ARDC Survey Reveals Challenges and Opportunities in Research Software Discovery

Research software comes in all shapes and sizes – scripts, code, notebooks, computational workflows, libraries, modules, frameworks, utilities, applications, and so on. Increasingly, it’s recognised as critical to research and as a first-class research output in itself, having a great potential to be reapplied and built upon. 

But for research software to be valuable, it must first be findable. Improved software findability benefits research institutes, research funding bodies, and infrastructure providers as well as researchers. As a key step to making research software more visible, the ARDC asked researchers how they find software in a survey and commissioned Dr Frankie Stevens to analyse the results. The report has recently been published, throwing light on the motives behind, approaches to, challenges in, and expectations for research software discovery.

“Understanding how researchers do or don’t find software, broken down by discipline and level of coding skill, is critical to building discovery infrastructure that actually meets researchers’ needs. We see this report as an important first step to building discovery infrastructure that is relevant and fit for purpose.” said Dr Tom Honeyman, Manager of the ARDC’s Research Software Program.

Survey Results

The survey asked researchers how they find software, what stops them from finding it, and what they would like to see in a software repository or catalogue. In addition to considering the responses of researchers as a single cohort, the results are further broken down in the report based on whether they can write code or not, what research field(s) they work in, and whether they identify as a research software engineer (RSE) or not. This is so that we could better surface nuanced solutions that meet researchers where they are at. The results breakdown sometimes showed uniform patterns of behaviour and priorities, and sometimes showed marked differences.

Considering researchers overall:

• on a 5-point scale from “not important at all” to “essential”, 79% of the respondents rated research software as “essential” to research, and 13% said it’s “very important”
• the most popular method of discovery is to use search engines like Google (83%), and the second most popular method is to ask peers for recommendations (70%). 

Considering hindrances to researchers who code and those that don’t: 

• non-coders most often said they don’t know how to search and how to evaluate the search results
• coders most often said their requirements for software are too particular. 

Considering hindrances to researchers in specific domains, respondents working in:

• humanities, arts, social sciences, and Indigenous studies are challenged by a lack of software searching skills and ability to evaluate the search results 
• environmental research more often have trouble accessing software referenced in publications. 

Further results can be found in the report.

Recommendations

Based on the survey’s findings, the report suggests several actions:

• Software infrastructure providers can optimise for general-purpose search engines to make their holdings more discoverable.
• Registries and institutional repositories can be improved on to offer more tailored resources, address the different needs of coders and non-coders, and incorporate the information software searchers value most.
• Researchers, especially those that don’t code, can be assisted to develop their software searching skills and the confidence to evaluate search results.

The Next Steps

Read the full report and the supplementary materials.

This report is just one part of the ARDC’s project to improve research software visibility. In the first half of 2023, the ARDC will begin a public consultation process to identify specific pieces of national level infrastructure to be developed to better enable software discoverability. If you’d like to share your thoughts on what additional information could better inform this or if you want to register your interest to be involved in the consultation, please contact us.

The ARDC is funded through the National Collaborative Research Infrastructure Strategy (NCRIS) to support national digital research infrastructure for Australian researchers.