Episode 267: Michelle Barker on the Research Software Alliance  ## Metadata - Author: [[Sustain]] - Full Title: Episode 267: Michelle Barker on the Research Software Alliance - Category: #podcasts - URL: https://share.snipd.com/episode/45de670e-89b3-4317-a02f-652107bed5a4 ## Document Notes ## Highlights - **Michelle's Open Science Journey** - Michelle Barker shared her journey from sociologist to open science advocate. - She integrated internet technology in education early, pioneering online research assignments in 2000. Transcript: Richard Littauer Open source and open science? Michelle Barker You've also promoted me. I should correct that I was one of the directors of Australian Research Data Commons, not the sole head honcho. How did I get involved in open science and open source software? So I'm a sociologist originally. That's what my academic calling was. And when I started working in that field as a lecturer in about 2000, I noticed that there was this thing around called the internet and technology seemed to be taking off as a way to enhance Collaboration and a tool that could be used in education and research and learning. So as a sociologist, my key areas are interest. How do you get large scale cultural change? And also how do you get organizational change? And so I started focusing on projects that looked a lot at how to use this newfangled technology like the internet to enable that. So for example, in 2000, when I was actually teaching in New Zealand, I ran the first ever course at the Victoria University of Wellington that required students to access the internet To do some of their research for the course each week. And a few of the students complained and there was an article in the local student newsletter because was unheard of, that it was a requirement of the course that they access the internet, Even though we'd arranged free access through the library, et cetera. And then over the course of my career, I just stayed interested in technology advancements and how to use them to get people talking to each other and how they're used in research. And over time developed a specialization in using those tools. And then when that became open science, I was a specialist in that. And particularly when I was at the Australian Research Data Commons, I was leading national funding for research software infrastructure. So that's when my specialization in research software really evolved. ([Time 0:02:00](https://share.snipd.com/snip/bfc61801-1c9b-45c4-aee5-268a24deeb15)) - **Need for Research Software Alliance** - Research software lacked international coordination unlike open data efforts. - This spurred the creation of Research Software Alliance to unify global community efforts. Transcript: Richard Littauer Why not wiki data? Why not something else? Michelle Barker When we started thinking about forming this, and Neil Chuhong was one of the co-founders with myself and Dan Katz from the US. It was an area that we shared a passion for and could see that there was no international coordination on. So at that time, there were good organizations doing coordination of research data, like the Research Data Alliance and CoData, part of the International Science Council. But there was no equivalent for research software or for any type of software used in research. But we could see that there were lots of fantastic national efforts and disciplinary efforts and thematic efforts like in training or around career paths, but they never came together To talk about how they could influence international conversations. So soon after that conversations like the formation of the UNESCO recommendation on open science happened and there was no way for the research software community to be represented Except through peripherally through open data representatives or through people like Neil and Dan and I talking to other people who are on the committee. So we wanted to create an opportunity for the community to come together and be represented internationally but also then for the community to work together internally. So where all those fantastic national disciplinary and thematic efforts existed, that we could see there was some siloization or duplication. So we wanted to ensure that collaboration started to form where they should. At that time in research software, I often say that if you wanted to know who else was working on something in your area, beyond your immediate circle of friends, you had to lurk in the Right corridors at the right conferences to access other networks, to find out where someone else in another country or another discipline or some other area was solving a similar Problem. That's not very democratic or accessible. We really wanted to take a lot of that knowledge out and put it in the open so that information flow could increase and people could benefit from learnings of what everyone else was doing And leverage those and then identify places where people were dealing ([Time 0:04:55](https://share.snipd.com/snip/6da23893-d37c-4b14-91d7-d053d0671bcc)) - **ReSA Engagement Strategies** - ReSA uses newsletters, databases, stakeholder forums, and task forces to foster collaboration. - Engaging diverse funders and communities enables effective knowledge sharing and problem solving. Transcript: Richard Littauer Like? Michelle Barker RISA GOLUBOFF, Yeah, we have an engagement plan which has three major pathways. The first is that we enable exchange of information to increase knowledge flow in the community. So we do a lot of standard things like a newsletter and social media. We have a few databases that we maintain such as what's available in funding specifically for research software or policies that support research software in research organizations. Then our second pillar is probably where we do the most of our work, where we aim to engage community collaborations to solve common issues. So we do that primarily through stakeholder forum. So for example, for three years, we've been running the research software funders forum, which over 60 different funding organisations have engaged with from national funders, Philanthropic funders, industry funders, and they come together to share best practices and network and identify where they could work collectively to solve common problems. We also have another forum on research software infrastructure providers. We support the International Council of RSE Associations, where the National Research Software Engineering or RSE Associ come together as well. So we enable those kinds of conversations across common stakeholder groups so they can think about where they might want to work together collaboratively. And then the second major area is what we call task forces, but are working groups where members of the community identify a problem. We help connect them with other people internationally with the same problem. They form a task force and we help promote the existence of that task force, find the members, promote their engagement with their community and whatever outputs they come up with. And some of the task forces we've had, for example, wrote the Fair for Research Software Principles two and a half years ago as a major output. And ([Time 0:07:26](https://share.snipd.com/snip/a24d9771-aaef-4531-9564-7015df267df8)) - **Defining Research Software** - Research software is defined strictly as software created within research context solving research problems. - Not all research software is open source; openness varies like data policies. Transcript: Richard Littauer The same as open source. Michelle Barker It's not exactly the same, but it's heavily overlapping. One of the interesting things about research software, which we define as software that's being developed within the research context to solve a research problem, and research software Is not the same as software used in research, which could include Microsoft or all sorts of other things that aren't specific to the research community. So one of the interesting things about research software is we don't actually know how much of it's open source. The majority, probably, we would guess anecdotally, but there's no data on that. And there's certainly some research software which is not open for very valid reasons. In the same way that research data is as open as possible and as closed as necessary, we can see this variation in research software. So most research software is open source software just ([Time 0:10:28](https://share.snipd.com/snip/a067c1a0-2b31-4419-a209-0fad41123e7d)) - **Challenges Measuring Social Infrastructure** - Measuring effectiveness of social infrastructures like ReSA is challenging. - Metrics like newsletter subscribers and forum participation exist but do not fully capture impact. Transcript: Michelle Barker It's a really good question. And I think it's still a common issue for many social infrastructures, which is what we consider ourselves, to be able to track engagement. So there's certainly metrics like the number of newsletter subscribers you have, or the number of times that a position paper is downloaded or cited, or the number of people that you've Engaged do a working group, the number of people that come to community consultations, or we just recently had a community consultation on our draft strategic plan for etc. I'm not sure that we still really got to the heart of how to measure the value of social infrastructures like the Research Software Alliance. So always ([Time 0:13:16](https://share.snipd.com/snip/74445ebc-6129-4917-9e84-2b248259a44a)) - **Social Infrastructure Defined** - Social infrastructure involves facilitating human interaction and community building rather than just technical systems. - Community managers play vital roles supporting collaboration, knowledge exchange, and belonging. Transcript: Richard Littauer We've heard digital infrastructure, right? I used to be the grant manager whatever you want to call it community manager for the digital infrastructure fund the dif is now the digital infrastructure insights fund diif i think Largely because they couldn't get the domain name and i was out hiking for a month in vermont but you know i'm sorry about that but it worked out and digital infrastructure is really easy To grasp hey everyone uses the internet without the internet you can't get things delivered from amazon and therefore we're all going to die right no we all have to have digital options There social infrastructures is a bit different and as a sociologist i can see why you would put that perspective on it hey we're out here providing the infrastructure for people to Connect with each other which sounds a lot like to me a lot of the work happening in open source. For instance, you could say the OSI is a social infrastructure group, because they just connect lawyers and people who are confused about licenses to talk to each other. And by connect, I mean, they have a website where you can talk about these things and look at them. Does that sound accurate to you? Is my understanding of that term sounds correct? Michelle Barker I think so. You're making me realize I don't actually have a set definition for social infrastructure ready to reel off, but certainly there's large parts of the community that utilize it. And I think there's been a growth in the last five to 10 years in understanding that a key part of... Of course, there has been a growth in the last five to 10 years in understanding that a key part of software development is engaging the community and thinking about the maintainers And the volunteers and the governance structures and all those things that involve people. So how you get people engaging with people, I guess, is to the heart of social infrastructure. We've also seen the rise of positions like community managers, particularly in the research community, but I think more broadly in recognition that they have a role in enabling the Social elements, enabling people to come together and do knowledge exchange and share best practices and network and find support and find a feeling of belonging and enable better Collaborations. So I guess that's what social infrastructure is. ([Time 0:14:09](https://share.snipd.com/snip/d0714461-ce2a-4790-b644-f9988cfbd977)) - **Effective Networking Techniques** - Build personal connections by highlighting shared experiences or interests when introducing people. - Research attendees beforehand to prioritize meaningful conversations and foster engagement. Transcript: Richard Littauer People realize that you actually have something that maybe they could learn or maybe you have a connection that they could learn from? Michelle Barker It's a skill set that I didn't realize not everyone had because it was something that is quite key to my personality. I am a social connector. So for example, if I'm introducing two people, I'll never say A, me, B, I'll tell them what their connection is. Hey, you both went to the same high school or you both have a passion for X and give them a way to connect. And I didn't ever understand that that was not normal until once I was talking to a data colleague about how I prepare to go to a conference. That I will go through the list of attendees if it's available and speakers and think about who I really like to prioritize chatting with. And I'll put them in groups. They'll be ones that it's going to be difficult to get their attention and so I might have to time it just right after their talk when they're leaving the auditorium to get a minute with Them to try and get their attention. And then there's people that I, a second group of people who happily talk to me and I can just write to them and ask, can we do lunch together? And then I always leave time in my schedule for kind of serendipitous outcomes to just talk to people at breaks because you never know who's going to actually be a really important person To meet. And my colleague was astounded by this. He said, he just turns up and talks to whoever happens to be available. He said, there's an algorithm for social connection. And I was astounded by his way of describing it because I'd never thought about that. And it's like, yes, there's a methodology for figuring out how to talk to people. And yes, if I'm really wanting to speak to someone that I don't know well, I might have researched their background so I can start the conversation rolling. You build an emotional connection or a personal connection or figure it out what I can say about me that's going to spark just enough interest that then when I email them a few days later, They'll read the email, for example. Sorry, let me just tie that back to your original question. For example, our newsletters have a broad range of things in them that might be of interest to stakeholders ranging from a national policy maker to a lead of a national research software Engineering association. And we'll put some of the context there, you know, how is this relevant to you? Why should you be involved? So I guess, you know, at the broader level, at organizational level, we're starting new forums or task forces. It comes back to that technique of encouraging people ([Time 0:19:56](https://share.snipd.com/snip/15529188-99b5-45a3-aad5-84f5d598c7b3)) - **Research Software's Unique Position** - Research software sits at the intersection of research community and open source software community. - It faces challenges stemming from both worlds simultaneously. Transcript: Richard Littauer Venn diagrams? Like, what do you do to think about this is where we want to be? And how do you decide which people are in that group? Michelle Barker I really like your framing about it. We're trying to define a field because of course there are other endeavours that do that. For example, a German based group, but IRSER looking at research on research software. So one of the early meta-science groups, but I guess they're defining the field in some ways as well. We have nice diagrams about all of the different stakeholders, but we don't have flowcharts or hierarchies. We just have lists of these are all the stakeholders that need to be involved. We have one Venn diagram that shows that research software is in the middle of the research community and the open source software community, they're both important in the broader Aspects. But yeah, that's as complicated as our diagrams ([Time 0:27:45](https://share.snipd.com/snip/47b2e511-f296-443d-8d98-301656811697)) - **Sustaining Research Software** - Sustaining research software means showcasing successes to demonstrate its critical value. - Support for personnel and infrastructure boosts research outputs and community resilience. Transcript: Richard Littauer Structural problems that you need to solve. And so what does it mean to you to sustain research software? How do you think about that question? Michelle Barker Well, there's a lot of challenges that exist. There's also amazing benefits. So a lot of the approach is to spotlight the successes that we see, especially where countries or organizations or disciplines have supported research software and its personnel And are then able to demonstrate the benefits in terms of increased research outputs. It seems like a very simple answer. Is that enough? Richard Littauer No, simple answers are the best. There are a lot of benefits. In fact, I was trying to end this on a high note because we are running up on time. So thank you so much. ([Time 0:29:51](https://share.snipd.com/snip/272c3dde-693c-49c9-b54a-eb547db8a80c)) - **Michelle's Spotlight on Research Software** - Michelle highlighted Softcite, a project analyzing software citations in research papers. - She also spotlighted grassroots research software engineering associations worldwide nurturing community and recognition. Transcript: Richard Littauer Today? Michelle Barker Michelle Pfeiffer 1 I'd like to highlight a very small project called SoftSite, which is a project studies how researchers use, cite and collaborate around software. And they produce data set, some software and some papers. There's a new paper just out that shares software mention extractions from 24 million open access papers. And the reason I love their work is because it's so critical for us to be able to demonstrate the impact of research software and its personnel. And as we've already discussed at length, it's very difficult to do that, partly because the software citation pipeline doesn't exist very well at the moment. So these kind of efforts, and this is just an example, there's others I could have highlighted are incredibly valuable. And I encourage people to go and mine that data set and use it to show the impact of research software and possibly even their own software if it's mentioned in that. I would also add that when I came to this question, I almost suggested that for me it be reworded to what's a community I would like to shine a spotlight on coming back to the social infrastructure Field. And I would have highlighted all of the national and regional and local research software engineering associations around the world, which is completely a social infrastructure In that it's usually run by volunteers. Not always. Sometimes people are in paid roles within institutions, enabling sharing of networks and best practices and support. And it's been running for 10 plus years now and has had a fantastic impact on recognition, understanding of research software. And it's just an amazing exemplar of what a community can achieve together. Led from grassroots up and just fantastic. Richard Littauer That's a very ([Time 0:31:49](https://share.snipd.com/snip/6152f30e-e0d0-4469-ad20-866772e6eb18))