The Research Data Management Adventure is interactive fiction that takes the player through the life cycle of a project, highlighting how decisions about research data can have far-reaching consequences. We explain how the game was developed, highlight some key features, and discuss researchers’ reactions to it.

Citation: Alexander Ball, Samuel Simango and Nushrat Khan (2022), ‘Work, Research Data and Play: the RDM Adventure Game’, in Research Support Games Day 3, 15 February 2022.

This presentation provides an overview of the history of the Metadata Standards Catalog and its antecedent resources, highlights from the v2.1 release, and plans for the coming year.

Citation: Alexander Ball (2021), ‘Metadata Standards Catalog: Current Status’, in Research Data Alliance Eighteenth Plenary Meeting, 3–11 November 2021.

This game is a piece of text-based interactive fiction that raises awareness of research data management (RDM) issues in a light-hearted and entertaining way. It can be used to test the player’s understanding of the topic independently of, prior to, or following on from traditional training; the feedback it provides can also act as a form of guidance in its own right.

The storyline follows a mid-career researcher managing a large project for the first time. The player tackles RDM challenges at five stages of the project, from initial proposal to publishing results.

Game play mainly resembles a traditional gamebook, but there are some more dynamic puzzles to solve. Points are awarded or deducted for the decisions made; the rationale is summarized on a scorecard viewable at certain points. The player’s actions have consequences for how the story unfolds: with more serious mistakes, the adventure ends early, but when played from start to finish there is a bonus section where players can (in character) reflect on their performance.

While the game works best when played in one go, each of the five stages can be played in isolation, making it suitable for use within a themed workshop.

Citation: Alexander Ball, Samuel Simango and Nushrat Khan (2021), ‘Research Data Management Adventure’, in 9th International Educational Games Competition (IEGC 2021), Brighton, 22 September 2021.

Licensing research data is an effective way of ensuring legal interoperability with other data sources, so long as an appropriate licence is chosen. This presentation explores the respective consequences of withholding and providing guidance and recommendations on appropriate licences, and the knock-on effects regarding licensed and unlicensed data. It also gives an overview of the approach taken by the University of Bath in terms of advising its researchers.

This presentation was addressed to the national Norwegian committee set up in 2020 to conduct a study of licensing for sharing and reuse of research data.

Citation: Alexander Ball (2021), ‘Consequences of Recommendations for Licensing Data’, in Meeting of the Committee ‘Utredning av lisensiering for deling og gjenbruk av forskningsdata’, Zoom, 6 May 2021.

The presentation provides an overview of the history of the Metadata Standards Catalog and its antecedent resources, and the internal data model it uses. It details the information that the Catalog can store about metadata mappings and crosswalks, shows how this information is displayed, and outlines the aspirations for the Catalog as a source of new crosswalks.

Citation: Alexander Ball (2021), ‘Registering Converters in the Metadata Standards Catalog’, in Research Data Alliance Seventeenth Plenary Meeting, Edinburgh, 20–23 April 2021.

The RDA Metadata Interest Group (MIG) Metadata Element Set is intended to be both a starting point for new metadata schemes and a conduit for conversion between existing schemes. It consists of seventeen high-level elements that group together more detailed elements that can be used to characterize a metadata scheme. This talk introduces Metadata Element Set and shows its current state of development.

Citation: Alexander Ball (2020), ‘Introduction to the Recommended Metadata Element Set’, in Research Data Alliance Sixteenth Plenary Meeting: "Knowledge Ecology", Costa Rica, 9–12 November 2020.

The first in this pair of presentations provides an overview of the history of the Metadata Standards Catalog and its antecedent resources, highlights from the v2.0 release, and plans for the coming year. The second details the progress made with the Metadata Element Set in the period from April to November 2020.

Citation: Alexander Ball (2020), ‘Progress to date on the Metadata Standards Catalog and Metadata Element Set’, in Research Data Alliance Sixteenth Plenary Meeting: "Knowledge Ecology", Costa Rica, 9–12 November 2020.

The RDA Metadata Interest Group (MIG) Metadata Element Set is the outworking of a vision for metadata interoperability that follows from the FAIR Data Principles and the RDA Metadata Principles. The vision is to define a metadata package for any given use case, map between the package and relevant metadata standards, and thereby use the package as a conduit for conversion between those standards. The Metadata Element Set provides the superset for these packages. It consists of seventeen high-level elements that group together more detailed elements that can be used to characterize a metadata scheme. This talk introduces Metadata Element Set, shows its current state of development, and explains how it might be applied in an Engineering context.

Citation: Alexander Ball (2020), ‘Towards a metadata Rosetta Stone: The RDA MIG Metadata Element Set’, in RDA Research Data Management in Engineering Interest Group Seminar: "Building Metadata Standards Within Engineering Disciplines and Communities", Zoom, 9 October 2020.

Journals that lead on open data issues can place demands on data repositories in a couple of ways. They require anonymous access to underlying data during the peer-review process, and expect a working link to the data to be provided prior to paper publication. One solution is for the repository to provide a private sharing link during peer review, then ask the authors to replace this with a DOI between acceptance and publication, and publish the dataset during the final proof stage. A drawback of this is the risk that the private sharing link might be published in place of the official DOI. It would be more convenient to use the same DOI link throughout, but the social contract around DOIs is that they should be reliably citable: changes in the light of reviewer comments would be impossible. Taking inspiration from papers that are released online in their accepted but non-final state, I propose establishing a convention for using the same DOI link for restricted reviewer access, pre-release ‘holding pages’, and finalized landing pages, in such a way that the citability or otherwise of the data remains completely clear.

Citation: Alexander Ball (2020), ‘Hold That Thought: DOIs and Early Access to Datasets’, in Open Repositories 2020: "Open for All", Stellenbosch, South Africa, 1–4 June 2020.

In 2016, the FAIR Data Principles were published, crystallising community expectations around the availability of research data. It is striking that the principles place so much emphasis on the metadata accompanying the data. As a research data librarian with responsibility for an institutional data archive, it is my responsibility to ensure that the items we ingest are accompanied by metadata of sufficient quality and quantity, and that this metadata is made available in line with FAIR Data Principles. Not only does this ensure the datasets meet community expectations, but it also helps us solve practical challenges for our service delivery.

Citation: Alexander Ball (2020), ‘Metadata for Better Data’, Catalogue & Index 198 (March): 17-20.

The University of Bath operates a Research Data Archive through which it publishes mainly long-tail data underpinning published research. These datasets are assigned DOIs and we do our utmost to ensure that researchers quote those DOIs in their data access statements and reference lists. In the vast majority of cases, this works exactly as intended, but sometimes things go wrong.

Because of this, we realised we needed a comprehensive and considered set of procedures for dealing with the various ways the deposit process might go wrong. It also seemed like a good opportunity to prepare for the natural end-of-life for our datasets. Thus the Tombstone Protocol was born.

Citation: Alexander Ball (2020), ‘The Tombstone Protocol: An Undertaking for Unfortunate Events’, in 15th International Digital Curation Conference (IDCC20): "Collective Curation: the many hands that make data work", Croke Park, Dublin, 17–20 February 2020.

In the Digital Curation Centre guide ‘How to Cite Datasets and Link to Publications’, the authors compared several dataset citation format proposals in order to come up with a list of elements that any dataset citation should include. One issue that came up was the relationship between a dataset identifier and its location, which leads to an interesting philosophical discussion as to whether identifiers such as DOIs do and should identify datasets as abstract, evolving entities or as static, reliably accessible versions and snapshots.

Citation: Alexander Ball (2019), ‘A Generalist Perspective on Data Citation’, in 2019 ESIP Summer Meeting: "Data to Action: Increasing the Use and Value of Earth Science Data and Information", Tacoma, Washington, 16–19 July 2019.

The University of Bath has operated an EPrints-based Research Data Archive since 2015, and in that time it has been heavily customized to suit local needs. In order to manage these customizations, an Enterprise GitHub repository was used first to record and subsequently to deploy changes to live and test servers. The Git-based workflow has enabled new features and fixes to be implemented and deployed much more rapidly and safely than before. For example, we introduced functionality to apply individual licences to all files in a dataset in one go; and a new responsive landing page design was introduced, including an updated JSON-LD header block for improved visibility through Google Dataset Search. One of the larger developments was to simplify the user journey for depositing a dataset, which previously required users to begin in the University's Pure CRIS before continuing in the Archive: work was done to improve the alignment of the data models in Pure and the Archive, and an XML feed was introduced to allow newly published datasets to be synchronized into Pure. These changes have improved the user experience of both end users and repository staff.

Citation: Alexander Ball (2019), ‘Scratching That Itch: Process Improvement and Problem Solving in the University of Bath Research Data Archive’, in Open Repositories 2019: "All the User Needs", Universität Hamburg, Germany, 10–13 June 2019.

In this talk, we discuss Creative Commons licensing and choosing appropriate licences for data.

Citation: Alexander Ball (2019), ‘Licensing Research Data’, in PGcert Biocuration Webinars - Module 3, Online, 28 May 2019.

Faced with Research Data Management (RDM) requirements for the first time, it is all too easy for researchers to interpret them as another administrative barrier to completing their research. Advocacy is needed to convince them that RDM deals with genuine issues, and that addressing them brings tangible benefits to the researchers themselves as well as the wider academic community. Librarians at the University of Bath and Stellenbosch University are collaborating on a serious game aimed at doctoral research students that will introduce a variety of RDM topics and demonstrate the possible consequences of good and poor practice. It takes the form of a text adventure, a form of interactive fiction used by early computer games such as Colossal Cave and Zork, and made popular in print by book series such as Choose Your Own Adventure and Fighting Fantasy. In the current game, the player takes the role of an early career researcher to whom RDM has been delegated by the principal investigator of a project. The player progresses through the RDM lifecycle from writing a Data Management Plan to publishing data underlying a journal article, encountering various challenges and setbacks on the way. As well as the narrative feedback for the player’s actions, the game awards more points for better practice, and terminates early after particularly poor choices. The game will be used to introduce students to the range of RDM activities, advocate the need for RDM, and allow students to self-assess their understanding.

Citation: Alexander Ball and Samuel Simango (2018), ‘Serious Gaming: Creating a Research Data Management Adventure’, in 15th Stellenbosch University Library Symposium: "Smarter Libraries: User Experience (UX) in Action", Stellenbosch Institute for Advance Studies, South Africa, 15–16 November 2018.

This presentation gives an overview of the work of the RDA Metadata Interest Group. Metadata are essential for FAIR data, which is one reason why the group published a set of Metadata Principles that explain the nature of metadata and their role in research, and set out a vision for achieving interoperability and machine understanding of metadata. This includes characterising metadata standards using the Metadata Interest Group Recommended Metadata Element Set and using the Metadata Standards Catalog to match and map between them.

Citation: Keith Jeffery and Alexander Ball (2018), ‘RDA Metadata’, in GO FAIR Metadata for Machines (M4M) Workshop, Leiden, The Netherlands, 15–16 October 2018.