Chapter 3 URSSI Plan
Throughout the URSSI conceptualization process, the team worked on overall planning iteratively, using slides at workshops and the URSSI website to disseminate current thoughts, using the survey and the ethnographic studies as inputs, using workshops (including slides and shared note documents) and the discussion forum (using Discourse on https://discuss.urssi.us) to gather and record feedback. Together, these activities plus internal discussions of the team and a focused mission and vision working group have led to most of the content of this document.
3.1 Mission and Vision
URSSI PIs began to develop mission and vision statements in late 2018. The intent of these statements was to succinctly state the purpose for URSSI’s existence (mission) and based on that purpose what URSSI strives to achieve (vision). The process for developing these statements was:
At the first two workshops, we asked participants about the goals and vision that a US research software sustainability institute should pursue.
The URSSI PIs then identified institutions with similar goals and collected their mission and vision statements for comparison.
Eight members of the Senior Personnel and Advisory Committee participated in a working group focused on developing URSSI’s mission and vision statement.
Each working group participant provided a written set of three mission and three vision statements based on the conceptualization proposal, workshop participant’s feedback, and guidance from the URSSI PIs.
Neil Chue Hong, director of the Software Sustainability Institute in the UK, facilitated a synthesis of the working group’s feedback. This synthesis process included a teleconference with participants to discuss and refine versions of each mission and vision statement.
The URSSI PIs then presented drafts of each statement to workshop participants at the final URSSI meeting in Chicago and refined a final version of each statement.
The mission and vision of URSSI are as follows:
Mission: Our mission is to improve the recognition, development, and use of software for a more sustainable research enterprise. We achieve this mission through collaboratively developing education, outreach, and software services that emphasize open, transparent reproducible, and cooperative practices. URSSI is an institute for software expertise as well as a social infrastructure that promotes an inclusive and diverse community of research software engineers, maintainers, contributors, and users.
Vision: Empowered people, building better software, enabling exceptional research
3.2 Planning assumptions, methods, and principles
The following assumptions are inputs to our planning process:
Our plan is based on the idea of proposing to NSF, and therefore, focuses on the US community. We will, however, work with like-minded organization both inside and outside the US, including the Research Software Alliance (ReSA). (Also see the potential partners listed the next subsection.)
We will propose an institute with a 5-year duration and a potential 5-year extension, based on NSF’s current institutes and published documents.
The budget of URSSI will be $3m-$5m per year. We take $3m as the baseline minimum funding level needed to operate an institute, below which an institute is not the right method to achieve the mission and vision, and then build higher cost and higher return activities on that baseline.
There is demonstrated interest in supporting some URSSI goals from private foundations and potential interest from other federal agencies, so URSSI activities should be planned as a set of reinforcing (but separable) activities, potentially on different timelines.
There is a strong need for improved software sustainability across all fields, and URSSI cannot solve this need in all fields by itself.
There are many partners who can help URSSI achieve some of its goals, as there is overlap between these goals and the goals of those partners (see the next subsection)
It is not practical for URSSI to directly work with the US research software development community (all of those who develop research software in the US, including in academia, national laboratories, and industry) due to the small size of URSSI and the very large size of this community, so URSSI needs to work indirectly by leveraging other groups.
Given the mission and vision, and our assumptions, we plan to use a set of methods to ensure that URSSI is successful:
URSSI must choose a set of initial targeted communities and efforts, focus on them for some period (until some predefined metrics of success are met or some predefined time period has passed without success), then move on to the next set of targets
URSSI must work closely with partners to amplify its efforts (see the next subsection)
URSSI must be clever about using resources, attempting to achieve multiple outputs for activities whenever possible.
Leverage existing organizations for authority, credibility, resources
Focus on people, not projects
Embed diversity and inclusion within all aspects of URSSI
Leverage available resources (software, services, credit systems, training materials, etc.) where possible rather than reinvent
Share resources we create to allow others to reuse and build on elsewhere
Respect and learn from volunteers
Activities should have an end or a sustainability plan beyond URSSI
Distinguish between software quality measures (badging, intrinsic) and software impact measures (reuse, citation)
Sustain software by sustaining its communities (stewards, developers, maintainers, leaders, active users)
Coordinate activities rather than start new ones when possible
Advocate and promote a variety of solutions to challenges, recognizing that no one solution is likely to work for all stakeholders or in all situations
Some topics are partially or completely out-of-scope for URSSI:
Software Commercialization. As an institute, URSSI will position itself as a broker to existing bodies of knowledge or expertise, and seek to fill needed gaps where no expertise already exists. Relevant to this positioning is a route to research software sustainability through software commercialization. We view this path to sustainability as already well supported by existing programs and initiatives, some of which are already funded by NSF. As a principle, URSSI is not opposed to or discouraging of research software commercialization. We will actively seek to learn from and collaborate with the commercial software ecosystem, including technology transfer offices that provide structured pathways to commercialization for research software in academic institutions. In this role, URSSI might guide people in deciding if commercialization is a realistic or good choice for their software project, and connect those who decide that it is to existing support programs like I-Corps. What we believe URSSI can uniquely provide, and what we find to be lacking, is support for alternative routes to sustainability through open-source or non-commercial fiscal sponsorship. While these alternative routes are obviously less straightforward, and more challenging they remain the most viable option for improving the long-term accessibility and impact of most research investments in software. A majority of even the best research software projects won’t be commercially viable given the size and scope of their audience. It is this critical gap, between commercialization and open-source, that URSSI seeks to close.
Software Development and Maintenance. URSSI staff will not develop or maintain research software, nor will URSSI provide funds to projects to do this, other than as part of focused incubation or training activity. Given the small size of URSSI and the large size of the software development and maintenance community, such hands-on software work is not a useful way for URSSI to have an impact.
Infrastructure. URSSI will not run infrastructure for the research software community. If new infrastructure is needed, URSSI will work with other organizations to instigate projects to create and maintain that infrastructure, run by those other organizations.
to create and maintain that infrastructure, run [TODO: what else should be out-of-scope for URSSI?]
URSSI will work with a variety of partners with whom we have overlapping goals. For the purpose of this plan, we list partners with whom we will seek to work, recognizing that these potential partner organizations have not made a commitment, and such commitments will depend on specific activities, and specific funding opportunities. In the following table, we list these potential partners and the areas in which we see an opportunity for collaboration. See the appropriate chapters (e.g. Community & outreach (Chapter 4), Education & training (Chapter 5), etc.) for the specific partnering opportunities in each area of URSSI’s work.
|Potential partner||Partnering area(s)|
|Other NSF SI2/CSSI Institutes & Centers of Excellence (e.g., SGCI, MolSSI, IRIS-HEP)||policy, education & training, incubator, community & outreach|
|Research Software Alliance (ReSA)||policy, community & outreach|
|Software Sustainability Institute (SSI)||policy, education & training, incubator, community & outreach|
|Better Scientific Software (BSSw)||policy, education & training, community & outreach|
|Fiscal sponsors (e.g., NumFOCUS, Code for Science & Society)||policy, incubator, community & outreach|
|US-RSE Association||policy, education & training, community & outreach|
|(UK) RSE Society||policy, education & training, community & outreach|
|Academic Data Science Alliance (ASDA)||policy, community & outreach|
|Research Data Alliance (RDA)||policy, education & training|
|American Association for the Advancement of Science (AAAS) Science & Technology Policy Fellows||policy, community & outreach|
|Campus Research Computing Consortium (CaRCC)||policy|
|Organisation for Economic Co-operation and Development (OECD)||policy|
|The Carpenties||education & training|
|Center for Scientific Collaboration and Community Engagement||community and engagement|
3.4 Desired impact
URSSI’s ultimate desired impact, as stated in the vision, is on scholarly research in all fields. URSSI aims to achieve this impact by enabling and encouraging the contribution of the resources and training to the research software ecosystem that will improve research software and to empower the people who create and maintain that software. In the remainder of this chapter, for each of these three elements of impact, we describe our activities that we believe will contribute to them. Additionally, in Chapter 10 (Metrics and Evaluation), we show the full set of activities and their intended impacts.
For software, URSSI aims to improve the sustainability of research software by
Developing and curating best practices for software projects, including examining practices in industry, for testing, governance, codes of conduct, continuous integration.
Sharing information and practices through communities calls, training opportunities, an incubator, and a community-developed book
The activities that have a primary impact on the sustainability of research software are described in Community & outreach (Chapter 4), Education & training (Chapter 5), and Incubator (Chapter 6). Additionally, some activities in Policy (Chapter 7) have a secondary impact on the sustainability of research software.
For people, URSSI aims to improve the careers of those who develop and maintain research software by:
In-person events and community calls
A summer school, and the incubator program
Promoting new career paths for those who develop and maintain research software
Developing and advocating for research software usage and impact metrics to be a factor in the hiring and promotion of software developers and maintainers, including promoting the publication of research software
Encouraging a diverse set of participants to enter the research software development and maintenance field and decreasing structural and systemic barriers to productive careers for members of underrepresented groups
The activities that have a primary impact on people and careers are described in the four following chapters, on Community & outreach (Chapter 4), Education & training (Chapter 5), Incubator (Chapter 6), and Policy (Chapter 7).
For the research software ecosystem, URSSI aims to improve the understanding and functioning of the ecosystem by
In-person events and community calls
A newsletter and and an active social media presence
Incentivizing contributions to public software
Disentangling software quality and software impact, including promoting software credit and citation mechanisms
Documenting the use of research software in research and providing systematic and regular analysis of the impact this software has on research
Promoting an increased understanding of the importance of research software in research
Advocating for increased funding opportunities for software maintenance
Creating an award program to recognize software contributions
Growing and participating in communities around the field of research software
The activities that have a primary impact on the research software ecosystem are described in Community & outreach (Chapter 4) and Policy (Chapter 7). Additionally, some activities in Education & training (Chapter 5) have a secondary impact on the research software ecosystem.