Collaborations between Research Software Engineering and the Humanities
Kate Court, Frances Hutchings, and Rebecca Osselton
This post explores some of the ways researchers from the Faculty of Humanities and Social Science (HaSS) collaborate with the Research Software Engineering (RSE) team at Newcastle University. It is aimed at making these collaborations seem more accessible to researchers who might already have an idea how software could help their research but are not sure where to turn for help and researchers who might not have considered working with the team but could benefit. At Newcastle Uni we are employing a model to structure some of these collaborations aimed at kick-starting funding applications within digital humanities. Researchers can apply to receive RSE support at no cost to build a pilot application. Here we will explore what researchers might need in terms of data before they work with us, what we might do during a collaboration, and what the overall project process might look like.
What might a researcher bring to the team?
Researchers bring us data in all forms, from carefully collected statistics to drafts of poems from a writer’s archive. Our aim in exploring some of this variety is to demonstrate that when RSEs talk about ‘data’ we are guided by the researcher’s notion of what this is. This mind map contains examples of just some of the variety of ‘data’ that researchers might bring to the team.
What might software built by the team do?
Broadly speaking, collaborations between the RSE team and Humanities researchers can fall into these (non-exclusive) categories; collecting, analysing, and presenting data. Although these are generalisations, it is helpful when thinking about what value software might add to a research project. We have illustrated each with just a small number of examples.
Collecting data
We can help researchers collect and manage data in larger quantities, with accuracy and in the form required for analysis.
Accessible and efficient data collection and review: In a recent project, an RSE worked with researchers from Speech and Language Sciences to build a tool for speech therapists to use alongside their patients. The tool summarises the patient’s ability at the end of an online test to help the speech therapist deduce their needs at the moment, and for comparison purposes at a later date, replacing pen and paper tools. The data was made available in the format of a PDF report and also raw JSON for further processing.
Reaching new audiences and increasing scope. The team built a mobile phone app that allows users to upload photographs and condition reports about rock art they find while walking in the countryside.
Connecting with participants remotely. Writing at the time of social distancing restrictions due to coronavirus, these applications also have the advantage of being able to remotely connect researchers with their participants or allow participants to complete data generation and entry themselves. The team developed an online game where players guess what language ‘aliens’ that appear on the screen are speaking, aimed at exploring our ability to identify language without additional cues.
Analysing data
Generally, data analysis is either conducted by the software itself, or by the researcher with assistance from the software.
Visualisations. The team generated chart based visualisations of a researcher’s data about investigative journalism to stimulate research questions. These included bubble, bar, and line chart multiples, plus some work to produce a word tree and node-link diagram.
Machine analysis. We work with researchers to build software that can analyse images or text-based data. The team used machine learning to analyse a set of images sourced from Flickr to explore people’s and machine’s sense of place.
Presenting data in new ways to generate new insights. Large or complex datasets can be challenging for researchers to ‘scan’ and hold in their minds in order to see patterns. A researcher brought the team an excel document containing details of translations of classic feminist texts into different languages. Presenting this time and place-based data visually on an interactive map brought it to life and helped the researcher analyse it for chronological and geographical patterns.
Presenting data
We can help researchers present their data and/or analysis to colleagues, students, and the public. A number of projects involve presenting data in addition to collection and analysis.
Digital archive. The team produced a pilot website to collect an archive of documents, images and audio and allow users to browse the archive in creative and intriguing ways. This archive centres around time and location-based temporary archives so aims to give user’s a rich experience of artworks no longer in existence.
Applying research findings for societal impact. The team built a mobile-friendly website that guides users on a ‘storytelling’ walk, including Augmented Reality experiences. This stemmed from the researcher’s work with a particular community group that found they felt excluded from a particular area of the city. The app aims to help them overcome their perceived barriers.
Ways of working
Although there is no standard structure for RSE and Humanities collaborations, there are some broad themes in what this journey looks like. This diagram aims to reassure researchers that although they do need some clear research questions and their data ready for the start of the pilot projects, they do not necessarily need a clear picture of the software they want to produce. In fact, even if they do, it is likely to change as the project progresses. The RSE team can work with the researcher to explore their ideas and bring them to life.
Want to find out more?
To see examples of previous pilot projects undertaken with HaSS researchers, and to find out if there is a current call for applications to work with the team, visit our website. Alternatively, the team always welcomes inquiries from researchers who are interested in including us in research bids. We can advise on timescales and costing, and contribute to technical sections of applications. If you already have funding and think software could add value to your research, please get in touch for an informal chat. All our details are here.
