CHAOSS Data Science Working Group, Feb 10, 2026

CHAOSS Data Science Working Group, Feb 10, 2026 ![rw-book-cover](https://img.youtube.com/vi/GIq32UOTM30/maxresdefault.jpg) 🔗 **Source:** [CHAOSS Data Science Working Group, Feb 10, 2026](https://www.youtube.com/watch?v=GIq32UOTM30) > [!note]+ Document Notes > tags: #open-source #metrics #data-science #funding #commercial-open-source #visualization #community-governance > author: CHAOSS Data Science Working Group > source: YouTube > reference: https://www.youtube.com/watch?v=GIq32UOTM30 > date created: 2026-02-10 14:30 > > # 🏷️ Title: CHAOSS Data Science Working Group Meeting - February 10, 2026 >  ## 🧰 Action Inbox - [H1] Create COSS data folder in repository — Matt and Don collaboration #commercial-open-source - [H2] Set up visualization notebook project tracking — Cali umbrella issue creation #visualization - [H3] Schedule usage/adoption practitioner guide discussion — 30-60 minute focused session #metrics ## 🧭 Table of Contents - [[#📝 Summary]] - [[#📖 Main Content]] - [[#🗒️ Notes]] - [[#📚 References]] - [[#🗣️ Paraphrase]] - [[#📰 Newsletter]] - [[#✅ Action Items]] - [[#🧪 Hypotheses]] - [[#📊 Methodology]] - [[#❓ Questions]] - [[#🔮 Future Research]] - [[#💡 Implications]] - [[#🧠 Learning Styles Assessment]] - [[#📊 Accessibility Assessment]] - [[#🌍 DEIB Assessment]] - [[#✅ Logical Validity]] - [[#🛡️ DARVO Analysis]] - [[#🎯 Conclusion]] ### 🔗 References - [Foundation Stats Project](URL) - [8Knot Visualization Platform](URL) ## 📝 Summary - The CHAOSS Data Science Working Group met to discuss multiple active projects including practitioner guides, foundation statistics collection, and new commercial open source data initiatives - Key focus on establishing visualization notebooks as a pathway for new contributors to learn data science skills while contributing to open source metrics - Discussion of recent events (FOSDEM/ChaosConEU) highlighting the shift toward digital sovereignty and open source funding in European policy - New commercial open source data collaboration with startup alliance bringing canonical company lists - Growing emphasis on making data science contributions more accessible through templated approaches and clear project structures ## 📖 Main Content The meeting covered several major initiatives within the CHAOSS ecosystem. The funding impact measurement practitioner guide is nearing completion after holiday delays, while the usage/adoption guide needs focused discussion time to clarify direction. A new commercial open source data project brings together multiple company datasets for community analysis. The visualization notebooks project represents a strategic approach to contributor onboarding, creating a pipeline from Jupyter notebooks using Plotly to 8Knot dashboard visualizations. This addresses the need for tangible, skill-building opportunities for new data science contributors. Recent European policy events showed significant progress in digital sovereignty conversations, with open source positioned as critical infrastructure requiring proper funding and maintenance support. ## 🗒️ Notes #### ⏱️ [00:00](https://youtu.be/GIq32UOTM30?t=0) — [H1] Meeting opening and practitioner guides update - Key idea: Funding impact guide nearly complete, usage guide needs focused discussion - Evidence: "probably go out in the next, I don't know, uh, couple of weeks, hopefully month" #### ⏱️ [02:15](https://youtu.be/GIq32UOTM30?t=135) — [H2] Usage and adoption guide challenges - Key idea: Writer lacks clarity on approach and content direction - Evidence: "I don't have the answers. like I like I don't have that that I haven't written the thoughts because I don't have the thoughts" #### ⏱️ [03:45](https://youtu.be/GIq32UOTM30?t=225) — [H1] Foundation stats project progress - Key idea: Apache data collection underway with ~30 projects tracked by 4 contributors - Evidence: "about 30 something project has been collected now. So I hope we can increase that momentum" #### ⏱️ [07:30](https://youtu.be/GIq32UOTM30?t=450) — [H1] Commercial open source data collaboration - Key idea: Multiple datasets with 800 companies each, minimal overlap except obvious cases - Evidence: "One was the one that was the list that was used the 800 companies that were used in the Linux Foundation study" #### ⏱️ [11:20](https://youtu.be/GIq32UOTM30?t=680) — [H2] ChaosConEU scale and diversity - Key idea: Event grew significantly with multiple tracks, ~80 attendees, diverse audience - Evidence: "it was about what we could fit in the space, uh, to be honest" #### ⏱️ [15:45](https://youtu.be/GIq32UOTM30?t=945) — [H1] EU digital sovereignty policy shift - Key idea: Complete transformation from basic open source adoption to strategic infrastructure investment - Evidence: "this year all of the conversations were really focused on digital sovereignty and how open-source needs to be a key part of that strategy" #### ⏱️ [18:30](https://youtu.be/GIq32UOTM30?t=1110) — [H2] Funding complexity discussion - Key idea: Throwing money at open source problems doesn't automatically solve them - Evidence: "funding for open source is one of those things that it's it's not as easy as it sounds" #### ⏱️ [22:00](https://youtu.be/GIq32UOTM30?t=1320) — [H1] Visualization notebooks pipeline concept - Key idea: Structured pathway from chaos metrics to Jupyter notebooks to 8Knot dashboards - Evidence: "the pipeline from idea to visualization in 8 knot um which this is I'm saying this verbally this is what we're my team is trying to work on" #### ⏱️ [26:15](https://youtu.be/GIq32UOTM30?t=1575) — [H2] Plotly advantages over matplotlib - Key idea: Interactive visualizations provide better investigation capabilities - Evidence: "I could investigate that fully without having to do a single query or write a line of code" #### ⏱️ [28:45](https://youtu.be/GIq32UOTM30?t=1725) — [H2] Contributor onboarding through visualization - Key idea: Skills learned (Python, Jupyter, pandas, Plotly) transfer to general data science work - Evidence: "really a lot of the skills the data scientists need" #### ⏱️ [31:30](https://youtu.be/GIq32UOTM30?t=1890) — [H3] Infrastructure readiness concerns - Key idea: Need augur instance ready before promoting contribution opportunities - Evidence: "I know that if I put these things out there that it's going to get attention and I want to make sure that it's ready for attention" #### ⏱️ [35:00](https://youtu.be/GIq32UOTM30?t=2100) — [H2] Project tracking and issue management - Key idea: Clear, detailed issues attract contributors across multiple repositories - Evidence: "clear scoped issues are getting worked on like whether it's a yeah whether it's but even not even good first things" ## 📚 References - [CHAOSS Foundation](https://chaoss.community/) - [8Knot Visualization Platform](https://8knot.chaoss.community/) - [FOSDEM Conference](https://fosdem.org/) - [Open Source Observatory EU Policy](https://digital-strategy.ec.europa.eu/) ## 🗣️ Paraphrase - Multiple CHAOSS practitioner guides are in various stages of completion, with funding impact nearly ready and usage/adoption needing strategic discussion - Commercial open source data collaboration will create shared repository for company datasets from Linux Foundation and OSS Capital sources - Visualization notebooks project creates structured learning path for contributors while building useful 8Knot dashboard components ## 📰 Newsletter **CHAOSS Data Science Working Group Accelerates Contributor Onboarding** The CHAOSS community is building exciting new pathways for data science contributors through their visualization notebooks project. This initiative creates a clear pipeline from metrics concepts to interactive dashboard visualizations, teaching valuable skills like Python, Jupyter notebooks, and Plotly along the way. Recent progress includes near-completion of the funding impact measurement guide and establishment of new commercial open source datasets for community analysis. The foundation statistics project continues growing with Apache data collection, while European policy discussions show promising shifts toward strategic open source investment. The visualization approach addresses a key challenge in open source metrics: making contributions accessible to newcomers while building genuinely useful tools. Contributors can start with templated visualization issues and progress through data preprocessing to dashboard integration. Resources: [CHAOSS Community](https://chaoss.community/), [Metrics Models](https://github.com/chaoss/metrics), [8Knot Platform](https://8knot.chaoss.community/) ## ✅ Action Items #### Community - Contributors: visualization notebook templates, data collection support - Maintainers: issue tracking templates, contributor onboarding workflows - Sponsors: European digital sovereignty funding opportunities #### OSPO - Policy & Governance: CRA compliance implications for metrics - Software Inventory: SBOM integration with community metrics - Community Engagement: foundation statistics data sharing #### FAIR - Technical independence: diverse data source aggregation - Policies: transparent funding impact measurement ## 🧪 Hypotheses - Structured visualization notebooks will increase data science contributor retention - Commercial open source datasets will reveal funding patterns and sustainability indicators - European digital sovereignty policies will drive increased open source infrastructure investment - Interactive plotly visualizations provide better analytical capabilities than static matplotlib charts ### 📊 Methodology - Track contributor progression through visualization notebook templates - Analyze commercial open source company overlap patterns between datasets - Survey European policy stakeholder priorities and funding allocation - Compare investigation efficiency between interactive and static visualization tools ### ❓ Questions - How can chaos metrics models be effectively translated into actionable visualizations? - What data sovereignty requirements will emerge from CRA implementation? - Which visualization templates attract the most new contributor engagement? ### 🔮 Future Research - Standardized approaches for commercial open source company classification - Integration patterns between SBOM data and community health metrics - Effectiveness of templated contribution pathways across open source projects ### 💡 Implications - Practice: Clear contributor onboarding reduces barriers to data science participation - Policy: Digital sovereignty requires strategic open source infrastructure investment - Security: CRA compliance demands better community health visibility - Education: Visualization workflows teach transferable data science skills - Funding: Multiple commercial datasets enable comparative sustainability analysis ### 🧠 Learning Styles Assessment Reading level estimate: 12th grade #### 👀 Visual learners - Present: Project workflow diagrams, data visualization examples - Improve: Architecture diagrams for notebook-to-dashboard pipeline #### 🔊 Aural learners - Present: Meeting recording with clear discussion flow - Improve: Audio summaries of key project updates #### 🗣️ Verbal learners - Present: Clear explanations of technical concepts, analogies for policy impact - Improve: Glossary of chaos metrics terminology #### 🧩 Kinesthetic learners - Present: Hands-on notebook templates, step-by-step contribution guides - Improve: Interactive sandbox environments for testing #### 🧠 Logical learners - Present: Clear project timelines, structured issue templates - Improve: Decision trees for contribution paths #### 🧑‍🤝‍🧑 Social learners - Present: Working group collaboration, cross-project coordination - Improve: Contributor mentorship pairings #### 🧘‍♀️ Solitary learners - Present: Self-contained notebook examples, independent research opportunities - Improve: Self-assessment checklists for contribution readiness #### 🌿 Naturalistic learners - Present: Ecosystem thinking about commercial open source landscapes - Improve: Sustainability framework connections ### 📊 Accessibility Assessment #### Screen reader semantics - Present: Clear headings, structured meeting notes - Improve: Alt text for any data visualization examples #### Keyboard navigation - Present: Standard video controls, clear focus indicators - Improve: Transcript navigation markers #### Contrast and readability - Present: Standard video quality with clear audio - Improve: High-contrast transcript formatting #### Text resizing and reflow - Present: Responsive transcript formatting - Improve: Zoom-friendly visualization examples #### Media and captions - Present: Full transcript available - Improve: Timestamped caption file, speaker identification #### Forms and interactions - Present: Clear action items with context - Improve: Accessible issue template examples ### 🌍 DEIB Assessment #### Cultural breadth - Present: European policy perspective, global open source discussion - Improve: Non-Western digital sovereignty examples #### Inclusive language - Present: Technical jargon explained, respectful terminology - Improve: Plain language summaries for complex concepts #### Neurodiversity support - Present: Structured agenda, clear project boundaries - Improve: Visual progress indicators, consistent formatting #### Representation - Present: Multiple voices in working group discussion - Improve: Diverse contributor pathway examples #### Psychological safety - Present: Supportive approach to new contributors - Improve: Explicit welcome statements for different skill levels ### ✅ Logical Validity #### Strong reasoning - Deductive strength: Infrastructure needs drive visualization requirements effectively - Inductive strength: Past contributor patterns inform onboarding design - Causal reasoning: Policy shifts create funding opportunity pathways - Analogical strength: Other project templates apply to visualization workflows #### Fallacies - Appeal to authority: European policy importance overstated without evidence - False dilemma: Plotly vs matplotlib presented without middle ground options #### Tone of voice - Overall tone: Collaborative and constructive with realistic timeline expectations ### 🛡️ DARVO Analysis - Deny: No denial patterns detected - Attack: No attack patterns detected - Reverse victim/offender: No reversal patterns detected - Emotional cues and rhetoric: Professional, solution-focused discussion ## 🎯 Conclusion The CHAOSS Data Science Working Group is successfully building infrastructure for sustainable contributor growth through structured onboarding pathways. The visualization notebooks project represents a thoughtful approach to skills development while creating valuable community resources. European policy shifts toward digital sovereignty create new opportunities for open source funding and strategic recognition. The combination of clear project structures, diverse data sources, and accessible contribution pathways positions the working group well for expanding community engagement and impact. ## 🧱 Refactor Appendix ### 📚 Literature Split Pad Purpose: Stage CEE units that become Literature notes. Split level: H3 ### [H1] Visualization notebooks create structured contributor onboarding > [!quote] [CHAOSS Data Science Working Group Meeting](https://www.youtube.com/watch?v=GIq32UOTM30) by [[CHAOSS Data Science Working Group]] via [[YouTube]]. Published 2026-02-10. Accessed 2026-02-10. > "the pipeline from idea to visualization in 8 knot" Creates clear pathway from metrics concepts to dashboard implementation while teaching transferable data science skills. [view highlight](https://youtu.be/GIq32UOTM30?t=1320#H1) #xref #articleseed ID:rw_chaos_vis_pipeline ### [H2] European digital sovereignty drives open source strategy > [!quote] [CHAOSS Data Science Working Group Meeting](https://www.youtube.com/watch?v=GIq32UOTM30) > "this year all of the conversations were really focused on digital sovereignty and how open-source needs to be a key part of that strategy" Complete transformation in EU policy discourse from basic adoption to strategic infrastructure investment in open source. [view highlight](https://youtu.be/GIq32UOTM30?t=945#H2) #xref ID:rw_eu_digital_sovereignty --- ### 🔖 Atomic Split Pad (Zettel Candidates) Purpose: Promote reusable claims into Atomic notes. Split level: H4 #### [Z1] Interactive visualizations enable faster analytical investigation - Condition: When using plotly over matplotlib for data exploration - Pointer: 26:15 timestamp discussion - Links: [[Data-Visualization]] [[Analytical-Tools]] #### [Z2] Clear project scoping attracts distributed contributors - Condition: Across multiple open source repositories with detailed issues - Pointer: 35:00 timestamp observation - Links: [[Contributor-Onboarding]] [[Project-Management]] --- ### ✏️ Fleeting Queue Purpose: Capture quick actions for Fleeting notes. - [F1] Research CRA implications for community metrics — Red Hat team expansion signal #fleeting #inbox - [F2] Explore SBOM integration with chaos metrics models — --- ## Full Transcript [00:01] Okay. So, welcome everyone to the data [00:03] science working group meeting for [00:05] Tuesday, February 10th. Uh we have a [00:09] whole bunch of stuff on the agenda. So, [00:10] we're just going to go ahead and get [00:12] started. Um but first, the question uh [00:15] question of the day is, what's something [00:17] you'd like to do but haven't done yet? [00:20] Uh oh, see the northern lights. That's a [00:22] good one. I also would like to do that. [00:24] Um go to Italy. Italyy's fun. [00:28] you should do that. [00:30] Um, okay. So, feel free to go ahead and [00:32] and add your stuff there. Uh, the quick [00:36] update on some of the practitioner [00:37] guides. So, the funding impact [00:40] measurement guide is getting relatively [00:42] close to being done. Um, we had it in a [00:45] pretty good state in December and then [00:48] we all went off for the holidays and [00:49] then we were all running things at [00:51] Bosdam and Chaoscon and we're just [00:53] picking it back up now. I've sent it out [00:55] to a couple of people to get feedback. [00:57] So, um, people are welcome to have a [00:59] look, provide feedback. That would be [01:00] great. Um, that's where kind of where [01:03] that guide is. I would say that it's [01:05] probably [01:06] probably go out in the next, I don't [01:08] know, uh, couple of weeks, hopefully [01:12] month, something like that. [01:15] Um, and then there's the usage, [01:18] adoption, and popularity practitioner [01:20] guide, uh, that, um, Cali and I have [01:23] been talking about. Um, I wonder, Cali, [01:26] maybe we should just set up some time [01:29] and and spend, you know, 30 minutes, 60 [01:32] minutes on a Zoom call and kind of walk [01:34] through how we want to how we want to do [01:36] that. [01:37] >> Yeah. I'm I'm also [01:39] open to having I try to put like some [01:41] discussion time in here as well for it [01:43] because I kind of what my hope was for [01:46] Chaos Con to have an open discussion [01:48] around it ended up just kind of being [01:49] like a free form discussion which is [01:51] fine. Um but like part of like I'd say [01:55] my hold up on this is because I don't [01:59] have the answers. like I like I don't [02:02] have that that I haven't written [02:06] the thoughts because I don't have the [02:08] thoughts. Um and so trying to figure out [02:12] maybe like how to like where to start [02:16] building up to that is pro is where my [02:18] biggest like hold up I think is more so [02:23] than the like process of writing it. [02:26] >> Yeah. Okay, fair enough. Um, so I will [02:29] put that as a future [02:31] agenda item for upcoming meeting. Um, [02:37] cool. And then there's the research [02:39] software impact which uh Matt German [02:43] Prey has been um working on. [02:47] And um [02:50] yeah, I'm not sure exactly where that [02:52] is. He's really busy this week with some [02:53] other other deliverables. So, it looks [02:56] like it's getting close to being done. [02:58] Um, I'll follow up with him again next [02:59] week and see if we can get that one over [03:01] the line because I think that one is [03:03] probably I think if he spends a few more [03:05] minutes cleaning some stuff up, I think [03:07] that one's probably also pretty close to [03:09] ready to publish. [03:12] Uh, any questions on the practitioner [03:14] guides? [03:17] Okay. Uh, project updates. Um, [03:22] uh, Ernest, I'm glad you made it. You're [03:25] just in time for an update on the [03:27] foundation stats project. So, I'll turn [03:29] it over to you. [03:31] >> Yeah, I'm so sorry I'm late. [03:33] >> No worries. [03:35] >> So, um, we were able to kick off the [03:38] data collection and we have about four [03:40] contributors if I'm not mistaken that [03:43] has, um, added a few projects on the [03:47] Apache. We have a page that we are using [03:49] to track that in case everyone anyone is [03:52] interested in knowing what's the update [03:55] on that front. I think about um 30 [03:58] something project has been collected [04:00] now. So I hope we can increase that [04:03] momentum going forward. [04:05] So that's really been the focus. So if [04:08] we are like halfway through then I'll [04:10] start making sure the eclipse data is um [04:13] ready as well. I think we can script [04:15] that conveniently. It might just be from [04:18] different u um places. [04:23] >> Cool. Awesome. Thank you. Uh feel free [04:25] to add some links if you want to add [04:26] some links to the to the notes uh later. [04:30] Um any other projects? I've mostly I'll [04:33] be honest, I've mostly stopped working [04:35] on the orcs and relicensing project [04:38] because I feel like I feel like I've [04:40] done all of the research that was [04:42] interesting to me to do on that. And [04:44] it's been one of those projects that's [04:46] just been hard to get anybody else [04:47] involved. So I've just kind of I've kind [04:49] of stopped stopped working on that [04:51] mostly. Um so I don't really have an [04:54] update on on that project. I don't know [04:56] there uh potential for some other other [04:58] research projects. So we'll talk about [05:00] that uh later in the agenda. Um and I [05:03] also wanted to mention so uh Mary [05:05] Blessing since you're on the call uh [05:07] Mary Blessing and I talked a little bit [05:09] about um you know data science [05:12] contributions [05:13] and Mary Blessing I would I would [05:16] suggest that maybe you look at some of [05:18] the foundation stats work that uh Ernest [05:21] and and team have been working on [05:24] because that is a lot of um it's a lot [05:28] of tracking down data that's been a bit [05:30] more it's been a bit more manual ual, [05:32] but it might be a good way to kind of [05:34] learn some of the some of the processes [05:36] and get involved in get involved in a [05:38] project without necessarily having a [05:40] whole bunch of data science skills. [05:51] Okay. Um, [05:54] cool. anything [05:56] anything else [06:00] on data science projects? [06:05] Um, okay, cool. Uh, so Matt is here and [06:10] we've been we talked in one of the um [06:13] past data science meetings about some of [06:16] the commercial open source data that [06:17] Matt's been working on. So I don't know [06:19] Matt, you want to give us kind of an [06:20] update on where you are with that? [06:22] >> Yeah. Yeah, sure. So, so as you may [06:25] recall if you in the last meeting, um [06:27] I'm I founded an organization called the [06:29] commercial open source u startup [06:31] alliance and as part of that I have been [06:34] the recipient of uh data trobes that [06:37] have been donated to me. Uh two in [06:40] particular are the uh or what are [06:44] attempting to be uh a canonical list of [06:48] commercial open source companies. Um and [06:51] I have uh two lists. so far that have [06:53] been contributed to me. One was the one [06:55] that was the list that was used the 800 [06:57] companies that were used in the Linux [06:59] Foundation study, the state of [07:00] commercial open source and the other one [07:02] is a private list that um JJ at OSS [07:05] Capital has been maintaining and it's [07:07] another 800 uh or so companies. The [07:09] really interesting thing is there's very [07:10] little overlap except the obvious [07:12] MongoDBs and ODOS and things like that. [07:15] Um so I've I've been I have these data [07:18] and I have this data sitting in a coded [07:20] database right now. Um I and I expect to [07:25] get additional data contributions and [07:27] this year's state of commercial open [07:29] source um uh study that data will all be [07:32] public and so I've been looking for a [07:34] place to put it. Um Gar at Beturgia had [07:38] suggested that uh we start a working [07:40] group. I came to this group and they [07:43] said you don't need to start a new [07:44] working group just work with us. And I [07:48] think that's great. I I think this [07:50] starts as simple as Don's going to [07:51] create a folder in the repo and I'm [07:53] going to start uploading data and we'll [07:55] see. No expectations other than uh a [07:59] common place to hold it and maybe it [08:00] takes off into something interesting. [08:04] >> Uh yeah. So what what do we want to call [08:06] that folder? Um [08:09] should we just keep it short to like [08:12] coss? Do you want like commercial? [08:15] >> Let's start with cost. That's fine. [08:17] Yeah, I think we can that that's broad [08:19] enough that and different enough from [08:21] anything else that Chaos studies. [08:25] >> That sound good to everybody else. [08:27] >> I got a thumbs up from the train. [08:29] >> Okay. [08:34] And I'll do that under the under the [08:36] data sets area. [08:40] >> Cool. And then you can just do pull [08:42] requests to add the data and we can we [08:44] can merge those for you. [08:45] >> Okay. [08:46] Uh, anything else on the commercial open [08:48] source data? I'm excited to get to get [08:51] that started. [08:52] >> Awesome. [08:53] >> Because I was I was telling Matt, we we [08:55] met uh we talked yesterday actually. Um, [08:58] and I was I was telling him that I I [09:01] would be interested in maybe doing some [09:02] research based on some of the commercial [09:04] open source data because it's something [09:06] that I've been I've I've always been [09:08] really interested in how companies and [09:11] organizations are working in and around [09:14] open source projects. So, it's uh kind [09:16] of Yeah, I find it really interesting. [09:21] All right. Um [09:24] Okay. So, the next thing on the agenda, [09:27] uh Fosdam and Chaoscon recaps. Uh so, a [09:31] number of us were at Chaos Con and [09:33] Fosdom. So, I was just curious, uh what [09:37] what some of you thought, what were some [09:38] of your highlights, things that you [09:40] found interesting? [09:50] say overall for the people who were not [09:52] there um it was a pretty big um shift [09:57] from prior Chaos Cons and scale of we [10:01] had multiple tracks, we had lunch, we [10:04] had um like just different communities [10:07] having meetups, data meetups. I mean, it [10:10] kind of had all sorts of new things and [10:15] kind of I think there was a lot of [10:18] different swings that were that were [10:20] happening and a lot of different new [10:22] things and I think for the most part um [10:24] they landed really well. There's [10:26] definitely stuff to improve on the in [10:28] the next year, but um yeah, it was cool [10:32] to see how the event has grown. I don't [10:36] know the exact number of attendees, but [10:38] I think it was somewhere around 80. Um, [10:41] which was great. So, yeah. [10:44] >> Yeah. And it was about it was about what [10:46] we could fit in the space, uh, to be [10:48] honest. So, um, and it was really nice [10:51] having all of that all of that space to [10:53] be able to do breakouts and workshops [10:55] and buffs and things, which we haven't [10:58] been able to do as much of in the past. [10:59] So, I I really I really liked the new [11:02] venue and they seem to be interested in [11:04] doing it with us again. So, hopefully we [11:06] can make that work next year. [11:08] >> Yeah. I think it was nice to see the [11:10] audience was so diverse and like where [11:14] people like you really didn't have like [11:16] one I didn't feel like there was like [11:18] one like whether it be academia or [11:23] corporate or a specific like um like [11:26] community that wants to do analysis. It [11:27] just seemed like the audience was a huge [11:30] mix of people that were coming at this [11:32] in completely different ways, which I [11:35] think is what makes Chaos Con EU um a [11:39] special event. And so I'm glad that it c [11:42] like that was what I how what I took [11:44] away with last year and I think this [11:46] capitalized on like that diversity. [11:54] Any other thoughts on on Chaos Con? and [11:56] then maybe we can talk about Fostam. [12:02] We got a S. So S post posted in chat so [12:05] she's on a train. Uh she was impressed [12:07] by the quality of the contributors and [12:09] and I would 100% agree with that. I [12:11] think that we had a lot of a lot of [12:13] really interesting people working on [12:14] interesting things in the room and so I [12:16] think it was uh yeah it was great. I had [12:19] a great time. [12:22] Um cool. Uh and then you know other [12:26] other events past past kaaskon I went I [12:30] went to the um OF policy event so it's [12:33] the EU um basically the EU policy open [12:36] source policy event that the OF does [12:39] every year on Friday. Um I will admit so [12:42] so I I wrote a blog post on it. I'll [12:44] drop a link in the chat or something. [12:46] But I was really impressed with that [12:48] event this year. So last year I will [12:50] admit I was kind of bored to death in [12:52] that event. But I felt like I have to I [12:54] I feel like it's important for me to go [12:56] to some of these kind of events because [12:57] I can hear the conversations that policy [13:00] makers, governments, um you know, people [13:03] at non nonprofit organizations that [13:05] aren't uh tech nonprofits uh how they [13:09] think about things because I think [13:10] that's important context to have. And so [13:12] last year, all of the conversations were [13:14] like, uh, you know, maybe maybe we [13:17] should use open source. Maybe this is [13:19] good enough. Maybe, you know, it was it [13:21] was basically the conversations that [13:22] I've been having since 2000, uh, with [13:25] people about why open source is okay to [13:26] use. Um, so that was that was really [13:29] boring. And this year it completely [13:31] turned around. So this year all of the [13:33] conversations were really focused on [13:36] digital sovereignty and how open- source [13:38] needs to be a key part of that strategy. [13:41] And for that to work, they actually need [13:43] to support maintainers and support [13:45] projects and and discussions about how [13:48] how this is a this is a global effort [13:49] and you really kind of need to fund the [13:51] open source projects wherever that [13:52] happens. It's not just like let's just [13:54] fund the you know the EU folks and um [13:58] and it was really it was really [13:59] interesting. I had a really I really [14:01] enjoyed that event this year because I [14:02] feel like it was it was really just [14:04] night and day from the conversations [14:06] last year which I think was really [14:08] amazing progress in a year. So, I think [14:10] that EU is making good strides in in [14:11] that direction. [14:16] And then there was Oh, go ahead, Cali. [14:18] >> Oh, I was just going to say that I think [14:19] that's a huge topic in general is the [14:22] transition to digital sovereignty, [14:24] especially in Europe is a big [14:27] conversation and so it's cool to see [14:29] that go across. It's something that like [14:31] Red Hat it it's it's a huge thing for [14:34] Red Hat because Red Hat's a pro like [14:35] we're a platform company and like ho [14:38] like hosting servers and things is kind [14:41] of our our thing. Um so it's cool to see [14:46] that that is going on outside of just a [14:49] platform company. Um because it's going [14:53] to feel pretty it's going to feel very [14:56] important from our vantage point. But [14:58] I'm glad to see that that vantage point [14:59] is not just limited um there and that [15:03] people like for whatever reason like for [15:06] the reasons that that caused this [15:08] conversation um oh well but at least [15:11] that it's landing on an investment in [15:14] open source and that open source is a [15:16] pivotal portion of um sovereignty inc [15:19] because I mean I'm biased because I [15:21] that's kind of always been my big stick [15:24] on open source so of it being you you [15:27] know, if you ever needed to like reset [15:29] up a electric grid, I that's how I talk [15:31] about it with like people who are not [15:33] open source people, but I'm like, if we [15:35] ever needed to, you know, reestablish [15:37] like technical infrastructure for a [15:39] nation, um things being more and more in [15:42] the open source means that that lift is [15:45] easier. Um and so it's I'm glad to see [15:49] people seeing that before um you know [15:53] the major catastrophe that could happen [15:55] where that is intimately necessary. [15:58] >> Yeah. Yeah. Yeah. And it was really [16:00] interesting because so there's there's a [16:02] lot of mistrust of American um tech [16:04] companies uh in in Europe. And one of [16:07] the things so it's it's created it's [16:09] created kind of a crisis for for the EU [16:11] frankly and but it was interesting I [16:13] heard several people several of the [16:15] speakers mentioned um that you should [16:17] never waste a good crisis um and [16:19] basically using the you know the [16:22] geopolitical landscape as the the push [16:25] to do things that they wanted to do [16:26] anyways. Um, so that was that was kind [16:29] of interesting I thought. [16:32] And then there was some good stuff at [16:34] Fosdom too. There was a funding dev [16:36] room. I spent some time in that. Um, and [16:38] I thought some of the conversations [16:40] there were um were really interesting. [16:43] It's, you know, funding for open source [16:44] is one of those things that it's it's [16:46] not as easy as it sounds. You can't just [16:48] always throw money at things and make [16:49] them make them better. And so I thought [16:52] that that um came across pretty well in [16:54] the the funding dev room that it's it's [16:57] uh it's complicated. It's not as easy as [16:59] it sounds. Um but it's something that we [17:02] we really need to get better at. [17:06] >> Any other thoughts from Postown? [17:17] I was just looking [17:18] >> I just wanted Yeah, I was gonna like we [17:20] could pause for a second because I [17:21] wanted to read Sal because I'm very [17:22] curious about the the the CRA stuff is [17:25] such a a a big deal. Um [17:30] and yeah, because I actually do I think [17:32] that the intersection between the CRA [17:34] and the importance of community data [17:36] analysis is pretty big. I mean that's if [17:40] we're talk if we're going to talk about [17:42] what is going on within Red Hat and [17:45] regarding like where I perceive people's [17:49] perceived value of this work to be the [17:51] most from the CRA is a big portion of [17:54] that. Um, and I've always been I've been [17:58] pushing for a while and so as Gorg of [18:00] talking about okay, if we're going to if [18:02] CRA is is such a big thing or we're [18:04] trying to look at like what are the [18:05] upstreams that are going into products [18:07] if you're looking at it from a company [18:09] standpoint can like how does that [18:11] intertwine with sbombs or doesn't [18:13] intertwine with sbombs because sbombs [18:16] don't have a agreed upon format or have [18:19] um a link to an upstream repository and [18:22] how this all comes to be and so I [18:27] yeah I'm I'll be very curious to hear [18:29] more when you're not on a train s cuz [18:32] this is there's like I would say for [18:35] like our team like the three biggest [18:37] like kind of pillars for what we're [18:39] supposed to be doing this year one of [18:40] the sides portions of it is like it's [18:42] generally quote unquote security [18:44] analysis but it has a lot to do with CRA [18:47] um I would say that I think it's a [18:49] pretty good signal that like Red Hat [18:51] thinks that it's a big deal whenever I [18:53] think there's going to be five people [18:54] added on to my um add to our team with [18:56] these specific goal or like specific [18:59] work around CRA. My team has 30 people. [19:02] So that's a that's a lot of people for a [19:06] small team to be focused just on CR that [19:08] CRA and so I think [19:11] yeah it's a big it's a big deal. [19:16] >> Yeah, for sure. Um, and as S pointed out [19:18] in chat, the foundation steps project [19:21] um, can really kind of give people a [19:23] baseline as we start to measure um, [19:26] measure improvement as the the C the CRA [19:29] gets implemented and people start to [19:31] actually learn what that means. [19:35] Cool. [19:37] Um, okay. Anything else on WDM ka [19:42] related related events? [19:50] Okay. Uh, visualization notebooks. [19:54] Yeah. I just wanted to start throwing [19:56] this conversation out here as with the [20:00] move over of all of the resources from [20:03] Repel and the archival of Repel and all [20:06] of the notebooks that we've moved over [20:09] for some context or people aren't as [20:10] familiar. These notebooks um go to [20:14] connect to an augur database and to do [20:17] analysis. Um the notebooks that were are [20:20] kind of the most active or what I see to [20:22] be the the like what I'm trying to try [20:25] to push some focus onto is notebooks [20:29] around making visualizations that would [20:31] eventually go into 8. So it's using [20:34] augur data doing the pre-processing in [20:36] pandis and then building a plotly [20:40] visualization and so we're starting um [20:44] kind of the pipeline from idea to [20:48] visualization in 8 um which this is I'm [20:52] saying this verbally this is what we're [20:54] my team is trying to work on to like [20:56] really for this isn't just it's a [20:58] checklist and this checklist will go [21:01] between 8 knot and this working group [21:03] and So that's why I wanted to start [21:04] bringing this up where right now a lot [21:07] of them are actually starting off with [21:09] chaos metrics models are the concept is [21:12] a chaosmetric model. The first step is [21:15] that how do you take that metric model [21:17] and put it in like kind of the vantage [21:20] point of something like 8 knot or [21:22] looking at how would you make a [21:23] visualization using augur data whichever [21:26] lens you would like to look at it from. [21:29] Um, I've kind of I do it from I think [21:32] about it from 8 knot because I know [21:33] that's the end goal of of the [21:35] visualizations that I'm working on, but [21:37] I don't think that that has to be the [21:39] limiting scope for some of these [21:41] visualizations. And then the first step [21:44] of that would be to build it in a [21:45] notebook that would be contributed to [21:48] this repository and to the folder we [21:51] have a folder for 8 knot visualization [21:54] notebooks. And then the ne the final [21:56] step would be to take that [21:58] pre-processing and then be put into 8 [22:01] knot. Um so what we're starting to look [22:05] at is on the 8 knot side having a more [22:09] templated approach of here's a [22:11] visualization issue. Here's the check [22:12] marks of what it is. And one of those is [22:15] going to be a pro a hyperlink to an [22:17] issue in this in this repository that is [22:20] just making said notebook. making the [22:24] visualization into the notebook. And so [22:26] if there are new contributors that are [22:28] coming in um that want to do some like [22:31] some analysis work, make a visualization [22:34] work with data. I think this is going to [22:35] be a good opportunity for people to do [22:37] that. And we are working on getting a [22:41] very small augur instance to be able to [22:44] be accessible for people to work on this [22:47] and work on eight knot. It would be like [22:49] pretty much the bare bones version of [22:51] augur. So like 10 repositories but that [22:54] is enough to make a to be able to build [22:58] these visualizations because you can [22:59] have that structure to be able to run 8 [23:01] knot locally. Um so [23:06] I don't have any like big action items. [23:08] I'm one more I'm saying this partially [23:10] to keep myself accountable to start [23:13] working towards making these issues um [23:16] because I really want this to be a [23:17] pretty [23:19] um just templated approach. So then if [23:23] people and because I think that the [23:24] visualization size side is something [23:27] that there's a lot of people that are [23:28] coming into this working group looking [23:30] for something tangible to work on and I [23:34] feel like this is a really good scope [23:36] for people who are trying to learn about [23:39] data analysis. [23:41] Um and so that's that's the vision. [23:46] >> Yeah, absolutely. and and to just second [23:48] that it really um it really is a good um [23:52] a good learning experience I think um to [23:55] work on this because the way the way it [23:57] not works I mean it uses dashplotly [23:59] which is one of the most common like [24:03] Python visualization sort of frameworks [24:06] um so the skills that you'd learn by [24:07] doing this would be Python Jupyter [24:09] notebooks [24:11] um [24:12] >> pandas pre-processing pandas um are [24:15] really a lot of the skills the data [24:17] scientists need. So, I think this is a [24:19] great project to get people involved in. [24:23] >> Yes. And I'm hoping Yeah, I have I have [24:25] big hopes. This was kind of this is this [24:28] was a part of the reason why I wanted to [24:31] move the notebooks over here. Um, and to [24:35] start this is kind of what I this is [24:36] what I've been trying to work towards [24:38] and seeing people come in with interest [24:41] of working on data science work. I think [24:43] that and I think it also um ties [24:46] together different portions of chaos [24:48] because really any and we'll figure I [24:51] probably will be asking for some help on [24:53] maybe documentation or like just because [24:57] like I'm going to open up issues for all [24:59] the visualizations that have been like [25:01] opened up on 8not but that's not like a [25:04] limiting scope. Um, if somebody wants to [25:06] make a metric or visualization based off [25:09] of a metric model or based off of [25:11] anything from the rest of chaos, then [25:12] that's fantastic. Or if they have an [25:14] idea and they want to build and build [25:17] from it, these are a set of starting [25:19] points, but it's not a limited. It [25:22] doesn't mean that those are the only [25:23] things. And as this develops, these are [25:27] the type of things that I would like to [25:29] to document well for people so they can [25:32] come in and work on what they want to [25:35] work on. I think most people will [25:36] probably want to start with a [25:38] visualization that has already has a [25:41] semi-defined scope because that's an [25:43] easier um thing to to chew on for the [25:47] first time. Um but there is so many [25:52] opport I think there's a lot of [25:53] opportunities here um with what you can [25:56] do with augur data um and trying to [26:00] build within the visualizations and I [26:02] think that it is worth a discussion of [26:04] whether in these notebooks or if we [26:07] would want a different notebook if [26:08] people are like I really want to use [26:09] mattplot liib I'm like I don't have a [26:14] for sure answer for that um I would [26:17] prefer people to use plotly because [26:19] that's what integrates directly within [26:21] Dash and the interactive nature of the [26:24] plotly visualizations is incredible in [26:29] comparison to Mattplot Lib. Um it's like [26:32] if you're doing a report mapplot like [26:35] the differences between I could see why [26:37] people like a static report I can see I [26:40] mean also like mapplot lib is like I [26:42] think the [26:44] goto it's what people think of when they [26:46] think of um um uh python data [26:50] visualizations um I think that um [26:54] dashplotly is kind of a far second but [26:57] it is so much more interactive even in a [27:00] jupyter notebook I don't know if people [27:01] have ever run draft plotly in a Jupyter [27:03] notebook, but you get the interactive [27:05] nature of the graphs in the notebook as [27:08] well as like so it actually you're like [27:12] what you see in 8 knot we that's nothing [27:15] all the interactive nature of those [27:16] graphs we don't do that if it's like a [27:19] toggle like that's like outside the [27:20] graph then that's us but all of the [27:23] things that make it to where you can [27:24] select and deselect things that's built [27:27] in with a plotly graph and that's also [27:29] the case if you just run one of those [27:31] graphs in a notebook. Um, and so that's [27:34] why I have drawn further and further [27:36] towards Dashplot is because especially [27:40] like yesterday I went to go investigate [27:42] some what I think might be some data in [27:44] accuracy or like with how our [27:45] pre-processing is working with some of [27:47] our graphs and because I could do so [27:49] much interactive nature in the graph [27:51] itself, I could investigate that fully [27:52] without having to do a single query or [27:56] write a line of code. Um, so [28:00] >> that's that's super cool. Um, okay. So, [28:03] as you were talking about this, um, it [28:05] occurred to me that we should probably [28:06] start tracking this notebook [28:08] visualizations as an actual project. Um, [28:12] so we should also probably do if you if [28:14] you have some time this week is create [28:16] an issue. um um and just create an issue [28:21] that we that we're working on this this [28:23] as a project and encourage people to [28:27] contribute to it. Um, so if you could [28:30] put some things in the issue about um [28:32] where people can go to contribute that [28:35] would be helpful because what we found [28:37] and I think you know Ernest and Sal can [28:39] can second this is like having it as an [28:41] issue where we can we can point people [28:43] and say you know hey if you're [28:45] interested in contributing [28:47] here's an issue with some details. I [28:48] think that's helped a lot with the um [28:51] foundation stats project. Okay, I have a [28:54] quick question because I don't know my [28:57] thought would be that I would be making [28:59] the issues for the each of the [29:01] individual like visual for context about [29:03] like 10 visualization issues at least [29:06] and eight not and my thought is that I [29:08] would be I'd be opening an issue for [29:10] each of those visualizations [29:12] in the data science working group so [29:14] then people could self assign and so [29:17] would you think that you wanted the [29:19] overarching issue before those sub [29:21] issues or because my brain works from [29:24] sub issues and then creating the larger [29:26] one because I don't because if I create [29:28] the larger one and they're like where's [29:29] the information? I'm like I haven't [29:31] gotten there yet. Um [29:34] >> I would say do it do it in whatever [29:36] order is most comfortable for you. Um I [29:38] would just make sure that they all get [29:39] labeled as project and then when you [29:42] create you can create it like you can [29:43] call it umbrella issue or something like [29:45] that. But I think we have another one [29:46] that's called something similar. Um [29:48] where you have like an overarching issue [29:50] that maybe points to some of the other [29:52] the other issues. Um or if you create [29:54] some of the issues in in 8 knot, you [29:57] could create an umbrella issue in the [29:59] working group and then point to some of [30:02] the things that people can work on in [30:04] the [30:04] >> that's my plan. Okay. is that each [30:07] visual so it' be like a check pretty [30:09] much like a to-do list for the eight [30:12] knot one because there's like steps that [30:14] you need to go through um yeah recommend [30:18] okay [30:19] >> yeah yeah I think I think what S [30:21] mentioned in chat looks [30:23] >> um and so point S makes a good point [30:26] these master issues are really great for [30:28] orienting people [30:29] >> and I think that's going to be the stuff [30:31] that I'll probably want to get some [30:32] feedback on because I'm too close to [30:36] this and have been working on some of [30:38] these things for way too long to really [30:40] like understand the vantage point of a [30:44] firsttime person. Luckily, um, Kyle on [30:48] our team is supposed to be working on a [30:49] new visualization and kind of going [30:51] through this entire cycle. And so, I'm [30:53] hoping through that we kind of start to [30:55] learn even though he's worked on the [30:57] software engineering side, he hasn't [30:59] done any of like the data visualization [31:01] work. And so he's still somewhat of a [31:04] new vision into this. The only thing [31:06] that I'm the other thing that I'm a [31:08] little bit worried about is like I is [31:10] the timeline of us like I know we want [31:13] to have a a um augur instance that is [31:17] available but whether that lands in our [31:20] and like the Red Hat infrastructure or [31:22] Sean's infrastructure. And so that's [31:24] something that I'm wanting to be careful [31:27] about of [31:29] I want to I know that if I put these [31:31] things out there that it's going to get [31:33] attention and I want to make sure that [31:34] it's ready for attention. Um [31:37] >> yeah, [31:38] >> fair. [31:39] >> And so that's like those are that's [31:40] where all of my my brain is going. So [31:44] I'll probably not create the like larger [31:46] like getting started on this issue until [31:49] at the very least the kind of templated [31:52] that we have it a little bit more like [31:54] hey here's the to-dos from the eight [31:56] knot side because I also want it to be [31:58] clear that if people just want to make a [32:00] visualization notebook and that's all [32:01] that they want to do that's fantastic. [32:04] somebody else can take over on the um on [32:08] the um like eight knot side super easily [32:11] if they want to be like that's not the [32:13] that's kind of the purpose of the [32:15] breakup is that it makes it to where you [32:17] can think about it from a data [32:18] prep-processing and visualization [32:20] standpoint and you can think about it [32:21] from an 8not standpoint. The needs for [32:24] the augur instance is just a bare bones [32:27] the same that what we would need for 8 [32:29] knot for a contributor to be able to [32:32] just like a startup. We want to have a 8 [32:36] knot instance with a couple of [32:38] repositories that just has collection [32:40] across the board. So then when they [32:43] query to different t to a table for a [32:46] repository there is data. Um, so it's [32:48] like the smallest I think that out of [32:51] the box augur real like maybe one other [32:54] repository if you want to have like a [32:55] more like populated one like maybe like [32:57] Rook Rook or just some medium-sized [33:00] repository, one of them would probably [33:02] be good just to have something that is [33:04] more representative of a larger um a [33:08] larger project. But yeah, if you have [33:10] any other questions, Adrian, please ask. [33:13] Um, but talking bear [33:18] Yeah, because that's my hope is that the [33:20] one for the working group and the one [33:22] for like we just we don't need multiple [33:24] for the different repositories. We just [33:26] need one so then new contributors have a [33:29] way to start working on these things. [33:33] >> Cool. And there is um so what I what I [33:37] might do there there is a um data [33:41] science project proposal um template for [33:44] an issue that's super high level that [33:46] basically just says we have this [33:47] project. Um do you want me to just uh [33:52] create something using that and just [33:55] point people to the directory and say we [33:57] have you know more details coming or do [34:00] you want [34:00] >> more details coming is great. Um, [34:02] anything [34:03] >> and then you can you can update it [34:04] whenever you're ready and you can add [34:05] whatever details, point people to [34:08] whatever. [34:09] >> Yeah, I probably am going to put I was [34:11] going to see with because I was looking [34:13] at the labels um in the notebook and so [34:16] you have it under project. I don't we [34:19] can kind of figure out in time if a more [34:22] specific like project like visualiz like [34:25] a visualization label would be useful [34:28] but I want to I was going to maybe [34:31] propose a block like uh I don't know if [34:34] a block is the right word but I I I was [34:37] thinking about if as I'm opening these [34:40] smaller issues to be able to signal to [34:42] people that they're like not that like [34:44] it's a maybe like a work in like cuz I [34:46] don't even trying to be able to [34:48] verbalize people like, "Hey, I'm still [34:50] working on this." I'm Yeah, cuz I'm [34:52] concerned about people getting which I'm [34:54] I want people to get excited and want to [34:56] work on it, but it's going to take me a [34:58] little bit to get everything set up for [35:00] the first time. Um, and I want to be [35:03] able to signal to people that this is [35:05] coming and we're working on it. It just [35:07] doesn't have all the details yet. [35:09] >> Yeah. Okay, cool. [35:13] Awesome. Okay, I will I will do that. [35:16] I'll create I'll create that just the [35:18] basic thing and then [35:20] uh yeah, we can create more labels. We [35:22] can yeah whatever whatever makes sense [35:24] for you. [35:25] >> Okay. Yeah, I'm going to think about the [35:27] right like wording because my hope is [35:29] that it's something that is concise [35:31] enough to like kind of get I guess I can [35:32] just do work in progress and do like the [35:34] sub the subdetails. I mean like this [35:36] issue is still getting the the details [35:38] necessary for somebody to be able to [35:40] work on it. [35:41] >> Um I might do that. [35:50] Okay, cool. Uh, anything else on the [35:53] visualization notebooks? [36:02] Okay, cool. Uh, well, we only have about [36:05] 10 more minutes. So, um, are there other [36:10] things that people want to put on the [36:11] agenda for, [36:13] uh, for future [36:17] future meetings? [36:19] >> I would say probably in um, I guess next [36:23] meeting, I'm out of office, but um, kind [36:25] of in the future, I'd say for me, I [36:27] probably am going to have updates kind [36:30] of indefs [36:32] under the project. I think I will have [36:33] updates most likely indefinitely around [36:36] the visualizations [36:38] um and more of just like hey this is [36:40] where we're at getting feedback from [36:43] people of hey was this a like a system [36:46] that was followable like what details [36:48] that you need looking for a lot of [36:50] feedback on just docs I written [36:53] communication is not my specialty [36:59] >> yeah fair enough um okay cool yeah that [37:02] sounds Good. We'll just we'll add that [37:03] to the project updates section. I also [37:06] uh Matt, just a note, um I think we [37:08] should treat the cost stuff the same [37:09] way. Um and just talk about it as a kind [37:13] of as a project within the data science [37:16] uh working group. And so we'll just do [37:17] updates on it in the in the meetings if [37:20] you have an update. If you don't, that's [37:21] that's cool, too. And then I'll I'll [37:24] also create a an umbrella issue like a [37:28] project proposal issue as well that will [37:31] point people to the directory that I uh [37:33] that I create. And so then you can [37:35] update that [37:37] as needed with more details [37:40] >> like like drop a read me in the in the [37:42] the folder. [37:44] >> Yeah. Yeah. I'll I'll create an issue um [37:47] and I'll create the folder and then I'll [37:50] in the issue I'll link to the folder. [37:51] Um, but then you can do PRs against that [37:54] folder and add like a read me, add the [37:56] data set, add add whatever you whatever [37:59] you think you need. [38:00] >> Okay. [38:02] But then we'll we'll just put it on the [38:03] agenda every week as a project that we [38:05] talk about so that people know that they [38:06] can know that they can get involved and [38:09] that'll be a great way if we you know if [38:11] there are some so you know one of the [38:13] things the foundation stats project has [38:15] been doing is basically filling in [38:16] missing data about Apache projects and a [38:19] lot of that is is actually kind of [38:21] manually going out and finding that [38:22] missing data. So, you know, if we [38:25] realize that there's some some, you [38:28] know, data that we're missing as a part [38:29] of the cost data set, um, we can create [38:32] specific issues and maybe get people [38:34] more involved in in that project, [38:36] helping us track down, you know, missing [38:38] data. [38:39] >> Yeah, that's a really good point. [38:41] >> Yeah, I'd say across the board, I'm [38:42] noticing that the population like [38:47] putting good details into your issues. I [38:49] don't really know how people are finding [38:52] repositories at such a rap and like [38:54] issues at such a rapid pace because I'm [38:56] seeing this like across not just here [38:58] not just in augur but in a lot of [39:00] different repositories [39:02] um clear scoped issues are getting [39:05] worked on like whether it's a yeah [39:08] whether it's but even not even good [39:09] first things that aren't labeled good [39:11] first issues are us are getting worked [39:13] on even not not just good first issue so [39:17] um the time spent putting a little bit [39:19] of extra detail I think is really coming [39:22] to fruition of people outside [39:24] contributors wanting to work on it. [39:30] >> Cool. Awesome. And I saw Sal's note in [39:33] chat about adding something about uhers [39:36] and um oning sponsorship. [39:41] So we'll talk about that in the in the [39:43] next meeting. Should be good. Um, the [39:47] only reminder that I have, if other [39:49] people have reminders, feel free to add [39:50] them there. The, uh, CFP for Open Source [39:54] Summit North America was extended to [39:56] February 15th. So, if you haven't gotten [39:58] your talk proposals in and you would [39:59] like to go to, um, Minneapolis, [40:03] um, feel free to submit a a talk [40:06] proposal. [40:09] We'd love to have talks about chaos data [40:11] at these conferences. So, [40:14] okay. Uh, cool. Anything Anything else [40:17] before we wrap it up in our last five [40:20] minutes? [40:22] >> Nope. [40:23] >> Looking good. [40:25] >> Okay, cool. Well, I give everybody five [40:27] minutes back and it was great chatting [40:29] with you all and take care. [40:32] Bye, everybody.  📄 **Full text:** ![[Full Document Contents/Articles/CHAOSS Data Science Working Group, Feb 10, 2026.md|CHAOSS Data Science Working Group, Feb 10, 2026 by YouTube]] ## Highlights - on data science projects? Um, okay, cool. Uh, so Matt is here and we've been we talked in one of the um past data science meetings about some of the commercial open source data that Matt's been working on. So I don't know Matt, you want to give us kind of an update on where you are with that? >> Yeah. Yeah, sure. So, so as you may recall if you in the last meeting, um I'm I founded an organization called the commercial open source u startup ([View Highlight](https://read.readwise.io/read/01kpb40yneyd6v45b6eekg5fvb)) - Tags: 👻 ai highlighted - would be interested in maybe doing some research based on some of the commercial open source data because it's something that I've been I've I've always been really interested in how companies and organizations are working in and around open source projects. So, it's uh kind of Yeah, I find it really interesting. All right. Um Okay. So, the next thing on the agenda, uh Fosdam and Chaoscon recaps. Uh so, a ([View Highlight](https://read.readwise.io/read/01kpb40yp32m01zaa27n159z2n)) - Tags: 👻 ai highlighted - some analysis work, make a visualization work with data. I think this is going to be a good opportunity for people to do that. And we are working on getting a very small augur instance to be able to be accessible for people to work on this and work on eight knot. It would be like pretty much the bare bones version of augur. So like 10 repositories but that is enough to make a to be able to build these visualizations because you can have that structure to be able to run 8 ([View Highlight](https://read.readwise.io/read/01kpb40ysmwtq9cg1negv7ee5m)) - Tags: 👻 ai highlighted - move the notebooks over here. Um, and to start this is kind of what I this is what I've been trying to work towards and seeing people come in with interest of working on data science work. I think that and I think it also um ties together different portions of chaos because really any and we'll figure I probably will be asking for some help on maybe documentation or like just because like I'm going to open up issues for all the visualizations that have been like ([View Highlight](https://read.readwise.io/read/01kpb40yw8417sttc57wxyp5dt)) - Tags: 👻 ai highlighted - come in and work on what they want to work on. I think most people will probably want to start with a visualization that has already has a semi-defined scope because that's an easier um thing to to chew on for the first time. Um but there is so many opport I think there's a lot of opportunities here um with what you can do with augur data um and trying to ([View Highlight](https://read.readwise.io/read/01kpb40yyy47cws7wp99xz2gre)) - Tags: 👻 ai highlighted - sounds Good. We'll just we'll add that to the project updates section. I also uh Matt, just a note, um I think we should treat the cost stuff the same way. Um and just talk about it as a kind of as a project within the data science uh working group. And so we'll just do updates on it in the in the meetings if you have an update. If you don't, that's that's cool, too. And then I'll I'll also create a an umbrella issue like a project proposal issue as well that will ([View Highlight](https://read.readwise.io/read/01kpb40z6et89y4fs6e5xh5esk)) - Tags: 👻 ai highlighted