CHAOSS Data Science WG, Feb 24 2026  🔗 **Source:** [CHAOSS Data Science WG, Feb 24 2026](https://www.youtube.com/watch?v=OdclF7fipLA) Yeah. >> Okay. I just clicked I just clicked record so you can kick it off while I get my notes up. >> Okay. All right. We are officially recording. Welcome in everyone. It's a slightly shorter lineup than usual because of LF members and like two other co-occurring conferences over this exact time this week, which is great. Um it gives us a chance to cover anything in depth that we want to. Um, but uh yes, we'll uh get that link out onto the chat real quick. Feel free to put in your name and something that you are looking forward to and uh then we will just run through this agenda. I suspect it's going to be a lighter than 50 minutes today. Um but if there's anything we want to dive into, we've got time. So on that note, I'm also going to add myself into that uh attendees, but we could start with uh practitioner guides. Uh do we have any on those? Go ahead. >> Uh yeah, I give some some quick updates. There's not a lot. Um I've gotten a lot of really good feedback on the funding impact measurement guide, which means now that there's an awful lot of work in trying to synthesize that feedback and figure out what goes what goes in and what what doesn't. and I just honestly haven't had enough time to do that yet. So, um, so that's where that is. I've gotten lots of good feedback. As soon as I have time to sit down with that guide, which will probably be maybe after my little wave of conferences or maybe while I'm traveling. Um, I'll get back to that and we'll get that one finished. But it's it's it's pretty close um in that I've gotten lots of feedback. I just need to just need to finalize it because not all of the feedback is um aligned with the direction of the guides because I've gotten a lot of feedback from people who aren't super familiar with the guides. So, I just need to >> need to figure figure that out. Um the usage, adoption, and popularity. Uh we haven't made any progress on that. I think we're going to put that on an agenda item for one of these meetings in the in the future. Um, and then the uh research software impact. Uh, Matt's been super busy and so I haven't been bugging him to ask him where that is, but that's the one that Matt's been driving along with some folks at Curious uh, from a university standpoint. So, that's what we have on the practitioner guides. Okay, that brings us into project updates and foundation stats. I will also point out that we have David on this call. Thank you David for joining us. So uh David works at uh University College London. So uh we had a Slack thread and I think maybe an email chain or something around the foundational stats and the the grant that you were pursuing. So if you can cover that in the update that would be great. >> Oh, I will do a quick update on that. Um so we were looking at the I think it was the SSI funding. Um, but I think the real honest thing there is they're looking for a project that's at slightly higher maturity where we can demonstrate uh like end user community value which I think we absolutely can inside of 12 months. I don't think we were ready for that grant this time. Um, but I am extremely curious because 8n and augur are going forward with at least the first pass of that. So, I think it would be a really really good idea for us to uh come together and get any lessons learned out of that submission. Um, but I also think it might be worth having a grant focused working group for chaos is something that I have seen come out of that interaction. That's a super super high level. Um, but Ernest, do you have anything that you want to And there's a couple of new people on this call. So, do you want to describe the foundation stats just overall as a project? >> All right. Thank you and uh good day everyone. Um, so foundational stats is um we are trying to feel the metadata um on availability or differences in how different foundations um report metadata. For example, Apache presents their communication structure quite differently from um CNCF and Eclipse just using those three as a case study. And we are hoping if we can normalize the structure of this metadata in the data sets, we would be able to answer sustainability questions on a cross foundational level basis because previous research has not necessarily been comparing what does sustainability means across different foundations. We've just been asking per project or per package or per foundation as it may. So we hope that if we could gather this data we we will have found additional work around I mean basic work elementary or beginning work let me not use that foundation because of the mix word we have those initial results to answer sustainability questions across foundations I mean OSS foundations yeah thank you >> thanks Thanks, Ernest. I don't know, David, if you wanted if you had anything to add about the like the Augur SSI um grant. I don't know exactly what they're what they're doing. >> Sure. >> S was talking, but in case you don't, please let me go. >> Right. So, yeah, I just joined to see exactly what definition STS were exactly and how this could help. So for augur essentially one of the thing that they need is a a UK based um person and well that's what we were saying also in Slack a and because we were playing with a back in October and we were making a lot of uh complaints in a good way like suggesting supporting things like saying why this is not working and and we have like pretty much three people almost 100% for almost a month. We did kind of a contribute in in quite a bit and and yeah we we in in in our role in the university because we are helping supporting a lot of teams. Last year for the first round we I think we were in 22 uh proposals as our team and this year I think we are way less that has been kind of a countdown after probably the feedback we got on the first round and we are now like around eight uh counting over I think. So yeah, the the idea for Aug uh if I remember correctly was pretty much on on helping uh including other forges like or better support for other forges be beside GitHub and some other things but yeah as Sal was saying it's more about the maintenance and and keeping how that project can become more solid. The problem is that their main one of the main questions is UK impact and and yeah we can promise that this is going to have an impact but how proving that has been quite hard at least from the previous round where in some in some proposals we have like a lot of papers from UK that they are using this software and they were like nope this is not enough and others is like uh yes you show a lot but it's too much and I'm like okay it doesn't make sense. Uh so yeah we are we are not clear yet of uh what will happen but this year one of the interesting uh things that for at least for the expression of interest is that people are going to be reviewing so it's going to be peer review by the people who are taking part on it. So that will maybe help which I think what happened last time that the reviewers were not too familiar with what is the call and and how funding was working because some some comments we receive are like oh this is too much for um uh overheads and like well that's the overheads that this grant like that's all that we can ask for. So it was kind of a quite I don't know it was quite separated like people reviewing the thing was not really understanding what they were reviewing in terms of the conditions that those projects were. So yeah I don't know what what other questions do you have that I can help answer. >> That was extremely helpful. >> Okay. >> Yeah. So I'm really interested at this intersection of funding open source and academia together. Um when I was doing open source and academia 15 years ago, you just got funding for Nadamia and happen to be doing open source. Now that there's a true birectionality there and I think this is a grant that represents that shared values intersection. Um so I am a little bit curious to see particularly the way that this grant possibly moves forward with like auguring 8 knot and see the response back. The first thing that I can think of from what you just described is I have found that specifically in the UK, if I tell them that my ability to get this to the UK means I'm going to go give a talk on the subject or workshop on the subject in London, Manchester, and Edinburgh, they're going to say yes. They're not even looking for quantitative value past that. They just want to see that you're going to be physicalized in the UK. So that might be something that we add and see if it makes a difference in the future because it's very easy to go give a talk in London, Manchester, and the UK um in 12 months time off of a grant. Um that might be worth adding or considering, but I'm also trying to get my sea legs around uh academic grant funding in the UK. It's different than it was 10 years ago. So, I'll take advice from anyone that's in it at this moment. >> But thank you for that. And if you got the mic is back to you. >> Yeah. No, I I So, this grant is completely new, right? So it was last well new last year but that was kind of a and it's kind of experiment that the sourceity is running and they say they say the way they are running the way they are doing the review the way they are doing the things they're quite open on those things so that's quite good uh but I don't think in open source in general on the rest of the of the uh funding schemes are much different than in the past but yeah I mean uh there's maybe more in research there's been more push of at least uh the figure of research software engineers has been more kind of a it has to be or if it's be if it's if it's if there's a research of engineer involved in a project it has more chances to be funded uh but not as much as the open source part of it some funding agencies uh say that it has to be but uh then how is enforce not that's a whole different thing. >> Yeah, I might use yellow. Sorry, what was that? >> Just connecting connecting the dots. Adrian came a little late. I was just catching them up. >> All right, right. Okay, cool. >> Oh, my chat window is empty. It's like all blank anyway. Yeah. So So yeah, I I don't know how much more info and I don't know. Yeah, like uh this particular call I find that it's a a lot as well about setting up governance and setting up community. >> Yeah. So I'm looking but well the one I the one I was more involved last year like writing it like I was in the project one side project I have uh it didn't pass the expression of interest but it was most focus on okay we want to improve something there's no new features no software to in include there but it's more about how we get contributors to be sorry how to be get users to become contributors how to be contributors to be maintainers that kind of a path and but yeah I I didn't I didn't mine didn't pass and from the 21 that we did I think that uh only a couple passed to the next phase and then they didn't got funded at the end and one was for mach league for example so it's it's >> this price of all sizes and this is not for open source only >> yeah exactly Yeah. Yeah. >> There's also type of compet compet competition going on. >> Yeah. Yeah. And I think uh when we're looking at this because kind of I think this is just a let's learn what we learn and I will take any advisement as we go and let's just soak up because it is a different call than it was a year ago and it could be a different call next year. But I want to get an idea of what it looks like and feels like to proposition the value of open source within academia in 2027. And I feel like that's going to be a different call to action than even this year. So interesting open thing to keep in mind. Um but thank you for candidly describing it as exactly how I would describe it from what I've read. >> I think right on the right on the target there. And well uh I'm also building an OSPO in in UCL. Uh so uh there is there's also the curious network as well building like kind of a OSO on on universities and academic institutions. So maybe there's also more opportunities out there that >> Yeah. >> Yeah. >> from >> Yeah. And I did half of my PhD at UCL, but I was in the functional neuroiming department, so you didn't see me. I was in a basement. But it's a good university and I'm excited there's no there. Um, cool. I would say for now I think that covers that topic. Um although it's a topic we should absolutely come back to as we get more feedback. Um next on the agenda here um virtualization notebooks with augur analysis. Cali is not here today because they're at the uh Linux Foundation membership meeting but that has been coming together and it is very cool. So come to the next session and I think you'll be able to get more insight on it. And then the costs, are we covering that today? >> Uh, no. Matt's not here. I suspect he's also at the at the member summit. Um, and I haven't had a I haven't seen any pull requests from him and I haven't had a chance to catch up with him and see if he wants any help moving some of that stuff over to the repo. >> Cool. Sounds good. In that case, then it looks like the last thing on the agenda is from me, and I'm excited about it. So this is about uh bringing in new maintainers, on boarding and sponsoring them, which is to say that I have identified earnest on this call as being excellent as a long-term maintainer to the foundation project because they came in as a contributor and then immediately provided more value than I was providing as the maintainer. And that's how you know when you've got your next best maintainer. Um, so what I'm trying to do there is to set up a sponsorship program for the maintainers. This might be unique to this space. It's not unique to me. I work in the security side of open source. And before we're able to maturely do a handoff with a maintainer, we have to ensure that we have not only taught but have demonstrated an instance of demonstration of proof around things like handling vulnerabilities. if they're disclosed to you, things like that. So, we're not doing a securityoriented sponsorship in this case, but I am in the process of building a chaosoriented data scienceoriented maintainer handoff, which both respects the culture of the community and also respects the complexity of data science repositories, which are slightly different from a typical software repository. Um, so as I'm building that out, I'm getting initial feedback now, but once we make that public as a possible sponsorship charter, I would be open to feedback from anyone on that subject. And I'm also open to any ideas on the call if you think the idea of a sponsorship is good, bad, or if you've seen something better. Um, but what I want to responsibly do is hand off that repository knowing that the next maintainer has all of the skill sets that I had while I was managing that. Um, any questions or comments there? I think that's something that I want to dive into in depth probably in the next call. Um, yes. And Adrien, I am going to I can add you in on the conversation or sort of bring you in. just have a MD that we're building right now around the sponsorship and I'm looking for anyone's insights on it before we make it official official. But as soon as we do, Ernest and I are going to sign a sponsorship agreement. Now, a sponsorship is only different from a mentorship in that I am the sponsor. I am the only person that can fail. So, I have to guarantee that the next maintainer has all the skills that I say and that I'm going to give them within a year. And that's how we do it in security. So, it's probably going to work better here where we're doing it with less stressful things to learn. Um, so yeah, that's what's happening there. It's still in progress. Um, but I'm hoping we can come to the next uh standup here and both myself and Ernest can talk about the experience of sort of deliberating that and what it currently looks like and uh look for any additional feedback. That's a document that can and should evolve and it should be probably bespoke to every repository depending on the requirements of the repository. So that's where that's standing. That's a lot of words where in the future in about 2 weeks I'll just show you the document and we can walk through it and understand it. Um, I also think that I wish when I was in academia that I had had the security experience of knowing how to do extremely hygienic handoff because we lose a lot in academia from not doing very good handoffs and this is a way to fix that problem. So, I think it's really good for chaos. Um, and that's it. That's the note on uh that update. >> Yeah, that's great. I'm I'm so glad that uh you and Ernest have been have been working on this. I think it's something that we can we can definitely use, like you said, in other places within within Chaos. I think it's something that's it's really valuable and we probably haven't done enough of before. So, I think that's I think that's good. And I just put it um you probably saw this under the agenda for upcoming meetings for the next meeting. We'll put it on the agenda again. >> Cool. Yeah, we'll be ready in two weeks. >> Awesome. >> Cool. Okay. Uh, and I think that puts us uh right now at any opens for the next meeting. Anything we want to cover. >> The silence is not deafening. It's a very nice break from what usually happens at the end of this call, which is like everyone's throwing down a lot of things they want to cover. That's really nice. Um, that means very likely in two weeks you're going to come here and you're going to see people have updated this. Um, but if we've got a nice loose schedule for next week, I would like to create more space within these calls, 10 minutes or so for new entries to the community because we're often getting those and it's a really nice to just make sure we have that time laid out to say, "Hi, hello, and are you on the right Slack? I don't feel so alone with over 3,000 members now. >> Yeah. >> Excellent. >> Awesome. >> Okay. Anything else that we are looking to cover? So, there's the upcoming meetings and reminders. Do you want to cover that? >> Yeah. I just added a couple of conferences that have open CFPs. So, um, the ASF's community over code is going to be in Glasgow this year and the CFP deadline is March 20th. All things open, uh, they have a slightly new format, so it's a slightly shorter event. Um, and their CFP deadline is March 31st. So, those are both, uh, I've heard great things about Community Over Code. I've never been to it. I've been to all things open a couple of times, and it's a it's a lovely conference. So, I would encourage people to submit talks if it's something that you're interested in attending. I can't believe All Things Open CFP is already open, but I'll submit. I'm ready. >> Yep. >> Okay. Excellent. Um, anything else from anyone on the group? Particularly if you're new, this is a very good time to soak up any wisdom from the call. Any questions, comments, or concerns? >> What is the archive project? I asked that. Yeah. So, so we uh we had this idea that um so so in particular so you get a lot of projects that are archived for various reasons, right? So you get projects that are archived because there's no activity. Um and and that is that's one case that's probably I would say I guess maybe the least interesting case. But you also have a lot of projects that get um archived even when there is activity. So you've seen this with some Google projects where they've archived the open source project and pulled the development internal and so you get you also see popular projects that have been archived um for for various reasons that um you know and it depends. Sometimes it's a popular project and you've got a maintainer who's just tired of working on it or they've you know they've had a baby or they got a new job or whatever. There are lots of reasons for people to stop working on projects. And so we started this project um with the idea that we would look at maybe I forget the top x number of projects on GitHub and manually look at them and try to decide why they were archived. Um and we so and we we found that that that just wasn't getting much traction. We the data that we were getting um from those manual assessments wasn't particularly interesting. Um, so I stopped working on it and but Calie, who's one of the other co-chairs of this group, was like, "No, I think there's I think there's still something there. I think we need to just rethink rethink it entirely. Just basically start over and maybe do something else, but something around projects being being archived. So So that's where it is right now. It is it it was an idea. That idea didn't work. So we put it on the shelf. Um, but we think there's still something there and we just need to figure out what that is. So, that's one of the things we're going to put on the agenda when we don't have anything else on the agenda and we're just going to spend some time talking about what that project should be. >> Cool. That's awesome. That's I'm really down with archives. So, >> awesome. >> Yeah. Yeah, that's a cool project. There's so much information that I want to get out of that, but it just is going to require a single person to think about it. And I'm not saying this is a full-time. and saying for 3 to four hours a week, I need someone to think about that for about three or four months and we will have that done. Um, and it will be a massive contribution. Um, the signals that you can pull from that if you could ever get to anything predictive and even if you can get to something descriptive, that's a massively valuable piece learning. Can I ask how you're separating all the people leaving GitHub and tracking that? >> Yeah. So, this is specifically for stewarded projects that were going through an incubation process. So, that's not whether or not they left GitHub, it's whether or not they were no longer in a stewarded ecosystem. So, they were literally dropped from the ecosystem. That's >> not individual open projects like projects that you guys have worked on then. >> Yeah. Uh yeah, so this is for projects that are within the Linux Foundation broadly. So it can be any sub foundation, the Eclipse Foundation, the Apache software foundation >> and other sources. So we're also working with Fedora as well, but I don't think we're pulling from these stats for that. Um but the archiving is particular to does it reach a maturity within a stewarded graduation process or does it get removed? If you're interested in the statistics around does it just move to code or elsewhere. So am I and I'm still working on figuring that out. Um, and uh, if I get the answer, I'll tell you, but that is not gerine to uh, whether or not it uh, gets archived. >> Yeah, because I was kind of trying to look at all like big projects on GitHub and looking at archive. And so this is one of the ways that we thought about narrowing it was looking at it in the context of the foundation stats project and looking at at those projects being being archived. So that's that's a a piece of it. But we could we could we could here. Okay. So here's the thing. This is the data science working group and we can uh we can study these in whatever way we want and build whatever data sets we decide are interesting. So if there's something that anyone wants to wants to look at, we can certainly we can certainly start just doing that. We don't you know we we have some ideas of of what we are thinking but it doesn't have to be it doesn't have to be that. anybody can work on really really anything that they find interesting and we'll see we'll see what we can do with it. >> Yeah. But you sound competent and you sound like if you were to have a scope co-developed, you might be able to get it done. So what I would really recommend is look over some of the documentation that we have available. Send me or nest questions around anything that we have this week because everyone else is at a conference. But if the archives project or that space is really interesting to you, I think 30 minutes with myself and Calie and Ernest all sitting down and talking, Don can be there because Don is the central wisdom. But in terms of getting you scoped in to like understanding where we could move on this and where you could have a specific impact, those are the three people I'd want to come together to advise you on point you where. um and then see what gains your interest and then all of us are there to support your code. >> Awesome. >> And I just I just threw a link in there um to the the data set section of the repository. So that has details about the foundation stats projects that has what we did on the archive project um and then stopped doing. So you can you can kind of poke through poke through some of those and see see if there's anything anything interesting. Um, yeah. But yeah, also I do recommend kind of starting with with uh Sal and Ernest and and Cali because I so I've been doing the the work I've been co-chairing the data science working group uh for almost three years and my plan is to step back at the end of March. So um so I'll I'll continue to participate but I'm going to step down as co-chair at the end of March. So that's that's part of why working with I think Sal and Cali is probably a better better step. >> Yeah. But it's a really good support system because on my side I'm s I'm very much the open source volunteer that's here because I want to know the answers and Cali is Red Hat. So they've got Red Hat behind them. So if we actually get something that's interesting or you need extra developer hours that's actually there in this circumstance. So uh yeah, you're not unsupported. You could actually have a good degree of corporate FTE hours on this if we scope it correctly. So worth diving into seeing if it's curious to you. And it's more valuable to me to know what questions you have after looking at these repos than anything else. So send them to me. >> Yeah, I'm taking a look right now. It's >> I love >> I'm a game developer by trade, artist by trade, but then I've put on a million hats. And I find that we've really as humanity, we've screwed up uh a lot on making things fun. And uh one of them is archiving. Like there's a huge amount of data out there that there's so many people that want access to it. And there's a lot of people like kids even that want to mess with things to learn but they just don't know how and it's really complicated and as an artist most programmers have no idea how to organize things for the general public. >> So like that's the biggest step like the biggest crossover for me um is making sure if there's stuff that needs to get worked on there are people out there wanting to work on it. Yeah, >> I've talked to so many maker spaces that don't know any of this stuff exists. Like they don't know where to start. Um, and it's kind of funny. So, this is where I'm starting to be the I guess I've been a tech artist a lot of times in projects. And that's where I think that the world needs a a little bit of a in between person to go in between uh tech and art and art is more general populace. >> That's where I want to step in. >> Well, good. I'm glad you're here. I get all of my dopamine from my GitHub PRs, both accepting and providing. Um, there's a lot of gamification in this process. >> Um, but yeah, I think for today, honestly, just go give it a look. All of your curio curiosity is valid and useful. You can throw an issue, a comment, a question, a Slack message. We will respond. Let's see over the next two weeks if you're just playing with the content, getting some questions answered, and then let's scope it in and we'll have a call with you because that would be interesting. Um, I do want to ask, is there anything else that anyone wants to either cover today or make sure that we write down on this agenda tentatively so we don't forget? Uh, profound silence. I love that. That's my favorite. Um, the last thing that I will say is we're not quite to the process of congratulating a new maintainer in open source, but we're very, very close. So, we're going to do just a very quick we're almost to having a new open source maintainer in this group, which is Earnest. So, we're going to give a small clap this week because in two weeks we're going to do the official handoff and they will be an official maintainer and then we have to give our official claps. Um, and I'm super excited about that. So, uh, all cheers to Ernest, who's been invaluable, and I'm so excited to have them here. >> Yeah, this is this is welld deserved because what was happening was Ernest was doing all of the the code reviews and providing all of the feedback and then pinging me and saying, "Hey, can you merge this for me?" Um, which just seemed silly to have me in the loop at all. And so, Sal and I talked about it and yeah, I think it's great. >> Yeah, >> congrats. >> Thank you so much, everyone. Thank you, Don. Thank you, sir. I'm happy to continue contributing. >> Yeah. Yeah. No, it's been amazing. I've never I literally in my entire career have never had to just stop and be like, "No, no, no. This is not a contributor. This is a maintainer." So, good. I'm glad. We'll get you trained in. Um, excellent. Well, it could be a slightly shorter meeting uh today if we've got the agenda all settled. Thank you all for coming, those that come often and those that are new. Um, excellent session and when you come back next time there will be a lot more faces on this call. It was really nice to have a smaller group here today. Um, so thank you for that and we'll see you I think next time. >> Thanks everybody. >> Have a good one. Bye everyone. Bye., CHAOSS Data Science WG, Feb 24 2026, YouTube, 01kp8h85vj95xnvxjf3kr28tvy.