Skip to main content
SearchLoginLogin or Signup

Research Software Management Workflow

An in-process sketch of stages and elements to consider when managing software for research

Published onOct 11, 2022
Research Software Management Workflow

Research Software Management is an emerging field coinciding with developments in Data Management, Open Science, Privacy and Security, and Digital Sovereignty. An important part of our work at the Research Software Community Leiden is to develop a mutable set of guidelines and practices that can generate dialogue toward helping people manage their research software.

Therefore, we should consider the following document an initial sketch of the process of managing software, in need of further detail, and open to debate.

As an initial mode of organisation, we should consider three distinct phases of time that indicate specific management practices: 1) before initiating a research project, 2) during a research project, and 3) after a research project is complete. In the following outline, we will list software management elements to consider during each of these phases, along with some key questions to pose under these elements.

  1. Before initiating a research project

    • Write a Software Management Plan.

    • Compose a development and maintenance budget.

      • Consider resources in terms of financial, temporal, etc.

    • Design the identity for your research software.

      • Does the identity conflict with other copyrighted material?

      • Does the identity lend to the find-ability and accessibility of the software?

    • Consider your publication plan.

      • Does your vision for the project’s publication include the publication of all research outputs, such as data and research software?

    • Consider preregistration.

      • Does your research design include research software?

  2. During a research project

    • Organise your research software through the support infrastructure of your repository / Git environment.

      • Are you utilising version control to record the development of your software?

      • Are you including metadata with your research software to improve its discoverability?

      • Does your infrastructure include issue tracking so that issues that arise within the software can be resolved?

      • Are your documentation practices enabling accessibility and reuse? [These multifaceted aspect of development asks questions on how the software developer is recording their work in progress and how their documentation language makes their software more actionable.]

  3. After completing a research project

    • Sustain your software through archiving practices.

This should be considered an initial and incomplete outline of practices for research software management. To contribute to this document, you have several options: 1) email us at Leiden’s Centre for Digital Scholarship to discuss research software management, 2) join the CDS for workshops and events that will discuss research software, and 3) create an account on PubPub to enable your ability to annotate and comment on these publications.


Photo by Vincentiu Solomon on Unsplash

Connections
A Comment on this Pub
Comments
10
Dan Rudmann:

External Comment: What I miss here in the sense of broader categories of information is Privacy and Security. Even without the data, the research software may contain personal data from an Access Management Module. If it does, or if the dataset requires it, information security is an issue we need to address. Even more generally, since the research software will in many cases contain data, more sections from the DMP may be relevant. Or do you expect people to write a DMP and an SMP when the content asks for it?

Dan Rudmann:

External Comment: There should be many more questions indeed, in this phase. Perhaps we should sit down with some people who aided researchers in creating research software in the past to come up with these.

Dan Rudmann:

External Comment: This is not a real question. It should be mandatory!

Dan Rudmann:

External Comment: This is very hard for researchers to do in many cases. These budgets need to be reviewed!

Dan Rudmann:

External Comment: Will there be a training, as we have for DMP’s? The demand will not be as high, so mandatory training like we arranged for the Linguists and the DMP-training will not make sense. So, perhaps we could create knowledge clips as training material. We would be willing to support that.

Céline Richard:

… and if you want to make your software Open Source. The CWTS Open Science policy details the reasoning nicely: “On the one hand, making the source code openly available may enhance the sustainability of the software because others outside CWTS may contribute to maintenance and development of the software. On the other hand, keeping the source code closed may enhance the sustainability of the software because it may make it easier to generate revenues from the software. These revenues can then be invested in maintenance and development of the software.”

Céline Richard:
  • I would also add a line about why a machine-readable citation file is better than the readme file only. Also, a useful link for citation guidelines: Katz DS, Chue Hong NP, Clark T et al. Recognizing the value of software: a software citation guide [version 2; peer review: 2 approved]. F1000Research 2021, 9:1257 (https://doi.org/10.12688/f1000research.26932.2)

Céline Richard:

I would probably mention a readme file here: do you have a README file that includes installation instructions, a license, citation instructions, a list of dependencies?

Céline Richard:

adding a few words about incentives for software management would be a nice introduction: if the software is well managed, it can be shared / reused/ be made discoverable /potentially published or monetizable, bringing then regonition for the researcher.

Dan Rudmann:

Here’s an example of in-line annotation, which is a great way to hold a broad discussion around any publication on this website. Also note that signing up for a PubPub account is free for all, and the Knowledge Futures Group, the organisation that develops PubPub, takes your digital privacy seriously.