Research Software Engineers:

Where it started and where it’s going

Simon Hettrick

October 2024 - US RSE 2024 - @sjh5000 ORCID: 0000-0002-6809-5195

Abstract: The concept for Research Software Engineers came from a discussion at a workshop. In only 12 years, it has grown into a career path and a self-sustaining movement that has spanned the globe. In this talk, I will look at the history of the RSE campaign and explain why I think the campaign was successful. I will move onto what excites me most about today’s RSE community: the way in which it is helping to change research culture from focussing not just on collecting results, but on collecting reliable and reproducible results. The impetus for this change comes from RSE’s contributions to the arguments around how we define “success” in research. RSE has had a huge, positive impact on this debate by proving that if we don’t recognise everyone who is vital to research, we limit our ability to conduct research. I’ll end with some thoughts on tomorrow’s research and the inevitable and hugely welcome growth of non-traditional research roles.

Could plot the number of history of RSE talks I've done - gives you an idea in interest in RSE
A bit about what it means to leave something you created
Why did it work? It was partly "Nothing is as powerful as..." which is great. But the really surprising thing is that it didn't happen earlier. I mean, there's a ton of software out there. If the planet was submerged in water, you'd expect people who provide boats to become pretty popular. But in this world our research institutions were all "Nope. I'm fine thank you" as they disappeared under the surface. It was weird. The reason is tied to what we see as success in research...
Talk a bit about what it means in reality to try and get a campaign like this working
- My top three tips on getting things changed
  - Stubbornness and stamina rather than astute evidence collection and clever arguments

This is about the past, present and future of RSE

I had a problem. Do something to make software important. Do policy, but with absolutely zero experience in academic policy.
The campaign came together because I had to choose what to spend my time on. What was going to change things?
I thought back to the Collaborations Workshop that had taken place that year.

Reproducibility

A training community

RSEs

Can't remember what the fourth one was

Something about volunteers and how hard it is to get something sustainable

The future is about getting an understanding of the value of RSEs. That's the thing. We need to demonstrate the value. There's ROI, of course. Rif a bit on how reliable those things are to the money people. Need to show the ways that RSEs are contributing. What would have happened without them:? Just as importantly, we need to ensure that universities use metrics that represent our work.
This is where you go "In the UK, there's this thing called the REF. Boring right? It's responsible for £X billions and everyone hates it."

VALUE is the thing that this talk will be about. RSEs have always been valuable. The future is about demonstrating that in a way that the sector understands.

"It's a great idea and all, but nothing will change."

When talking about the reliance on papers in academia

Point out that the people in the group that coined the name "RSE" wrote a paper to sort the problem out.

Could give some hints on working in policy

not always the most intelligent, but you can often be the nicest

Try and say the thing that people want to say but are restricted from doing so by a fear of hitting things head on