With the growing demand for openness in research, more and more researchers are obliged to publish their datasets and processes in ways that are easily accessible, searchable, machine-readable, reusable, and citable. This requires organizations to strategize, define, and implement a robust open-data management system as a vital part of the research mechanism. This course will focus on some of the best practices to open up research processes and data. Participants will learn to develop and enhance open-data pipelines.
The course will begin by exploring how to plan an open-data pipeline. We will address key questions: What is open data? Why go open? Then we will look at the main data categories of the research process, major elements of the open-data life cycle, and open-data management plans and policies. We will move on to understanding key components of the pipeline, including documentation, open-data acquisition methodologies, metadata standards, open-data ontology, and open-data storage structures and schemas. We will discuss data analysis and outcomes, open-source codebase, and open-data visualization.
Next we will examine how to scale up an open-data pipeline. Topics will include open-data ethics, privacy, and security; open-data quality checks; publishing platforms; open-data licenses; open issues and bug tracking; indexing, searching, and reusing open data; and changelog and version controlling. We will then turn to examples and case studies, as well as how to measure success and failure and optimize sharing and collaboration. The course will wrap up with a summary, group activities, and a Q&A session.
If time allows, we will have targeted group discussions on topics including:
Sharing our struggles in opening up research, data, and process.
Exploring common best practices across streams for publishing open data, code, and analysis.
Collaborating with different research bodies to co-create open research and data, and understanding the hurdles and benefits of this collaboration.
Limits on participation: None.
Intended audience: Early-stage to experienced researcher scholars, data enthusiasts, data practitioners, data scientists, and data librarians.
