Evidence shows that practice-based learning is beneficial to students’ understanding of threshold concepts in all disciplines. Teaching activities that draw on research data provide students with real-world examples of how they might apply this new knowledge, and this reinforces their understanding. While research data from projects in disciplines such as computer science or economics is widely published, it is only recently that humanities scholars, in particular historians, have started considering publishing their research data in digital format. Using a case study of a 5-year funded collaborative project between historians and computer scientists, this paper discusses how the research data was created and applied in a classroom context, providing students with a ’real-world’ experience of working as a historian. It shows how the project data developed from a basic transcription of historical records into a fully enriched open dataset that can be used by teachers from a range of disciplines including history, historical geography, demography, computer science and medicine/medical humanities. It concludes that lessons planned around such open research data comprise a valuable educational resource to teachers and students.

Census returns are a critical source of information for governments globally. They underpin a wide spectrum of public planning including health, housing, work and education. Historically, census forms have captured names, places, dates, age, occupation, family structure, and religion. In more recent times, sexual orientation and ethnicity, queries that can be intrusive to vulnerable communities, have been added to the criteria, and for such reasons data security is of paramount importance. Most governments restrict access to individual census returns, presenting the data in aggregate report format. The Irish government is particularly strict, enforcing a statutory closure period of 100 years. An exception was made for the Irish 1911 census which were digitised and released for free online consultation in 2009 [1]. They are an excellent source for genealogists and historians alike but exist as separate digital siloes. This project uses an eXtreme Model-Driven Development (XMDD) environment to create linkages between both datasets. It will discuss the development process of the CensusIrl application and the process used in developing the matching algorithm used. We will discuss the census records and the data cleansing process used in creating the initial proof of concept application. We detail the different approaches to the development life-cycle of the application and describe the different utilises used in the sanitation of data points in the records and the match-making process.

Death and Burial Data: Ireland 1864–1922 (DBDIrl), is a digital humanities project, which uses historical civil registration of death as its primary dataset. The overarching aim of this project is to provide enriched and clean historical Irish data for analysis, in a eXtreme Model-Driven Development (XMDD) fashion. This paper discusses how e-learning environments were used to enrich these partially indexed data in an online, hybrid and blended learning group instruction format over four years. It describes how the DBDIrl data entry application, called Historian Dime App (HDA), evolved over a number of iterations to create a more user friendly interface, in an interdisciplinary collaboration of historians and computer scientists enabled by the XMDD approach. It discusses how the development process of HDA benefitted successive cohorts of history students engaged in a curricular Practice-based learning (PBL) project that follows a transcribathon model as defined by the Folger Library11https://folgerpedia.folger.edu/Transcribathon, We adapted the model for postgraduate teaching and learning in the humanities and took a reflexive approach to student/user feedback to evolve the HDA over four versions. It resulted in enhanced features, higher rates of user satisfaction, and a more responsive data curation and storage mechanism. This effort achieved our original aim of obtaining clean and accurate outputs from the students' project work.

Abstract With the heterogeneity of the industry 4.0 world, and more generally of the Cyberphysical Systems realm, the quest towards a platform approach to solve the interoperability problem is front and centre to any system and system-of-systems project. Traditional approaches cover individual aspects, like data exchange formats and published interfaces. They may adhere to some standard, however they hardly cover the production of the integration layer, which is implemented as bespoke glue code that is hard to produce and even harder to maintain. Therefore, the traditional integration approach often leads to poor code quality, further increasing the time and cost and reducing the agility, and a high reliance on the individual development skills. We are instead tackling the interoperability challenge by building a model driven/low-code Digital Thread platform that 1) systematizes the integration methodology, 2) provides methods and techniques for the individual integrations based on a layered Domain Specific Languages (DSL) approach, 3) through the DSLs it covers the integration space domain by domain, technology by technology, and is thus highly generalizable and reusable, 4) showcases a first collection of examples from the domains of robotics, IoT, data analytics, AI/ML and web applications, 5) brings cohesiveness to the aforementioned heterogeneous platform, and 6) is easier to understand and maintain, even by not specialized programmers. We showcase the power, versatility and the potential of the Digital Thread platform on four interoperability case studies: the generic extension to REST services, to robotics through the UR family of robots, to the integration of various external databases (for data integration) and to the provision of data analytics capabilities in R.