Research Software Engineering in the Arts and Humanities
Community Interest Group Survey
Number of participants: 24
1. Average completion time
3 hours 18 minutes
Survey completion time (in minutes)
2. Where is your institution based?
3. Please select your organization type
4. What is your official job title?
5. What is the nature / contract type of your current employment?
6. Please select the discipline in which you work. Please select all that apply.
7. What is the highest level of education you have attained?
8. In which discipline is your highest academic qualification?
9. Do you consider yourself a professional Research Software Engineer (RSE)?
10. How many years of software development experience do you have?
11. On average, how much of your time is spent developing software?
12. On average, how much of your time is spent on research?
13. On average, how much of your time is spent on management and/or administration?
14. On average, how much of your time is spent on teaching/training?
15. What are the barriers for someone to choose a job as an Arts/Humanities Research Software Engineer (AH RSE)? Please rank each of the following items in order of importance with #1(top) being the most important
1. unrecognised ah rse activity
2. lack of long term career prospects
3. lack of clear career paths
4. lack of clear promotion criteria
5. lack of clear salary benchmarking
6. not integrated into/contributing to uk rse initiatives and communities
16. Do you always work with the same researchers, or do you regularly change the researchers you work with?
17. Are you part of a dedicated research group within your institution?
18. Describe your work pattern:
19. Are you normally working on a single or multiple projects?
20. Are you currently working or have you worked in industry projects as well?
21. Have you participated in any pan-institutional RSE collaboration or activity?
22. Are you a member of an association of Research Software Engineers, or more generally research technical professionals?
23. Would you be interested in joining such an organisation?
24. What would you hope to get out of such an organisation? Please rank each of the following items in order of importance with #1(top) being the most important object to #3(last) being the least important
1. collaboration frameworks
2. policy interventions
3. professional rights
25. In general, when your software contributes to a paper, are you acknowledged in that paper?
26. Have you contributed to research articles as the primary author?
27. Have you contributed to other types of research outputs (reports, prototypes)?
28. Do you feel that your contribution to research is recognised by your supervisor/line manager/ the researchers you work with/ your institution?
29. Have you presented your software work at a conference or workshop?
30. How many software developers typically work on your projects/team?
31. Describe briefly your team structure
Pool of RSEs available for projects in any discipline. Individual projects get one, maybe two, RSEs doing development work and one Senior RSE overseeing as project manager (~2 hours a week).
Small Team linked to General IT Department
I often work alone or with a single domain expert in Classics.
Usually me alone. Occasionally collaboration with a contract developer or externally commercial agency.
Innovation lead with 1 PM who serves as engineering lead and 1 to to 4 developers
One Director, Principals that face each of the four institutional challenge area, Leads under them, that face each mission in each challenge area, then more fluid staff of Seniors and without prefix RSEs. We are moving to better align reporting structures to this structure, presently its matrix. In the same team, and we treat everyone on a continuum are Research Data Scientists
Divided into lab support (Director, Deputy Director, Lab manager Project manager, Principal RSE) and solution development team (RS Analysts, UX Designers and Systems manager).
Almost flat laboratory structure
Team is made up of developers, analysts, UI/UX designers, project and lab management and sysAdmin staff.
Research Software Engineers; Research Software Analysts; Research Software UI/UX Designers; Research Software Systems Manager; Research Software Project Manager; Research Software Lab Manager
A typical team would be the Primary Investigator, me as the technical manager/lead developer, a couple of postDocs and a couple of technical hires.
My main role is as digital skills manager but sometimes I perform RSE tasks to help researchers with their projects (mostly data wrangling/collection/analysis)
We have a small technical team of three developers and a systems administrator. We are based within the Research IT group of a University, which provides IT services to researchers. We do not primarily contribute to the work of the Research IT group, however, as we are hired as part of a consortium delivering a National Digital Repository for Humanities, Social Sciences and Cultural Heritage data. We participate in various research projects in addition to our core work of developing and maintaining the Repository infrastructure.
Emerging
I directly reported to th Digital Humanities manager. We both worked on multiple research projects developing web based applications to present, create, update and maintain research data.
Very flat (1 manager and 10 developers all at same level), no defined roles or responsibilities, impossible to scale.
RSE works alone but with students.
none
I answer to the project PI.
We all take projects depending on our skillset wen they come in. Mostly everyone works alone on their projects unless a project requires multiple skillsets that may be spread out in the team.
2 RSEs with an academic supervisor.
One of presently two RSEs working on multiple projects at multiple institutions based within a research institute at one institution. Line managed by the Director of the institute.
32. Do your research software projects typically include a (technical ) project manager?
33. What is the bus factor of your most important software project? (The bus factor designates the minimal number of developers that have to be hit by a bus (or quit) before a project is incapacitated)
34. How often do you licensing software you produce under an open-source license?
35. How often do you associate your software with a Digital Object Identifier (DOI)?
36. Do you have an ORCID ID?
37. How is your current research software work funded?
38. Are you willing and/or able to act as PI on grants?
39. Are you supported in your funding applications by your institution (policy-wise/research office)?
40. How do you assess existing funding provision in the arts and humanities for the following areas: technical assessment of funded projects and research outputs
41. How do you assess existing funding provision in the arts and humanities for the following areas:
sustainability of funded projects and research outputs
42. How do you assess existing funding provision in the arts and humanities for the following areas:
infrastructure provision, including access to HPC / scalable compute
43. How do you assess existing funding provision in the arts and humanities for the following areas: devoted AH RSE work (pool)
44. How do you assess existing funding provision in the arts and humanities for the following areas: RSE compute capability
45. How do you assess existing funding provision in the arts and humanities for the following areas: maintenance of technical infrastructure (data repositories, services, systems)
46. What do you consider as a RSE AH community training priority (choose all that apply)
47. What do you consider as RSE AH community Policy priorities
48. Can we add your email address to a list run by the UK-IE CIG for AH RSEs? We'd like to share information on community events, learning and networking opportunities (optional)
49. Any other comments?
Question 15 is a "do you A or B" qustion but the answers are "yes/no". I said yes because I meant A
RSE communities provide a useful networking medium. But in AH at institutional level the number of actual working RSE's is usually so small a pool it's effectively moot. There are several underlying issues. Technology development and implementation in AH is typically at a very modest level.
The cause is both prevalent traditionalism but also the meagre state of funding awards for AH/digital projects. It's impossible to build anything to scale. Projects that were possible 20 years ago with the AHDS/JISC and other funding councils are pipe dreams in the current environment.
At the risk of being overly negative, an AH RSE community is likely to reflect these issues. It will be small and unambitious.
In Ireland, it would be beneficial to universities to allow Digital Humanities departments more flexibility in how they hire RSE developers. Wages are low for the kind of experience and talent they ideally want to hire. Red tape in the hiring process makes getting and keeping a RSE role risky. Benefits are practically non existent compared to industry standards (healthcare etc).
Also, I, and possibly other self employed devs and agencies would, be able to offer services to the departments on a flexible, as hoc basis if contracting policies were not blocking this from being possible. I think many projects would only need a developer or agency for a free weeks or months during key parts of the research project, rather than fixed term contracts. However this is not possible in Ireland due to university contracting policies.
Finally, dev ops need to be embarrassed by universities. Having a rigid IT structure makes it difficult or impossible to setup CI/CD workflows to make software infrastructure easier to deploy and maintain long term. Again, this is where policy for 99% of university departments is massively holding back Digital Research and Digital Humanities projects. Perhaps this would work best if there was a dedicated RSE department that works with Digital Humanities (and other departments) who specializes in delivering software projects for universities?
As a member of a RSE Group that is actively trying to engage with AH Groups, then the biggest barrier is just communication for a couple of reasons:
1. Software Engineers typically come from STEM disciplines and don't understand AH
2. AH personnel typically have limited understanding of STEM including Software Engineering
3. It is difficult for us to engage with AH Groups and find projects in the AH Domain. Project that do arise are typically simple web apps, simple databases or somebody doing "AI" whatever that is. However, I'm convinced there are some really great potential projects out there is we could join the dots.
Anything which helps to resolve the first 2 items will probably significantly help towards item 3.
Using the ERROS Connectionist Database (CDb) which I conceived, created and patented, you can create applications incrementally, simply by entering data and application definition into the ERROS Ontology which is part of and also defines the CDb (i.e. the CDb is defined by itself), Major, complex, multi-user, multiple institution, collaborative, high performance applications that work over the internet and that can handle billions of records can be created without any new program ocde. It is particularly suited for AH and GLAM and is well tested. It really does all that I claim.