 I'm very happy to welcome Dr. Heidi Serbel to this keynote. Today's sponsor of the day is Roche. Thank you very much for your support for the conference. And yeah, we are very happy to have Heidi as a keynote speaker. Heidi is a biostatistician. She studied in Munich and then came to Zurich for her PhD. She co-founded the Zurich R user group that is kind of at the origin of this year's edition of the youth R conference. She's an open science and reproducibility advocate. And she has recently been awarded the newcomer of the year MAI award from the German Informatics Society. She will tell you a bit more about her career in the keynote, so I will not tell you everything now. But she's now working at the Jörner Institute where she's working on advancing methods in teaching and research. And I will not talk any further, but leave the floor to you, Heidi. Thank you very much. Thank you so much, Dorothea. Do you want to add something? If you have questions during the keynote, please ask them in the Q&A. And we will happily get to them at the end of the keynote. Thank you very much. Perfect. Thank you so much. I'm going to start sharing my slides now. Someone just wanted to come in. Sorry. Yeah, so thanks so much for having me here. It's such a great honor to be speaking at this year's youth R. So my first youth R was in 2015 in Alborg in Denmark. And I looked at the keynote speakers there and it was like, wow, what amazing people. And it's incredible to be a part of such an esteemed group now. So thank you so much for inviting me to give this keynote. I'm going to talk today about research software engineers and what they do, what it means to be a research software engineer, what the troubles are, and all of this in the context of academia. You can follow me on Twitter. My Twitter handle is at Heidi Bayer. And as Dorothea said, I work now for the Yuna Institute, which is a company that deals with medical devices and certification of medical devices. And we're doing a lot of education and research now. And I'm also an independent researcher with ICDOR. So if you're an unconventional researcher, ICDOR might be a good home for you. So you should check it out. So my slides and everything are licensed under a CC by license. So you're welcome to reuse any of the things that I have in my slides, including my drawings. So let's see. Yeah, so as Dorothea mentioned, I'm going to talk a little bit about my own path before I dive into the topic because I think it tells a little bit about where I came from and why I'm interested in this topic and why I'm so passionate about these things that I'm talking about today. So I did my bachelor's and master's in Munich in statistics. So I actually learned R in my very first year of university and fell in love with it directly. My bachelor thesis already was an R package. So if you're interested in analyzing unexploded bombs, then the high-risk package is for you. Then after that, I went to Zurich. I did my PhD there under the supervision of Thorsten Hothon and Achim Zaylas, which many of you might know as one of the big people in R, one of the important people in R. They really gave me everything to pursue my passion in R in software and in open science. After that, I went to do a postdoc back in Munich in a medical informatics project. Then I got the chance to do a professorship for half a year. So in Germany, we have this thing where when a professor leaves, then they sometimes allow younger researchers to play a professor for half a year. So it was Professor Heidi for half a year then. And afterwards, I started two postdoc positions at the same time at LMU and at the Helmut Center in Munich, where I was already kind of a bit of an unconventional postdoc working mostly on open science and reproducible research with a focus on AI and medicine. And then I applied for a group leadership position at Helmut's AI, which I was very lucky to get. And I really thought that this was going to be my dream job. And then it turned out that it kind of wasn't my dream job, because I wasn't a real researcher anymore at the time. And I just figured I need to change something. And I need to tell people that, yeah, this is not my kind of not my identity anymore. And I'm going to talk a little bit more about that in a minute. And now I'm actually more on like the meta level. So I'm thinking a lot about how we can do science and education better and try to implement that in. Yeah, a kind of new way and with new ideas in a new setting as well. So this is where I'm at now and through all these things through all these many jobs that I had already in my short career. One or two things were always with me along the way and these were software and open science. So as I said, I already started using our during my bachelors. I got into open science because my first real research project during my masters turned out to be such a big mess. I had files flying out all over the place on my computer and I thought I need to do this better I need to become a better researcher. And that's how I got really into reproducible research and through that path I learned about this whole open science movement and the community behind it. And yeah I stuck with it since then. Why am I giving this talk. It's a little bit born out of this idea that I kind of was in paradise. And I had the dream job so I scored a group leadership position. I was really in a great place and I was still pretty I mean I'm still pretty young and being a group leader at that age was really a success and everyone thought oh Heidi's going to become a professor at some point. And so, at the same time I felt like really struggling and really like I was in this paradise, but somehow something was not right so this is here depicted by the snake strangling me. And I tried to figure out what it is and somehow it was that I wasn't fitting into this clear picture of what a researcher needs to be and what a researcher needs to do because I was doing a lot of things that wasn't exactly writing papers. Now, what does it have to do with RSE. First of all, let me talk a little bit what research software engineers actually are and what it means to be a research software engineer, because maybe in this community. The term is already a little bit known, but like in general people don't really know what a research software engineer or RSE actually is. So what do RSEs do. RSEs obviously develop software but they develop software that's specific to research so they need to not only be able to. Yeah, right software but they also really need to have an intricate understanding of the research field that they're working in. They generally code so that can also be analysis scripts and stuff like this. Sometimes they teach about software to researchers, they support research projects in terms of software. Yeah, they consult researchers with any kinds of software problems or I mean there's also specialized RSEs who work on very specific problems and help a lot of researchers with that. So let me go into a couple of personas to give you a bit more of a deep understanding what research software engineers are and what types of research software engineers are out there. Well let's think of one person she's a PhD student. She writes software as part of her research project, but her supervisor tells her hey you have to have like three papers at the end of your PhD or otherwise you're not going to get your PhD. So she would actually like to code more but she needs to think about her career and write papers. And then there's this guy and I assume that every institution has some a person like this, so he's the person that everyone goes to in case of problems so imagine you have an issue you don't know how to implement something and are, and he's the person that you can go to any time he's super kind and always willing to help, although of course it's not his job to do that. Then there's the researcher she works in a lab she needs software and other software skills to really do her research in the best way possible and she learns what she needs to do so but she doesn't do much more than that. There is a reproducibility guru he like is really into reproducible research and learns all about the software tools that he needs in order to make his work really really reproducible and also maybe helps other people to do so as well. And then there's the classical software person she's a trained software engineer, she's hired to work on software for a research project so imagine maybe an app for questionnaire or something like this. Eventually she decides to leave academia because companies value her skills more, and she can actually have a career and a great salary as well. And then there's me. I really liked research I really do, but I love, I really love and enjoy thinking about open science and reproducibility I love coding in our. And actually, as you see today, I get invited to talk about open science about reproducibility about research software engineering, and that much more than about my actual research so just FYI. My research has been on model based trees and random forests. So if you're interested, check out the party kit package. And, yeah, I, I use that for the context of personalized medicine and personalized treatment effect estimation, but people weren't as interested in my actual research then they were interested in the things I did on the side, the science work. So, I actually just left the classical academic career in order to be able to do these things that I enjoy doing so much. And it's kind of weird to go to a company in order to be able to pursue the passion of promoting good research practices. So, what I want to leave you with, and this is really important software and RSEs are everywhere in research nowadays. So there's almost no research project these days that has. Yeah, nothing to do with research software so let's think about biological research for example, software everywhere. Digital humanities is a field that's by definition digital and need software medical research psychological research or even one of my favorite projects that has been going on in recent years, the mosaic expedition where they sent a ship into the ice and drifted through the ice for earlier. There's a group, a team of RSEs in the background, just figuring out how all this data can be managed in the end. So this is really something where we see very clearly how research software engineers are so much needed for these kinds of projects that really, yeah, that are really visionary and think about how will this earth be in a few years and what is happening to an environment currently. So all the big questions that we're doing in research right now are dependent on research. And of course in research, but also in research software. So, I want to give you a moment to think about maybe whether you're an RSE and I think, actually, most of the people listening right now are, in fact, research software engineers because you're here because you're our users. Because you're somewhat interested in research, I guess. And so I think most of the people who are listening right now are, in fact, research software engineers. And it's important for me that you think about that because if you identify as a research software engineer, then we'll manage to figure out all the problems that I'm going to talk about later in this talk. And now I want to do a little bit of an excursion and talk about my favorite topic which is open and reproducible research. And what that actually has to do with research software engineering, because I strongly think that these two things are super closely connected. And we cannot think good quality research nowadays without thinking about good quality software. And I do that a lot if I talk about open science. There's a couple of things that I usually say. So, for example, I tell people to make their code available and reusable for others. I tell people to publish preprints. I tell people to use version control. And in general, I tell people, well, why don't you just make everything available for everyone. And I am aware that this is not easy. So I'm always trying to be proactive and show also how to actually do all these things. And I also, I mean, I tell people that they don't have to do it all at once and of course they don't have to do it. But this is would be the perfect world if everyone would try to work openly. So, if I'm talking about these things and then of course I show them which tools I personally use, and which I would recommend and this is one of the slides that I sometimes show and we were not going to go into every detail here but I just wanted to show you that there's a lot of different tools that I'm using to make my work reproducible. There's a lot of, especially technical things that I am using in order to make my work reproducible. And this is why I'm showing you the full slide with all these words and not to like go into the depth of everything but to make you aware how technical reproducibility and, and, yeah, computational reproducibility at least really is. So, I talk about things like version control. I talk about things like virtual machines or Docker. I talk about make files. I talk about raw oxygen things like those and they're all like highly technical and what we get in the end are researchers that look like that. So, they're trying to do a good job, and they're trying to implement these things that I'm recommending. But let's be honest, not every researcher has the time and the background to learn all of these things and I personally think not everybody has to know everything and be able to do everything. So, yeah, the main point here being open science and reproducible research require tremendous software skills. Yeah, and somehow I think we need to figure out the way how we can use these modern reproducibility tools without getting completely frustrated right so we don't want to see researchers like that in this GIF better just like giving up saying I don't want to do this anymore. Let's just quit, quit it all entirely. So, what do we need? I think we need to work together we need to have teams of people who are able to do different things. And so I think in terms of software and reproducibility we need people who develop helpful tools. We need people who help research and finding and learning these good tools we need people who support researchers with computational reproducibility and support them in writing software. So if you're surprised, surprised, these people are called research software engineers. So I think research software engineers could help as parts of teams of research teams, but also as yeah teams of consultants that go into different research teams and help there. And this right now already in statistics so we have statistical consulting at universities where people can go to and ask for help with statistics. We have statisticians as part of medical research teams for example, and who do all or who help with with the statistical work. Why can't we do the same thing with software as well, because this seems to work. And we're already now seeing that, at least in the UK or in places where universities and research institutions are implementing this. So it really seems to work and I think that this is the way forward for for research is to allow for different kinds of people to be part of a research project. Let's get to the tricky part of us because, of course, the UK. There we see that a lot of this is already being implemented and research software engineers can have a career, although it's only getting started there as well. But in most places what we see is that research software engineers are hired on short term contracts. Sometimes they're just PhD students doing software on the side, they get a shitty salary, they have no job security. And if they really want to focus on research software and not focus on writing papers all the time, then they also have no career path. And at the same time, these same people are not only highly skilled at software engineering, they're all so highly highly skilled in the specific research field that they're working in. So they're some of the most skilled people that we have in research. And of course, since software is so much needed in research, these people are very much needed for research project. So we have these two sides of shitty salaries, no security, no career path. And at the same time, they're really so much needed and should be valued so much more. And so I think one of the big questions that come with this is why are career paths and in particular academic career path so so stiff. So because right now you're having a career if you're good at writing papers, lots of papers if you're good at publishing in high impact journals, then you're well off, then you'll have a career, then you'll be professor at some point, and you'll get a permanent position. But what if you're interested in other things? What if you want to do a great job and promote research by writing good software? Of course you can write software papers as well but really should this be the path we have to go if couldn't research software by itself be a valuable output. And I think this is one of the major points that we have to address if we want to allow for careers for research software engineers and essentially also many other people who are super important for research. Let's think about data managers. Let's think about scientific journalists. Let's think about scientific managers. There's all kinds of people who are super, super important for academia for the scientific endeavor, but they're not as valued as people who just write papers all the time. And I'm not saying that writing papers is easier. Writing papers is not something that we want, right? Of course this is also one possible research output, but we need to think a bit more about like allowing for a different range of research outputs. Couldn't a podcast be a research output. Couldn't software be a research output. Couldn't data be a research output that we value as much as we value papers. So, going back to this image is academia really a paradise for everyone. And I think it's definitely a paradise for people who enjoy the things that we value highly in academia. But it's not as much a paradise for people who don't enjoy it as much. And for people who want to do a bit other things, have other kinds of research outputs. And this I think is a problem because right now we're like, okay, here's this t-shirt, this researcher t-shirt, and put it on one size fits all. But it's not the case. Not everyone is the same. Not everyone enjoys or is even good at this one task that we give researchers, right? And I also really, really think that it shouldn't be this way because it's great that we have people who write software. I think it's great to have people who are good at designing great websites for the public. I think it's great that there's people who produce videos or other kinds of research outputs that communicate what you do other than just papers. And I think this is really important. And we shouldn't think in this one size fits all way. So what would my optimal world look like? And I am aware that this may sound naive at first, but I really think that if we think would think about this a bit more and if we would support people who really have good ideas. Then I think an optimal world would look like that everyone gets to work on something that they're passionate about. And I think this might not be possible like entirely and fully and for every part of your work and I think that never will be the case. But in some sense, I think this can be possible. Now the question is, of course, how do we get there? And yeah, I really think that the best path towards this goal is by growing and building communities around topics where we need the big change. And luckily for us, there is already an RSE community and luckily this community has already reached many, many goals, at least in parts of the world. So where does the community come from? So the term RSE was actually coined in the UK in March 2012. And from there on, the movement in the UK started. So we see that for almost 10 years, research software engineers in the UK have been pushing for the goal of having careers for research software engineers for making software first class scientists in science with the slogan better software, better research, which makes total sense to I think everyone who's listening right now, but which is important to say again and again to everyone. In Germany, that was to my knowledge, the second community that started on like a country basis. The community was actually founded at an RSE conference in the UK in 2016. The association was founded in 2018. And we actually had our first conference in 2019, which was like such a great success and it felt like a start of something big. And I think we're going to grow something really nice out of this. Because RSE community is popping up now all over the world. And I put a link in here. So there's Australia, New Zealand, there's Belgium, there's the Netherlands, there's the Nordic countries, there's the US, who already have RSE communities. And you can found one in your country, wherever you are. Because there's so many RSEs out there. And I think a lot of the people who are listening right now are now thinking, oh, I'm an RSE as well, what can I do and feel free to, I don't know, discuss this in the slack or whatever, and just get together and form local chapters and local communities. Because I think this is the starting point of how we can achieve something and change research for the better in terms of research software. And now I also want to talk a little bit about what the R community can do, because obviously the R community consists of many, many research software engineers and these two communities, the RSE and the R community are very much overlapping in terms of what they want and what the goals are and also the people as well. And so I think the R community and the RSE community should work hand in hand. And for this, we had an incubator session actually on Tuesday at USAR, just about this topic, what can our community do and how duties to communities fit together. And here are the things that we came up with or at least a short summary of this. There will also be a blog post about the incubator and what we came up with, but here just a short summary, what we think the R community could do. First of all, obviously, be part of the RSE movement. We could do things like building an R community within the RSE community, so get people together who both are users and RSEs. We could host events together. We could give RSE talks at our conferences, like I'm doing now, and vice versa. So we can just work together as community, as one movement. And of course, we should talk about it. We should create awareness. So I wrote down these three words, talk, teach, try. So talk about the term tell people that RSE is a thing. And yeah, really, every time I talk about this, someone comes to me afterwards and says, oh, I just noticed I'm an RSE too. And this is really a nice feeling to have. And teach about research software, teach about how to make research reproducible, how to implement good research software, good research software practices, and so on. And maybe try it. Try to become a research software engineer if that's something that you want to do, although right now it may be still pretty hot depending on where you are. Another thing is of course, lobbying. And this is what I think we need to do a lot now is talk to the people who are in charge and tell them, hey, this is kind of important software is such an important part of research nowadays. We need to do something to improve research, and not just always waste money by having one, for example, one PhD student work on a software project and then this one PhD student leaves and a software project dies. And then five years later comes a new PhD student who writes the same thing over again. Right, this is just a waste of money, a waste of resources, a waste of time for everyone. So we need to talk to the people in charge we need to talk to the funders, we need to talk to the heads of institutes to to politicians and tell them, hey, this is important. This could really change and improve the way we do science. And actually, I've talked to a couple of heads of institutions in the Hamilton's Association recently, and they are quite aware. And I was really surprised how aware they are of how important research software is. And I do think that we have a pretty good shot at changing the way things are here because people are already noticing and they want digitization and everything so I think there's some something we can do. If you're in this position that you can hire people, you can consider hiring someone. So actually, I gave a talk about this, a similar topic once and afterwards I got a message of someone who's told me, well, I heard your talk and now I'm hiring an RSE. And I was like, such a great feeling for me. So, if you do so, tell me because this is going to be a nice outcome of this talk, if someone does. And then, of course, we need to think about how we evaluate people in science. And yeah, obviously, it's easy to count the number of papers or the number of papers in whatever high impact journals or other stupid, stupid measures. But maybe we need to think about, like, we can't only do these super simple measures because people are going to try to have a career and they're going to try to follow these simple measures. And it's not necessarily the best for science. Maybe we need to think about people having a portfolio rather than just a list of papers, a portfolio of projects, portfolio of research outputs in general. And then finally, we also talked about other things that we could do, like, maybe we could implement something like a podcast series called something like Meet the R Engineers. There could be books about research software engineering with R. Actually, in this, I looked into the notes of the session on the books and it was really nice how they came up with, like, not only one book idea, but several book ideas. And I think there's something coming. So stay tuned for books and podcasts on topic as well. And yeah, finally, there's so many other things that you can probably do that we didn't even come up with in these 45 minutes that we had this short time. But if you have any ideas of how we could improve the situation for software and science, then just go ahead and do it. There's nothing, nothing much, at least stopping you. And stay tuned for the blog post about the incubator outputs. Now I'm almost at the end of my talk, and there's a couple of things that I want to still mention. So this is more of an advertising end, but it's very much connected to the things that I've said before. So there's two initiatives that I want to mention in this context, which have helped me personally to figure out how to build community and how to deal with this whole idea of reproducibility, working on open projects and stuff like this and these projects are called Mozilla Open Leaders and The Touring Way. So let me first talk about Mozilla Open Leaders. This is a training program for people leading open projects. There's actually several folks now that go into specific directions, for example, open leaders in the life sciences and such. So we could maybe think about forking this idea of training leaders for something like research software groups or something like this. So this is one thing that I wanted to mention. I was trained through the training program, and I learned so much through it. And it's really a great initiative. The other thing that I want to mention is The Touring Way, which to the outside world seems like an online book. So it's an online book about reproducibility, collaboration, project design, ethics and so on. But really, it's much more than that. It's an entire community and it's a really cool community for people like us, right? So the whole thing is focused on data science. And a lot of the people who are involved are research software engineers, others are data scientists, others are ethicists who work in data science. So it's really a cool project that anyone can get involved in. And yeah, check it out. It's a really good book. If you want to learn about how to make your work reproducible, this is, I think, in my opinion, the number one resource. So that's why I wanted to mention this as well. And finally, I want to make two advertisements about things that I have done recently and that I will do recently, which are my two podcasts. The first podcast is called Open Science Stories. It's very short episodes. Each episode is between three and 10 minutes. And it's little stories by people and very personal stories, actually, by people about something related to open science. And it's, I think it's turned out really great. And everyone who created an episode, all the storytellers really did a great job. So if you're interested in open science and like just getting into it and trying to figure out what it actually means, I think Open Science Stories is for you. I'm actually, I've published the last episode of the first season this morning. And I'm actually currently looking for a new host because I think for this kind of podcast, it would be nice to rotate hosts every season to get a little bit of different ideas in to get a little bit of different kinds of people into the podcast. So if you're listening to this and you think, oh, this is a great idea. I would love to do something like this. Get in touch with me because you can become the co host. You can also do it together as a group and be co host together, whatever is fine for me. I would love for this to to go on and to move on with someone who loves the idea of the podcast. And then there's one other thing that I want to mention, which is a podcast that I'm starting now. I've just recorded the first interview and it's so this one is an interview podcast called reboot academia and I'm sorry it's going to be in German. So this is for the German speaking folks out there. But this is essentially about the topics or some of the topics that I've talked about today. So it's about how we can rethink the way we do education in universities and in research and rethink the way we do actually research. I'm currently just getting started, but you can follow me on my journey, creating the podcast and I'm creating a blog so the QR code here sends you right to YouTube to the playlist that is the blog. So blog is a video blog for those who don't know. And I'm just recording a couple of videos with my thoughts but also with the journey itself. So if you're interested in that, check it out and the podcast doesn't have a website yet because it's not there yet. But if you follow me on Twitter, I will let you know about it. And with this, my keynote is already getting to its end but I'm really looking forward to your question questions. And finally, I want to thank again the amazing team behind this conference. I've felt really, really informed and really well taken care of throughout this whole journey. As a keynote speaker, so this is very new for me to be in such a prominent role. And I know how much work it is to organize use are as I was in the team last year. So I think I really think you did an amazing job of organizing this and you can feel extremely proud and how professional everything is and how you're welcoming people and. I think it's amazing. Thank you. Well, well, thank you Heidi for this great talk. I have a couple of questions lined up we will see how far we get in the 15 minutes that are left. The first question, are there any funding bodies that have specific funding sources for RSEs? That's a very good question. And there definitely are. And I'm probably not aware of everything. So in Germany we had recently a funding call for actually maintenance of existing research project research software projects, which I think is a great idea. That was by the DFG. I know that there's other kinds of rather specific funding bodies that help with research software or focused on research software. I can look them up and post them in Slack because I don't have them on my mind right now but maybe others can also chime in and we can collect something in Slack. I think there was supposed to be a group also in the incubator but I don't think we had enough people on that session right. Yeah. Yeah, just join the Slack group. I think it's called HashKeySybal. And there we can continue the discussion on how to score funding in this direction. So I'll post the links that I have there because I don't have it in my mind right now. Excellent. Thanks. So yeah, please join this channel on the conference Slack also to continue the discussion after the keynote. I have another question, a bit from the other side. So in your opinion, what would you suggest to a funding agency that wants to support RSEs? Is there a need to shift incentives? What is the one thing that they should start with? That's a really good question and that's a really big question that I'm not sure I can answer like in a few words. A couple of thoughts on this is that we want to really fund, if we want to really fund software and support software projects, then we don't need to only think about kickstarting research or software projects, but we need to also think about long term maintenance and long term funding for these key projects that continue over years and years, right? Because maintenance is a big issue that we have, I think especially in research software, because it takes a lot of time, it's not very prestigious. And I mean, this is one thing where I think we should start. Definitely incentives. So if we give out money, lots of money to research projects in general, I think we should think about not only counting again how many papers came out of this research project, but also did the research project end up with a good software, with maybe even a software community, because community management and this whole thing is also super important and super time consuming. Yeah, these are things that I think we should start thinking about when we think about how we can we improve the funding situation for software and science. Yeah, these are just a couple of thoughts, but that's definitely not everything. Yeah, I could probably talk another hour about this. Thank you very much. So another question was where to draw the line between what a PhD or postdoc should do in terms of programming and where the help of an RSE should be requested. So, where do you draw this line? That's, I don't know if that's a question that I can answer. So, I think it depends on, first of all, is there actually anyone you can ask, because if there isn't then, well, what, what choice do you have. And also depends on you really. So I really enjoy getting like deeply into these technical things and like figuring it out reading books about good coding and like, I really enjoy doing that. And I did that a lot as a master's student as a PhD student. And that's how I got my skills in that direction. I also joined like courses. So I took a lot of courses at the Zurich R courses and stuff like this, because I just enjoyed it and I wanted to learn it for myself. So I think it depends a lot on you and what you feel capable of doing. I don't think I can answer that in like in general. If you feel lost, then probably then at least it's time to ask someone. Thank you. So another question was how do can we explain to people the difference between an RSE and a normal software engineer? Do you have any thoughts on that? Yeah, and I think that's really, really important because I know that universities have had the idea or recent projects have had the idea to just hire software engineers from a company. And so from what I've heard at least, this didn't go too well because they don't understand the way research works. So they don't understand, they don't have the knowledge they need about the research itself because every research topic is different and they're all super complex because that's why we do research about it. And the way research as an endeavor works is often very different to how software projects and companies work. And so I think the big, big difference between a research software engineer and a regular software engineer is the knowledge and the intricate understanding of the research project at hand. And also like how research in general works. So there has been a survey actually about research software engineers and they were asked what kind of degree they had. And I think almost everyone had at least a master's degree and most of them had a PhD. So they really understand research as well as they understand software. Thank you. Another question was so for maybe from when you're in a not so high level position. How do you advocate for IC? How can you convince your professor to go for that road? And maybe another question that is kind of linked as this is not an established field. How do you build yourself a career in research software engineering? Yeah, that is the big question. So this is really something that I have been struggling with myself because I'm not like maybe a classical RSE, but I'm also a person who's like, yeah, very much somewhere, definitely in research, but not the classical researcher. Right. And I've really been struggling in the past years with finding a place for myself because I was doing a postdoc. And I was saying, okay, I want to be a postdoc in your group. But I, I'm kind of different than the others. And so that's how like, that's how my job interviews were like, I do this, this and this. I'm probably not going to write as many papers as the other postdocs. I hope this is okay with you. So I mean, I was really lucky to be working with people who understood where I was coming from. I think if you work with someone who doesn't understand it at all, and I don't think you have much of a chance. But I, I always try to be as open as possible and just tell people where I'm at and what I want to do. This maybe also led to more job changes than other people have in, in this amount of time. But I've been always very lucky with the people I've been working with in the sense that they understood this issue for me and they, they enjoyed that I was being open about it. So I just try to talk about it a lot and be very open about my thoughts on this and advocating for research software engineering and stuff like this. Building your own career, when there is no career path in this area is definitely a huge risk. So I can't say it any other way. If you're in a country where RSE is not a career path that's already there, then if you go towards that route, it is a risk. That's, that's what it is. And in the end, I think the lucky position that we're in is people who are able to work with software able to work with data that this is also a skill that's useful in many areas. So maybe we're in the great position that this is a risk that we can take. So, but I don't want to say that, hey, just go for it. It's easy because it's not, you're going to have doubts and it's going to be crazy and you might not succeed. I don't know. You might not never get a permanent position at a university if you just focus on software depending on where you are. Yeah, thanks. So, another question was, if you have thoughts about RSE outside of academia, so like in the corporate world, in NGOs, in government agencies. Yeah, I'm kind of that now. So I switched to accompany the UN Institute, because there I was promised that I was going to be allowed to do the things that I wanted to do without the pressure of having to write papers but with the possibility to write papers. So there are definitely companies that can do that, where that can work. And I think their career paths are maybe a bit more flexible, although I'm going to be honest, just because I've worked at a company for a month now I'm not like, I don't understand the corporate world yet. So I'm probably not the best expert to talk about NGOs. Sure. I think, I mean, that's probably very similar just with a different spin. But if they're doing research, then software is going to be a topic for them, just as well as at other research research projects and research institutions. I hope that answers the question. Well, otherwise, feel free to come back to Heidi in the Slack afterwards. So we are almost at the end already. I think I would like to ask one question myself. So if someone thinks, well, this sounds really cool and I want to start advocating for RSE, but there is no local RSE group or etc. Is there any support that they can get and where? That's a very good question. So definitely, if you're here now, right now, and if you say, oh, I want to do this, I think start writing in the Slack actually, there might be someone else from your country who thinks the same right now. And otherwise, get in touch with the existing communities. So in the slides, and I posted the link to the slides also in Slack. In the slides, I have a link to the list of all the existing communities and I think especially the community in the UK is super knowledgeable and helpful in that sense. And they can maybe help you reach out to people. But all the communities are, I'm sure, super helpful and open because we all have the same goal. So just, yeah, there's also an RSE Slack, which you can find through the same website or, I mean, Twitter might work as well. Yeah, that's my recommendation for now. The hashtag that you can use for RSE on Twitter is RSENG, so NG in small letters, because I think RSE itself was already taken. So most people use the RSENG hashtag and you might find someone through Twitter as well. There's different routes, but I think these are probably decent ones. So thanks for the question. I think that was very good. Well, thank you very much. Thanks again, Heidi, for coming to give this keynote. I think there were some very important insights. And if your question was not answered now during the keynote, please reach out on Slack and ask again and continue the discussion there. We now have a 15 minute break and then the next sessions are starting. We have a session on data visualization and a session on our introduction. So please join us there. And thanks again, Heidi, for coming and thanks to all of you for coming to this keynote. Yeah, thank you.