The WissKI User Meeting 2024 (WAT24) took place on 20-21 November in Nuremberg. This year again we were pleased to welcome a growing number of 70 participants (35 of them on-site)! It was the first user meeting to be held almost entirely in English, as we are increasingly addressing international users. Robert Nasarek (GNM Nuremberg), Community Manager and WissKI developer, opened and moderated the programme. (PRESENTATION) (NOTES)
-- 20.11.2024 --
We started with a keynote...
presented by Harald Sack (FIZ/KIT Karlsruhe) and focused on "Knowledge (Graph) Driven Research Data Management in the Age of LLMs". It explored the complexities of research data management across disciplines, highlighting the diversity of research data and the challenges of metadata standardisation. The presentation highlighted the role of knowledge graphs in research data management, discussing centralisation versus federation and the NFDI4Culture approach using modular ontologies for data management. It explored the use of Large Language Models (LLMs) in ontology engineering, addressing challenges such as ontology design, evaluation and maintenance, and emphasising the support of ontology-based applications by improving query answering, recommendation systems and text summarisation. Sack noted that the integration of LLMs and ontologies can improve the accuracy and reliability of AI systems, especially in complex domains such as healthcare and finance. The session also covered the National Research Data Infrastructure (NFDI) programme in Germany, which aims to make data a common good for research, and whose consortium NFDI4Culture is developing, among other things, an NFDI4Culture Knowledge Graph, which will serve as a central index for decentralised repositories and support data integration and access. (PRESENTATION)
In the WAT24 news...
we learned from Robert Nasarek (GNM Nuremberg) that Drupal is currently putting a strong focus on the Starshot initiative, and that the development of recipes, the project browser and automatic updates are being pushed forward as part of this. This is a happy coincidence, as it coincides with the SODa project's (see below) plans to provide pre-configured WissKI instances quickly and easily. Of course, Drupal is not unaffected by the AI hype, and DrupalAI was part of many presentations and discussions at the Drupal Developer Days in Burgas. In the medium term, it will also influence the Drupal/WissKI world, especially in the areas of semantic search, content generation and standardisation, normalisation and quality control.
In addition to the release of the Drupal 11 compatible WissKI 4.0, there are also some innovations in the WissKI context. First prototype experiments with Drupal recipes are underway to provide ready-made WissKI environments. Docker images, default data models, ontologies and module setups for different use cases are currently being developed (see SODa Repository). Since not everything can be solved with recipes, more and more Drush commands are being implemented for various tasks. One of the most important developments, however, is the Open Graph Database, which extends the RDF4J triplestore with a user and rights management environment using token authentication. We now have a powerful, flexible, secure and most importantly open and free graph database for our Triple! (PRESENTATION)
There is a place for enthusiasts of semantic data networking...
and it is called Association for Semantic Data Processing (Interessengemeinschaft für Semantische Datenverarbeitung - IGSD e.V). Mark Fichtner (GNM Nuremberg) informed about the general activities and special services of the association. It is pleasing that for the first time it was possible to employ staff for a research project, which further strengthens the role of the association as an actor in the community.
Our WissKI community...
continues to grow and diversify. Johannes Schäffer (HU Berlin) presented some interesting aspects of our users and our community survey. A large part of our users are still located in German-speaking regions, especially in Bavaria, Baden-Württemberg and Berlin. However, we are also seeing an increase in members at institutions in Austria and Switzerland, and have welcomed online participants from India, China and several European countries.
Our non-representative survey gave us a small insight into our users. Most of them know WissKI from colleagues or friends or because it was used in a project. They mainly come from the GLAM sector and most of them have been working with WissKI for more than a year. The users generally appreciate a guided installation, the availability of interfaces, the reference or linking of standard data and the possibility of semantic data modelling. They are mostly data modernisers, project managers and site builders and have already gained experience with other research data management systems (such as Omeka S, WikiBase, Museumplus or Research Space). Almost all respondents have access to competent IT support and have already worked with WissKI in other projects. The most frequently used community services are the informal contact with developers and other WissKI users, the WissKI user meeting and the news section on the WissKI website. The biggest need is (unfortunately still) good documentation and tutorials. We are very pleased that more than half of the participants rated WissKI as better than other research data management systems and that two thirds of the participants would recommend WissKI to others.
A intelligence cluster for collections, objects and data Literacy (SODa)...
was launched in December 2023 and Johannes Schäffer (HU Berlin) gave us an overview of the goals and structure of the project. The SODa project (german acronym for "Sammlungen, Objekte, Datenkompetenzen") focuses on improving the management and use of university collections in Germany. There are approximately 1,400 collections at about 130 universities and colleges, comprising millions of objects from various disciplines. These collections are primarily used for research, teaching and public knowledge transfer. The mission of SODa is to support university collections staff and researchers in the management and publication of collections data and in data-based research. The project emphasises data literacy and focuses on the life cycle of research data management. This includes creating and enriching object data, ensuring data quality and compatibility, and addressing legal and ethical considerations. The project also aims to ensure the long-term availability of data in accordance with the FAIR principles. Data-driven research is a key component, involving digital provenance research, analysis and visualisation of structured data, new digitisation methods (especially for 3D data), automated data analysis and machine learning, and documentation of conservation and restoration processes. The project structure includes different areas of expertise such as cataloguing and data management, ethics and law, documentation of conservation and restoration, machine learning, digital provenance research, 3D visualisation, infrastructure development and science communication. The main institutions involved are the Humboldt University of Berlin, the Friedrich-Alexander University Erlangen-Nuremberg, the Germanisches Nationalmuseum and other partners. SODa offers a range of programme activities including consulting, workshops, open forums, open educational resources, participation in working groups, a summer school, a barcamp, a semantic coworking space, and science communication through blogs, showcases and social media. (PRESENTATION)
A Semantic Co-Working Space is build...
as part of the SODa project and shall provide all the necessary tools for a complete research data lifecycle in a single digital workspace. Robert Nasarek (GNM Nuremberg) shows how the SCS will facilitates data handling and lifecycle processes such as data modelling, acquisition, enrichment, publication, transformation, migration, analysis, storage and processing. Therefore applications such as WissKI, SQL, Jupyter Notebook, Protégé, Drive, Triplestore, Draw.io and OpenRefine will be included. The Semantic Coworking Space will provide out-of-the-box solutions with features such as single sign-on and integrated applications, and should be free of charge. (PRESENTATION)
Much work was done in the WissKI Lab...
at the FAU Competence Center for Research Data and Information. Michael Kohlhase (CDI/FAU Erlangen-Nürnberg) gave an introduction to the lab, which provides comprehensive services for researchers using the WissKI platform. These include project consulting, hosting, maintenance and data modelling support. The lab offers a sustainable funding scheme, with basic services free of charge and additional services integrated into project proposals, coordinated with the DFG. It hosts numerous WissKIs, including the FAU-WissKICloud and "Objekte im Netz", and focuses on improving WissKI by automating tasks and ensuring that datasets are FAIR (Findable, Accessible, Interoperable, Reusable). Initiatives aim to strengthen WissKI as a virtual research environment and to support a global WissKI Commons. The lab develops tools such as the WissKI Distillery, Data Viewer, API for data import/export and TIPSY for visualisation of path builders. They have worked on projects such as NFDI4Objects and SODa. (PRESENTATION)
A new way to visualize Pathbuilders...
was developed by Tom Wiesing (CDI/FAU Erlangen-Nürnberg). Pathbuilders are key components of the WissKI system, which operates as a middleware between the Drupal content management system and the RDF triplestore storage. The presentation aimed to help users understand the data model created by Pathbuilders, which organises data in a graph database using triples. Tom explained the concepts of bundles, fields and paths, where bundles represent object types and fields their properties, each associated with a path consisting of concepts and predicates. The presentation introduced TIPSY (Tom's Inspector for Pathbuilders), a tool designed to simplify the visualisation of these paths. Built with TypeScript, React and Vite, TIPSY runs in a web browser and allows users to visualise their pathbuilders locally, without sending data to a server. The presentation included a live demo of TIPSY, showing its ability to generate visualisations by drawing nodes for classes and edges for predicates. The presentation ended with an invitation for questions and a mention of an upcoming workshop for further exploration of the tool. (PRESENTATION)
Data integration and harmonisation...
plays an important role in the context of the NFDI4Objects project, which focuses on making data from science and research FAIR (Findable, Accessible, Interoperable, Reusable). NFDI4Objects addresses the tangible heritage of human and environmental history and involves 14 institutions and over 60 participants, including FAU Erlangen-Nürnberg. Sarah Wagner (CDI/FAU Erlangen-Nürnberg) gave a brief overview of the project and how it is organised in several Task Areas, with Task Area 6 focusing on data qualification, integration and harmonisation. The FAU Competence Centre for Research Data & Information and the Klassik Stiftung Weimar are developing the N4O Objects Ontology based on CIDOC CRM for describing research data and creating a data model for object biographies using WissKI. Object biographies apply the concept of biography to cultural heritage objects, documenting their contexts, changes of ownership and research activities. The Behaim Globe serves as a use case to illustrate the complex challenges of history and data integration. The project is also developing a minimal dataset to connect core data across different task areas, facilitating integration into the N4O Knowledge Graph. The presentation highlighted the importance of detailed provenance information and the potential of object biographies to provide a meta-perspective to cultural heritage documentation. (PRESENTATION)
A report on the use of WissKI for anatomy collections...
was given by Sara Doll (University of Heidelberg). The presentation discussed the use of WissKI for cataloguing and documentation. Since 2022, WissKI has been used to manage the complete documentation of the collection, which includes a wide range of anatomical objects such as skulls, models and instruments. The migration to WissKI involved transferring data from previous systems and implementing restoration reports. The documentation process records various details such as inventory numbers, object types, descriptions and provenance. Challenges include non-standardised restoration protocols and terminology inconsistencies during data transfer. WissKI has improved the accuracy and efficiency of documentation, supported by meetings and technical assistance. Looking ahead, the team faces the challenge of implementing embryo sections into the system, with questions about scanning methods and presentation in WissKI. The report highlights the ongoing efforts to adapt and optimise the use of WissKI for the specific needs of the anatomy collection. (PRESENTATION)
Eight workshops...
were held at the end of the first day of the WissKI user meeting; one introductory workshop for beginners (Mark Fichtner, GNM Nuremberg) and seven advanced workshops. The latter covered 2D (Robert Nasarek, GNM Nuremberg), 3D (Julian Cremerius, FAU Erlangen-Nuremberg), research data management (Anna Gnyp HU Berlin, Rebekka Reichert, HU Berlin, Gudrun Schwenk, IGSD München), search and retrieval (Robert Nasarek, GNM Nuremberg), law and ethics (Katharina Leyrer, FAU Erlangen-Nuremberg), TIPSY (Tom Wiesing, CDI/FAU Erlangen-Nuremberg) and useful modules (Robert Nasarek, GNM Nuremberg). Thanks to the many SODa speakers who made this possible!
-- 21.11.2024 --
To connect TEI-based digital editions and WissKI databases...
is a quite complicated task, which Gustavo Riva (UB Heidelberg) has been adressed in his presentation using the duerer.online project as a case study at the beginning of the second day of the WissKI User Meeting. The initiative is part of Special Research Area 933 - Material Text Cultures and aims to integrate Text Encoding Initiative (TEI) files with WissKI in order to improve digital editions. The project involves linking TEI entities with GND (Gemeinsame Normdatei) references to provide structured information about persons and other entities within digital editions. This integration facilitates the online presentation and indexing of TEI files, using the Heidelberg Infrastructure for Digital Editions (heiEDITIONS) and eXist-db for data management. The aim is to create a seamless connection between TEI-based content and WissKI, enabling enriched digital editions that are more accessible and informative. The presentation highlighted the technical structure and workflow of the integration process and demonstrated the potential for enhanced digital scholarship through this approach. (PRESENTATION)
There is an open repository for hypothetical 3D reconstructions...
and it is part of the project Computer-based Visualization of Architectural Cultural Heritage (CoVHer). Igor Bajena (University of Mainz/ University of Bologna) gave us insights. The CoVHer project is funded by European Commission and aims to improve digital literacy in higher education by developing guidelines and methodologies for the creation, visualisation and evaluation of 3D models of non-existent or hypothetical architectural artefacts. The project is aligned with the UNESCO Charter for the Preservation of Digital Heritage and follows the FAIR Principles and the London Charter for Documentation. The goals of the repository include enabling web-based exploration of 3D models, ensuring user-friendly access, and maintaining an open service. The documentation process includes the use of ontologies and data models such as CIDOC CRM and its extensions to address gaps in 3D model documentation. The project also explores the use of a 3D viewer for Drupal to improve the user experience and plans future work to offer the 3D repository as a WissKI cloud service, define use cases and share it with the community. (PRESENTATION)
WissKI for Africa Multiple...
is the name of the digital research environment at the University of Bayreuth, which works with international partners in Nigeria, Kenya, Burkina Faso, South Africa and Brazil and focuses on the transformation of African studies through 72 interdisciplinary and multinational research projects. Neeraj Thandayan Viswajith, Marisol Vázquez Llamas and Jiayu Yang (all from the University of Bayreuth) explained how WissKI was chosen for its open source framework, flexibility and ability to handle semantic information, and gave an insight into the supporting tasks associated with the project, such as research data management training, data collection, curation and standardisation. The research data workflow involves the management of various data types, including artefacts, audio, images and digital content, using a flexible data pipeline supported by MongoDB. Technical, ethical and policy challenges arise during the project, as well as compatibility issues, digital literacy gaps and cross-border data policy compliance. Solutions include improving guidance, enriching metadata with NLP, optimising system efficiency and ensuring data security. Future work aims to improve user experience, multilingual support and offline data collection processes. (PRESENTATION)
How WissKI helps with insights from the short 11. century...
was addressed by Tara Andrews (University of Vienna) in her presentation on RELEVEN. This project aims to re-evaluate the eleventh century by using the STAR (Structured Assertion Record) model to link historical events and entities. The STAR model extends the generic Linked Open Data (LOD) model by incorporating CIDOC-CRM standards to represent assertions, evidence and authority in historical data. The project uses WissKI as its data entry system due to its flexibility in handling complex data structures, despite challenges such as the need for numerous subclasses and the inability to filter by specific properties. The team has developed methods such as scraping scripts and structured web forms to manage data entry, while also exploring the use of WissKI's API to integrate data into non-Drupal websites. The presentation highlights the innovative use of WissKI to manage historical data, acknowledges the challenges and "abuses" of the system to meet their needs, and points to ongoing developments to improve data integration and usability. (PRESENTATION)
Role in the barrels, taste all the flavours...
could be the motto of the VIGO project "WissKI Barrels" recently approved by the DFG and presented by Robert Nasarek (GNM Nuremberg). The aim of the project is to develop specialised WissKI systems, known as "WissKI Flavours", tailored to different domains such as museums, archives and 3D models. This should improve the usability and accessibility of WissKI, allowing users to focus on the unique aspects of their projects without the need for extensive setup, thereby supporting diverse research and cultural heritage applications. The initiative builds on the existing strengths of WissKI, which has been instrumental in synthesising subject data models from several projects. By identifying recurring basic entities, modules and system setups a WissKI Flavour provides users with pre-configured systems that streamline the management of research data. These systems will undergo expert review and feedback before being released as online services. The workflow involves collecting pathbuilders and ontologies, building and reviewing subject-specific WissKI systems, and releasing them for use. Volunteers are encouraged to contribute by exporting and documenting data models to ensure consistency and effectiveness. Deployment options include WissKI Docker environments and cloud services, with a focus on flexible and efficient data management solutions. (PRESENTATION)
The WissKI World Café...
was a format to discuss various topics such as support, development, modelling and infrastructure,
modelling or infrastructure to collect and discuss needs and problems in changing small groups, but also to socialise or to
and discuss needs and problems, but also to socialise or to contribute and exchange wishes and solutions.
We closed our WissKI User Meeting...
with a sum up of the topics and insights and encouraged everybody to join our WissKI Chat Channel, stay tuned on Mastodon and on our news section, visit our quarterlies and - for sure - the next WissKI User Meeting from 12-13 November 2025.