontextual character, and its interconnectivity.

The final section is devoted to the current in-depth renewal of ODIS (2022–2027), which is being executed owing to a research infrastructure grant from the Research Foundation Flanders (FWO). At the core of the project is the development of a new triple-store database and user interface. The device will offer new ways to query, analyse, and visualise data sets, allowing ODIS to respond to the challenges of digital humanities effectively. The development of an interoperability and discovery layer will foster interconnectivity, and this layer is the cornerstone of a durable open access policy. The new system will also be profiled as a tool for citizen science. In that respect, the crowdsourcing project Chapels in Flanders (using an Omeka S website related to ODIS) is an interesting test case.

Keywords

Citizen science; Civil society; Contextual disclosure of heritage; Digital humanities; Heritage and research database; Linked open data