CHAOSS OSPO WG, April 2 2026 - CourtneyR.dev's 2nd Brain

CHAOSS OSPO WG, April 2 2026 ![rw-book-cover](https://img.youtube.com/vi/6RXrSMM5OLg/maxresdefault.jpg) 🔗 **Source:** [CHAOSS OSPO WG, April 2 2026](https://www.youtube.com/watch?v=6RXrSMM5OLg) > [!note]+ Document Notes > tags: #chaos-metrics #open-source #anaconda #ospo #community-assessment #narrative-reporting > author: Travis Hathaway > source: CHAOSS OSPO Working Group > reference: https://www.youtube.com/watch?v=6RXrSMM5OLg > date created: 2024-12-19 19:00 > > # 🏷️ Title: Using CHAOSS Metrics at Anaconda for Open Source Project Health Assessment >  ## 🧰 Action Inbox - [H1] Develop narrative-driven open source reports combining quantitative metrics with maintainer interviews — 15:45 #storytelling - [H2] Evaluate CHAOSS Community Activity metrics for comprehensive project health assessment — 18:30 #metrics-implementation - [H3] Create standardized questions for annual open source project reviews — 28:15 #process-standardization ## 🧭 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 - [CHAOSS OSPO Working Group](https://chaoss.community/) - [LFX Analytics](https://lfx.linuxfoundation.org/) - [OpenHub](https://www.openhub.net/) ## 📝 Summary - Travis Hathaway from Anaconda presented their approach to measuring open source project health using CHAOSS metrics - Anaconda supports major Python ecosystem projects like Jupyter, PyScript, Bokeh, and Conda through staffing and NumFOCUS sponsorship - The company is developing narrative-driven reporting that combines quantitative metrics with qualitative maintainer interviews - Key challenge addressed: moving beyond sterile statistics dashboards toward engaging storytelling about project health - The presentation showcased early prototypes and discussed the "Cambrian explosion" of open source metrics tools ## 📖 Main Content Travis Hathaway, a software engineer at Anaconda working primarily on the Conda package manager, shared his 10% time project focused on assessing the health of Anaconda's outbound open source projects. His background in sustainability and social sciences brings a unique perspective to open source metrics, emphasizing the social aspects of project health assessment. Anaconda, founded in the early 2010s as Continuum Analytics by Travis Oliphant and Peter Wang, emerged to solve installation and packaging problems in scientific computing. The company now employs 300-400 people globally and maintains significant open source projects while sponsoring NumFOCUS, a nonprofit supporting scientific computing tools. The presentation demonstrated Hathaway's evolution from building custom metrics dashboards to leveraging CHAOSS metrics for more structured assessment. His early prototype dashboard tracked basic GitHub metrics but was superseded by more sophisticated tools like LFX Analytics. What differentiates CHAOSS metrics is their user-story-driven approach, particularly the Community Activity metrics that begin with "why it matters" and specific user scenarios. The innovative aspect of Anaconda's approach lies in combining quantitative data with qualitative storytelling. Rather than presenting sterile statistics, they plan to interview project maintainers to create narratives around data spikes and trends, producing engaging annual reports that explain the human stories behind the metrics. ## 🗒️ Notes #### ⏱️ [00:00](https://youtu.be/6RXrSMM5OLg?t=0) — [H1] Introduction and Background - Key idea: Travis Hathaway introduces his work on open source metrics at Anaconda - Evidence: "I'm going to talk about using CHAOSS at Anaconda" #### ⏱️ [02:15](https://youtu.be/6RXrSMM5OLg?t=135) — [H2] Personal Background and Interest - Key idea: Social science background drives interest in open source community metrics - Evidence: "Studies program was sustainability, society and the environment... it kind of involves both aspects of social science but then also technology" #### ⏱️ [04:30](https://youtu.be/6RXrSMM5OLg?t=270) — [H3] Conda Package Manager Context - Key idea: Day-to-day work involves managing an open source ecosystem beyond just packaging - Evidence: "It's an open-source ecosystem of packaging... we have our own sort of philosophy towards packaging" #### ⏱️ [06:45](https://youtu.be/6RXrSMM5OLg?t=405) — [H1] Anaconda's Open Source Foundation - Key idea: Anaconda was founded to solve scientific computing installation problems - Evidence: "Back then the sort of act of just getting stuff installing and running on your computer was actually quite difficult" #### ⏱️ [09:00](https://youtu.be/6RXrSMM5OLg?t=540) — [H2] NumFOCUS Partnership and Project Support - Key idea: Anaconda supports open source through staffing projects and sponsoring NumFOCUS - Evidence: "We support open-source development by basically staffing projects... and Anaconda is heavily involved in sponsoring this organization" #### ⏱️ [11:30](https://youtu.be/6RXrSMM5OLg?t=690) — [H3] Early Metrics Dashboard Prototype - Key idea: Initial custom dashboard provided basic GitHub metrics but was limited - Evidence: "This was merely just a prototype... measuring things like activity on the project and then also just the number of issues and pull requests" #### ⏱️ [15:45](https://youtu.be/6RXrSMM5OLg?t=945) — [H1] CHAOSS Metrics Differentiation - Key idea: CHAOSS metrics excel through user-story-driven approach versus pure statistics - Evidence: "What I really enjoyed was the fact that there it begins with sort of a you know a why it matters and a user stories" #### ⏱️ [18:30](https://youtu.be/6RXrSMM5OLg?t=1110) — [H2] Community Activity Metrics Implementation - Key idea: Focus on CHAOSS Community Activity metrics for comprehensive assessment - Evidence: "I've decided that it would be really great for us to take the metrics in the community activity" #### ⏱️ [21:00](https://youtu.be/6RXrSMM5OLg?t=1260) — [H3] New CLI Tool Development - Key idea: Building organization-focused tool on top of existing CHAOSS infrastructure - Evidence: "This is a CLI tool... where we're mostly interested in gathering and generating reports for entire organizations on GitHub" #### ⏱️ [24:15](https://youtu.be/6RXrSMM5OLg?t=1455) — [H1] Narrative-Driven Reporting Vision - Key idea: Combining quantitative metrics with qualitative maintainer interviews for storytelling - Evidence: "Have a format where yes, there are statistics, but there are stories behind these statistics" #### ⏱️ [28:15](https://youtu.be/6RXrSMM5OLg?t=1695) — [H2] Annual Review and Standardization Plans - Key idea: Creating standardized questions and processes for annual project reviews - Evidence: "Maybe even standardize some of the questions that I'm asking as far as like, how is this year and the software" #### ⏱️ [30:45](https://youtu.be/6RXrSMM5OLg?t=1845) — [H3] Tools Proliferation Challenge - Key idea: Rapid growth in open source metrics tools creates navigation challenges - Evidence: "We're in this Cambrian explosion of tools... there's so much stuff out there" #### ⏱️ [45:30](https://youtu.be/6RXrSMM5OLg?t=2730) — [H2] Community Guild Structure - Key idea: Cross-project coordination through community guild enables knowledge sharing - Evidence: "Meta group of all the different sort of open source efforts at the company" #### ⏱️ [52:00](https://youtu.be/6RXrSMM5OLg?t=3120) — [H3] Academic Research Computing Evolution - Key idea: Research computing sector transformation due to AI and LLM requirements - Evidence: "Once the LLMs hit it's like oh yeah where do we have the gigantic computing resources that have GPUs" ## 📚 References - [CHAOSS Community Activity Metrics](https://chaoss.community/metric-community-activity/) - [LFX Analytics Dashboard](https://lfx.linuxfoundation.org/) - [OpenHub Project Analysis](https://www.openhub.net/) - [NumFOCUS Organization](https://numfocus.org/) - [CMS OSPO Metrics Reports](https://github.com/CMS-Enterprise/oss-community-metrics) ## 🗣️ Paraphrase - Anaconda is developing a narrative-driven approach to open source project health assessment that combines CHAOSS metrics with maintainer interviews to tell engaging stories about project evolution rather than presenting sterile statistics dashboards - The company's community guild structure enables cross-project coordination and knowledge sharing among different open source initiatives, representing a mature organizational pattern for managing multiple community projects - The proliferation of open source metrics tools creates both opportunities and challenges, requiring organizations to navigate an expanding ecosystem while focusing on tools that provide user-story-driven insights rather than just data visualization ## 📰 Newsletter **Anaconda Pioneers Narrative-Driven Open Source Health Reporting** Anaconda is taking an innovative approach to open source project assessment by combining CHAOSS Community Activity metrics with qualitative storytelling. Rather than relying solely on statistics dashboards, they're planning to interview project maintainers to create engaging annual reports that explain the human stories behind data trends. This approach addresses a key challenge in the open source metrics space: how do we move beyond sterile data presentations toward meaningful insights that engage stakeholders and tell compelling project stories? The company's community guild structure also offers a valuable model for organizations managing multiple open source projects, enabling cross-project coordination and knowledge sharing among maintainers and community leaders. **Key Takeaway**: The future of open source metrics lies not just in better data collection, but in better storytelling that combines quantitative insights with qualitative human experiences. Resources: [CHAOSS Metrics](https://chaoss.community/), [LFX Analytics](https://lfx.linuxfoundation.org/), [NumFOCUS](https://numfocus.org/) ## ✅ Action Items #### DevRel - Developer advocacy: Create narrative templates for project health storytelling combining metrics with maintainer interviews - Education: Develop workshops on CHAOSS metrics implementation for community managers and OSPOs - Developer experience: Build CLI tools that generate organization-level reports from CHAOSS metrics - Community management: Establish standardized questions for annual project health reviews #### OSPO - Community Engagement: Implement cross-project coordination through community guilds or shepherds groups - Strategic Alignment: Develop KPIs that balance quantitative metrics with qualitative project narratives - Software Inventory: Create comprehensive dashboards tracking health across portfolio of supported projects - Developer Support: Provide training on narrative-driven reporting approaches for project maintainers #### Community - Maintainers: Develop interview templates for annual project health reviews and success story documentation - Contributors: Create onboarding materials that explain project health metrics and community assessment approaches - Ecosystem: Establish coordination mechanisms between related projects within larger organizational umbrellas #### WordPress - Community: Implement narrative-driven reporting for WordCamp and meetup impact assessment - Training: Create lesson plans teaching community health assessment using CHAOSS metrics - Meta: Develop organization-level health reporting for WordPress.org project ecosystem - Documentation: Create guides for combining quantitative metrics with qualitative community stories #### The WPCC - Programs: Develop metrics and storytelling approaches for fellowship and grant impact reporting - Projects: Create templates for project health assessment that combine data with maintainer narratives ## 🧪 Hypotheses - Organizations that combine quantitative metrics with qualitative storytelling will achieve higher stakeholder engagement than those using statistics-only approaches - Community guilds or cross-project coordination structures improve knowledge sharing and reduce duplicated effort in multi-project organizations - User-story-driven metrics frameworks like CHAOSS provide more actionable insights than generic analytics dashboards - The proliferation of open source metrics tools requires curation and standardization to prevent decision paralysis ### 📊 Methodology - Survey comparison between narrative vs. statistics-only reporting engagement rates - Case study analysis of organizations with and without cross-project coordination structures - Longitudinal study tracking CHAOSS metrics adoption versus custom dashboard usage - Tool landscape analysis measuring proliferation rates and adoption patterns ### ❓ Questions - How can organizations systematically balance quantitative metrics collection with qualitative storytelling efforts? - What organizational structures best support coordination across multiple open source project communities? - How do we prevent metrics tool proliferation from creating decision paralysis for OSPO practitioners? ### 🔮 Future Research - Development of standardized narrative frameworks for open source project health reporting - Investigation of optimal organizational structures for managing multi-project open source portfolios - Analysis of LLM applications for automating aspects of project health assessment and storytelling ### 💡 Implications - Practice: Move beyond dashboard creation toward integrated storytelling approaches that engage stakeholders - Policy: Develop guidelines for balancing automation with human insight in community health assessment - Education: Create training programs teaching narrative-driven metrics approaches to community managers - Funding: Allocate resources for qualitative research alongside quantitative metrics collection ### 🧠 Learning Styles Assessment Reading level estimate: Flesch-Kincaid Grade Level 12 #### 👀 Visual learners - Present: Dashboard screenshots, metrics visualizations, organizational structure diagrams - Improve: Process flow diagrams showing narrative integration with metrics collection #### 🔊 Aural learners - Present: Full video presentation with discussion and Q&A - Improve: Podcast-style summary highlighting key narrative approaches #### 🗣️ Verbal learners - Present: Clear definitions of CHAOSS metrics, community guilds, and narrative reporting - Improve: Glossary of open source metrics terminology and storytelling frameworks #### 🧩 Kinesthetic learners - Present: CLI tool demonstrations, hands-on dashboard examples - Improve: Step-by-step guides for implementing CHAOSS metrics and interview processes #### 🧠 Logical learners - Present: Comparison between different metrics approaches, organizational structures - Improve: Decision trees for selecting appropriate metrics tools and reporting formats #### 🧑‍🤝‍🧑 Social learners - Present: Working group discussion format with multiple perspectives and experiences - Improve: Facilitated workshops for implementing narrative-driven metrics approaches #### 🧘‍♀️ Solitary learners - Present: Self-guided exploration of metrics tools and frameworks - Improve: Reflection prompts for evaluating current organizational metrics approaches #### 🌿 Naturalistic learners - Present: Ecosystem thinking about project relationships and organizational environments - Improve: Analogies connecting project health to natural ecosystem indicators ### 📊 Accessibility Assessment #### Screen reader semantics - Present: Clear heading structure, descriptive video content with dialogue - Improve: Transcript with timestamp navigation, alt text for dashboard screenshots #### Keyboard navigation - Present: Video player supports standard keyboard controls - Improve: Skip links for different discussion topics, accessible video controls #### Contrast and readability - Present: Standard video quality with clear audio - Improve: High contrast transcript option, adjustable text sizing #### Text resizing and reflow - Present: Responsive video player and transcript display - Improve: Text-only version that reflows appropriately at high zoom levels #### Media and captions - Present: Full transcript provided with speaker identification - Improve: Synchronized captions, downloadable transcript formats #### Forms and interactions - Present: Basic video interaction controls - Improve: Accessible feedback forms for presentation evaluation ### 🌍 DEIB Assessment #### Cultural breadth - Present: International perspective with US/German experience, global company context - Improve: Examples from diverse geographic regions and cultural contexts #### Inclusive language - Present: Professional technical discussion avoiding exclusionary terms - Improve: Plain language summaries for technical concepts, avoiding industry jargon #### Neurodiversity support - Present: Structured presentation with clear agenda and discussion sections - Improve: Content warnings for rapid topic changes, consistent visual formatting #### Representation --- ## Full Transcript [00:02] Okay, cool. Um, we are recording now. [00:06] Welcome to the chaos working group [00:08] meeting. Today is April 2nd and we are [00:11] having today we have Travis Hathaway who [00:14] will be talking about using chaos at [00:16] Anaconda. So I will pass it on to you [00:19] Travis to kick us off. [00:23] >> Great. I'll go ahead and uh start [00:24] sharing my screen. [00:26] Share the whole desktop. Cool. [00:30] All right. Yep. So, um, today what I'm [00:33] going to talk about is more it's going [00:35] to be more of a story and a sort of a [00:39] look into what, um, is currently going [00:42] on at Anaconda um, to better um, measure [00:46] and kind of assess the health of the [00:50] open-source projects that we support. [00:53] And yeah, so I'll be kind of using a [00:56] combination of this HackMD document I [00:59] made plus a couple. We're going to be [01:01] going through some of my browser tabs [01:03] that are just going to be kind of [01:04] telling the story. So yeah, just uh [01:07] that's how the the presentation today is [01:09] going to work. And then if anyone has [01:10] any questions at any point, please uh [01:13] interrupt me. I would be happy to kind [01:15] of um discuss. But um otherwise, I will [01:18] have some kind of some discussion items [01:21] um that we could think about at the end. [01:22] So yeah, so um for all of you that don't [01:26] uh know who I am, um so I'm I've been [01:30] working at Anaconda now since 2022 and [01:34] um yeah, I'm originally from the United [01:36] States uh and I moved to Germany [01:39] actually in uh 2019. Before that I did a [01:43] master's studies in um in Germany and [01:47] the studies program was actually nothing [01:48] related to software engineering. It was [01:50] the it was called sustainability, [01:52] society and the environment. And so it [01:54] was very much a social science and kind [01:57] of geography focused degree. And when I [02:00] came across the kind of topic of open- [02:04] source metrics, assessing project health [02:07] and kind of the the social aspects that [02:10] had that really interested me, right? [02:12] Because it's um it kind of involves both [02:14] aspects of social science but then also [02:16] um technology. [02:18] And so that's kind of why me as a [02:20] software engineer I' I've taken a a [02:22] particular kind of interest in this. And [02:25] um I should also mention that the um [02:28] work that I'm doing is part of our [02:31] community guild within our company and [02:34] therefore it is not per se my sort of [02:39] day-to-day task. My my day-to-day task [02:41] or manage is the cond package manager. [02:44] Um and so I'm allowed to donate about um [02:48] I would say 10% of my time to these [02:51] projects. Um yeah, but then regardless [02:54] of that, we us as the community guild, [02:57] we have set a goal um to deliver a [03:00] report on sort of a handful of our [03:04] outbound um open source projects um [03:07] projects that we basically staff and [03:09] yeah, kind of make available for [03:12] everybody to use. Um yeah and so the one [03:15] some of the reasons that we're doing [03:16] this is that so we we I think a lot of [03:19] us know that these projects do have a [03:21] lot to offer and we on top of assessing [03:25] the health it's also a what what I I'm [03:29] kind of personally interested in is sort [03:30] of telling the stories of these projects [03:34] and yeah we'll I'll talk about that more [03:35] in a second but yeah but just to give [03:37] you an idea of some of my more my [03:39] day-to-day work um I'm involved with a [03:42] kind of packaging packaging manager [03:44] which the Honda um packaging manager is [03:47] more than actually just a packaging [03:48] manager right it's this is it's an [03:49] actually it's an open-source ecosystem [03:52] of packaging and it have we have our own [03:55] sort of philosophy towards packaging um [03:57] we have our own set of standards for [03:59] packaging and so this community exists [04:03] obviously within Anaconda and was kind [04:06] of the the genesis of this was at [04:08] Anaconda but we also very much are now [04:12] for for the last yeah since actually [04:13] I've been at Anaconda we've been really [04:15] trying to make this a community that's [04:18] supported by multiple stakeholders um [04:20] and not just Anaconda and um yeah but [04:24] let me get on we'll move to the next [04:26] slide I can't see [04:28] oops all right so um yeah so at Anaconda [04:33] uh we do I mean one of the things about [04:36] Anaconda if uh if people don't know they [04:39] were um a company that was started in [04:41] the early 2010s and back then it was [04:43] known as Continuum Analytics and the [04:47] company sort of started [04:49] a out of um kind of the the offices in [04:55] Austin, Texas. And it was two people [04:57] that were instrumental in in founding it [05:00] which was uh Travis Alifont and Peter [05:02] Wang. And they were essentially solving [05:06] problems for developers back then. um [05:09] especially and especially developers in [05:11] scientific computing. And so back then [05:14] the sort of act of just getting stuff [05:17] installing and running on your computer [05:18] was actually quite difficult. And so the [05:21] sort of flagship open source project [05:23] from Anaconda helped solve this um help [05:26] solve a lot of these problems. And so [05:28] then that sort of flourished into the [05:30] business that is today. Um, and I think [05:32] now the company currently employs over [05:35] about 300 to 400 people and is has a [05:38] presence in I think several continents. [05:41] I know North America, Europe, Asia and [05:44] Australia for sure. Yeah. And then um so [05:48] a lot of the work that we do is we we [05:50] support open- source uh development by [05:53] um [05:55] basic staffing uh projects uh like the [05:59] project I'm working on but then other [06:01] projects such as uh beware um pi scripts [06:05] uh numba and um dask and hollow viz and [06:08] so um this is also done by supporting um [06:15] a organization called Num Focus. And Num [06:17] Focus is a nonprofit that's specifically [06:21] uh focused on giving scientists uh [06:24] better tools uh for any I mean and [06:27] there's just such an amazing variety of [06:29] packages here. Um everything from data [06:31] visualization to really popular machine [06:34] learning uh libraries to um astroi which [06:39] is uh used heavily in astrophysics. Um [06:42] yeah, so and Anaconda is is heavily [06:46] involved in sponsoring um this [06:49] organization. [06:50] And so um yeah, so where does this take [06:54] us? uh back to sort of Anaconda's [06:56] outbound open-source metrics and kind of [07:02] now I'm going to talk about the early [07:05] sort of days of things that we've kind [07:08] of tried out to measure the the impacts [07:11] of these of these projects. And I want [07:14] to show you all a tool of of mine that I [07:18] developed but never really um got to [07:22] publish because like I said a lot of [07:24] this work sort of takes the yeah it sort [07:28] of takes a backseat to my primary work [07:30] on condom. And so this was merely just a [07:33] prototype that we um that I kind of took [07:36] maybe about a month to complete. And um [07:39] but the idea here right was that we had [07:41] a open-source uh measuring open source [07:44] dashboard where it would be a dashboard [07:47] where you could go to assess the health [07:50] of various open- source projects and um [07:55] was actually I mean it it didn't it it's [07:57] pretty minimally useful um but still [08:00] pretty neat. I mean we have as you see [08:02] we these are like the the big projects [08:04] we support. So, um, some Jupyter [08:07] projects, Piscript, Beware, um, some the [08:10] Hall of Data visualization and [08:13] ultimately this was meant to sort of be [08:15] expanded on even more, but at the end of [08:17] the day, it was only just measuring um, [08:21] things like activity on the project um, [08:25] and then also just the number of issues [08:28] and pull requests. And this was all [08:30] being pulled from from the um yeah from [08:34] GitHub's APIs directly. But yeah, we can [08:37] kind of see the various projects here. [08:39] Yeah. And um and flip there to see sort [08:43] of these uh projects as well. [08:46] But yeah um [08:49] so [08:51] after writing this and a little bit time [08:54] some time comes later I I'm also [08:56] introduced to things like the LFX uh [08:59] dashboard which is is an amazing tool um [09:03] in my opinion that offers like a lot of [09:05] really cool statistics and basically [09:08] renders like you know things like this [09:10] kind of completely useless in my my [09:12] opinion because this is um just [09:14] organized in such a better way. And so, [09:18] um, yeah, so this was, um, a really neat [09:21] project. I've been keeping an eye on it [09:23] and also sort of assessing it for our [09:26] use internally to see whether it could [09:29] be kind of part of how we kind of [09:32] collect and uh, display data um, for uh, [09:36] the open source projects that we that we [09:39] maintain. [09:41] And then um but additionally right I I [09:44] also wanted to kind of take a moment to [09:46] talk about the chaos metrics because the [09:52] what what I felt where I feel that chaos [09:56] metrics kind of offer a benefit and [09:58] where you kind of sometimes can get lost [10:01] here is [10:03] these are just kind of statistics [10:05] dashboards. This one offers a really [10:07] nice sort of like these really great [10:10] themes with contributors and everything. [10:12] And this is definitely a a big step in [10:14] the right direction, but where um the [10:18] chaos um health metrics sort of I think [10:21] differentiate themselves is that they [10:24] are very much specifically like tailored [10:28] to um specific to specific users, right? [10:32] the so if we look at the one I was [10:34] interested in which is the community [10:35] activity um to measure kind of what's [10:38] going on with um with our our the [10:40] projects we support um what I really [10:43] enjoyed was the fact that there it [10:46] begins with sort of a you know a why it [10:48] matters and a and a user stories um kind [10:52] of summary to really see um hey like [10:55] this is exactly the person that we want [10:57] to target and um these are the types of [10:59] metrics that you could use in in order [11:02] to address um these these questions. [11:06] And so right now um internally what what [11:09] we're doing is I've decided that it [11:12] would be really great for us to take the [11:17] the metrics in the community activity [11:20] and then um and then basically like uh [11:25] most of this is coming from GitHub. A [11:27] couple other things are coming from some [11:28] other sources. I mean, for example, like [11:31] community meeting rosters, right? Um, [11:33] I'll have to can't really automate that [11:36] or I can't. Well, sometimes you can, [11:38] sometimes you can't, but that that'll be [11:40] kind of a fun one to collect everything. [11:42] But the idea is is that um what what [11:46] what my goal is is is to create a report [11:51] um that's some it's kind of a you can [11:53] think of like probably like an annual [11:54] report at first just to sort of go [11:57] through the process of creating this [12:00] report once to see what it what it feels [12:02] like and then um being able to kind of [12:04] pass this around the company and [12:07] trying to get traction too within the [12:10] company to see if we raise anyone's [12:12] eyebrows. trials if anyone can say like, [12:14] "Hey, yeah, this is really great. I [12:16] really learned some things from this [12:18] report." Um, and so in order to do that, [12:22] I've I've written I'm kind of starting a [12:26] a new a new tool that's kind of written [12:29] on top of some of the existing [12:33] uh tools that I've seen floating around [12:36] in the sort of chaos sphere. Right? So [12:39] this is a this is a CLI tool that's [12:41] written using uh underlying is using [12:44] pers which is a really amazing um CLI [12:48] tool and so this is a where where PCAL [12:53] is more geared towards kind of gathering [12:56] statistics for specific projects. Um, [13:00] this one is kind of like sits a little [13:02] bit like a layer above that where we [13:04] we're mostly interested in in in [13:07] gathering and generating reports for [13:09] entire organizations on GitHub. And um [13:13] yeah, so this and this tool is still in [13:15] its infancy. I'm really hoping that um [13:19] some of the LLM kind of um yeah, LLM [13:23] tool um code tooling things are going to [13:25] help me go a bit faster um when [13:27] developing this. Um and actually and [13:30] right now I'm kind of this has been I [13:32] kind of got started and I'm it's been [13:34] taking a backseat because I'm actually [13:35] going to be presenting at uh this [13:37] month's PI data conference in Germany on [13:40] something completely different. So, um, [13:42] but yeah, after that con conference, I [13:44] really want to get going with this and [13:47] and be able to sort of have the the [13:50] metrics here, [13:53] but at the same time, [13:57] I think the the the the really important [13:59] part for me too is to kind of [14:04] get away from having just a a sterile [14:08] looking sort of statistics. um [14:11] dashboard. What I really want to be able [14:14] to do is to have a format where yes, [14:19] there are statistics, but there are [14:22] stories behind these statistics. And um [14:27] and so for example, if we look at the I [14:31] think maybe we can actually see this. [14:34] I'm going to go back here. I'm going to [14:36] look at beware and [14:39] yeah [14:41] February anyway. [14:44] Yeah. So there are for example if we [14:49] look at the closed pull requests um and [14:52] maybe if we even look at new issues [14:53] that's a little bit more noisy but for [14:56] example I know that the maintainer of [14:59] this project um Russell he actually he [15:03] uh we we were able to meet and um at a [15:05] couple at a conference last year Euro [15:08] Python and he told me about um a lot of [15:10] success that he had um for a sprint in [15:14] beware [15:15] and and how there was a lot of [15:17] first-time contributors. And you can see [15:19] some of these um these spikes in closed [15:22] poll requests, right? And so the thing [15:25] that I'm really interested in is to kind [15:27] of get get a format where you can see [15:30] these graphs and then have sort of for [15:33] example we would probably do like a 2025 [15:35] year review and then interview the [15:40] maintainers of the people involved with [15:41] the project to sort of shape the [15:44] narrative. So in that way we're both [15:46] taking quantitative and qualitative data [15:50] in order to really sort of like yeah [15:52] push push forward a a story and in the [15:57] project and sort of yeah how people are [15:59] dealing with it, right? and this Yeah. [16:02] And it's it sounds like a lot of work [16:03] and yeah, obviously it's going to be a [16:05] lot of work, but um that's sort of like [16:08] my kind of pie in the sky kind of goal [16:11] with all this is that and I think that [16:14] if I could focus on sort of that angle [16:17] with reporting these statistics, then it [16:19] then it then it becomes something that's [16:22] um that's way more um way more engaging [16:26] for people that are actually reading it. [16:28] and and and not only kind of it's not [16:31] just a statistics dashboard, right? It's [16:33] a way for people to learn about the [16:35] project and um yeah, and it could be a [16:38] really fun and and and if I can kind of [16:41] get this software working well enough [16:42] and kind of get the um maybe even [16:45] standardize some of the questions that [16:47] I'm asking as far as like, yeah, how is [16:50] this year and the software? Um that [16:52] could be um yeah, it could be a really [16:54] fun thing to kind of build on, but yeah, [16:56] we'll see. Um yeah, I guess that that [16:59] brings me back at [17:01] maybe um so I can kind of start to so [17:04] that I'm at the end of the stuff I [17:05] wanted to talk about and then so yeah [17:07] the on top of any questions that you all [17:09] might have some qu discussion items [17:12] questions I I thought about too was this [17:14] this uh there are there are so many [17:16] tools out there how do we how do we find [17:18] our way I mean I I actually and I right [17:21] before signing on today I saw I I really [17:23] I like the podcast that you all do I [17:25] think it's it's super amazing And I just [17:28] saw like this uh person's sort of AI [17:30] assisted repo health tool and I was like [17:32] oh and I looked at it I was like okay [17:34] yeah interesting and and it's just I I [17:37] know that with the and even with LLMs [17:39] right like the LLM is allowing me to [17:41] write my own tool and it's just this it [17:43] feels like we're we're in this Cambrian [17:45] explosion of tools. Right. Right. It's [17:48] just pretty crazy. So maybe maybe if [17:50] anyone had any idea just how to how to [17:52] find your way in this like sea of just [17:54] like oh there's so much stuff out there [17:56] and [17:57] The other thing too is just I'm really [17:59] really interested to hear what people [18:00] think about this sort of this narrative [18:03] approach, right? And combining [18:05] quantitative and qualitative data and [18:07] sort of trying to put more of a human uh [18:10] yeah kind of human- centered approach to [18:12] these dashboards. But anyway, thank you [18:14] very much. I appreciate your time and [18:16] yeah, happy to to talk about any of this [18:18] with all you. So [18:26] >> awesome. Thank you for sharing, Travis. [18:28] It's really cool to see all this work [18:31] and what you're doing behind the scenes [18:33] with open source at Anaconda. Um, okay. [18:36] Does anyone have any questions or would [18:40] like to discuss any of the action items, [18:42] the two bullets points below? [18:45] >> Yeah. [18:46] >> Okay. Go ahead, Remy. Oh, those aren't [18:48] action items, by the way. They're just [18:49] discussion items. [18:50] >> Discussion items. Excuse me. Yes. No. [18:52] >> No action necessary. Yeah. Yeah. Yeah. [18:55] Um, yeah. Remy, can you uh go ahead? [18:58] >> Yeah. Thank you for your presentation [19:00] today. Really interested to to check out [19:03] your uh chaos, you know, tools that are [19:05] in their infancy. Um, we've actually [19:08] talked a couple times within the OSPO at [19:11] CMS about more of a narrative approach [19:14] to, you know, going beyond the [19:15] dashboards and like trying to tell a [19:17] story. Um, a long long time ago, you [19:20] know, very long time ago, back when [19:22] like, you know, uh, [19:26] was sort of starting, uh, there was a [19:28] project called olo.net net that would [19:30] like scrape all of the open source [19:32] projects and then try to [19:34] >> put some kind of like metrics on top of [19:36] them and describe them a little bit. [19:39] >> And there's a a summarization tool that [19:42] I still think is like one of the best [19:44] ones for characterizing the activity [19:47] within an open source project. [19:48] >> Uh it was purchased or acquired by [19:51] Blackduck once upon a time and then [19:52] rebranded to OpenHub. So like if you go [19:55] and check out this page right here and [19:57] like you're still screen sharing so if [19:59] you just click on it in the chat it [20:00] should take you to the to the page for [20:03] itself which is one of the ones I tried [20:05] beware I tried pi script but [20:07] >> um the in a nutshell [20:10] >> uh section right in the center is one of [20:12] my favorites where it's just like um [20:14] it's actually up yeah they got a lot of [20:16] dashboard stuff too. Oh, right, right, [20:18] right. Yeah. Right here. [20:18] >> The narrative right here where it's like [20:20] it has this many commits by these many [20:22] contributors with this amount of code. [20:24] It's mostly written in this. It's well [20:26] established with a very large team with [20:28] increasing year-over-year commits, 22 [20:31] years of effort. First commit, last [20:33] commit. Like that narrative structure is [20:36] like if you can get those metrics, you [20:38] can plug and play them. And like [20:40] >> we've been experimenting in the OSPO [20:43] about like how do we pull certain [20:45] project metadata and then fill these [20:47] narrative structures so that you can [20:49] tell more of a story about the [20:51] repository instead of being like check [20:53] out this graph. Like a lot of people [20:55] don't really want to check out a graph [20:56] as much, right? So [20:58] >> yeah. Yeah, that's Yeah, [21:00] >> I love this. [21:01] >> Cool. That Yeah, thanks a lot for [21:02] sharing. And it's like um this this [21:04] reminds me cuz I like I said before the [21:07] the thing that I heard about on the [21:08] podcast the repo health I one of the [21:13] so the the the the thing so obviously [21:16] I'm I'm a I'm a maintainer one of the [21:18] maintainers of the the cond [21:21] um this repo health had a it's geared [21:25] for people that want to contribute to [21:26] the project and then it had a really [21:28] funny section that was at least funny [21:30] for me cuz it was showing ing the um [21:34] like top three or top four pull requests [21:37] that had been rejected and then giving [21:39] like an LLM kind of like reason for like [21:43] this was why it was rejected and then [21:45] this is what the mistake that was made [21:47] and and so I wonder and I and I so I [21:49] wonder now like with um with this in a [21:52] nutshell, right? Like I think that there [21:54] there's a lot of opportunity but also [21:57] also there's the opportunity to sort of [22:00] make these a little bit more richer with [22:02] LLMs because they can use their sort of [22:06] yeah the more humanlike kind of [22:08] expressions right um to to to summarize [22:12] the projects but at the at the same time [22:15] it's it's as a as a project maintainer [22:17] myself I always kind of think I always [22:21] look at it and go and I kind chuckle [22:23] because I'm like, "Well, [22:26] I guess why don't you just ask us, you [22:28] know, like you could I we could tell you [22:30] like how to make a good poll request and [22:32] you don't have to you don't have to use [22:34] an LLM." Like you can we I mean, for [22:37] example, we have a Zulip channel and [22:39] anybody can come by and um we we want [22:42] and we want people we want people to [22:44] kind of connect with us, right? Rather [22:47] than I Yeah. cuz I but but but I but I [22:49] totally understand like I mean because [22:51] if you want to just if if you if you're [22:54] if you're reviewing hundreds of open [22:57] source projects, right? Um it's like the [23:00] having taking all of that time to go and [23:03] and make a meaningful connection with [23:05] each of those project maintainers. [23:07] That's that's absolutely you can't do [23:08] that, right? And so I totally see the [23:10] the value of having those types of [23:12] tools. But yeah. [23:14] Yeah. Cool. [23:20] Yeah. [23:26] >> Yeah. Uh Laura, um please uh ask your [23:29] question. [23:30] >> Um I'm really interested. I'm a [23:33] community manager for um the University [23:36] of California network of open source [23:39] program offices. And so I'm really [23:42] interested in the community angle and in [23:45] particular how [23:48] prominently [23:50] community features in [23:54] the the like is it a KPI? Is it like how [24:00] how is it valued? How does it how is it [24:03] featured in in in the values? And um how [24:08] does that play out in terms of what gets [24:12] prioritized? I you were saying that you [24:13] can you can use 10% of your time for [24:16] this which is awesome. Is that like [24:18] common or is that special for you? [24:21] >> Oh yeah. No. So the let me and that's [24:23] actually a really good that's a really [24:25] great question because I I work in um so [24:29] my my des I'm a I'm a software engineer [24:33] and I'm working on this the on theonda [24:36] project and then I would say and so the [24:40] Anaconda doesn't actually have an OSP [24:43] office but they do have um a really [24:46] great team of uh developer relations and [24:49] so those are the people that are [24:52] currently like really leading the torch [24:54] on um on on a lot of that work and and a [24:57] lot of community building and so um I [25:00] and I work and so two of their names are [25:02] Dawn and Dana and yeah I'll actually be [25:05] seeing Dawn soon um in Pi for Pyon [25:08] Germany and so yeah they really are [25:11] doing a lot of that work so don't yeah [25:14] and so that's um I I'm not [25:17] representative but it's just because uh [25:19] I because I think I've I've worked in [25:22] companies in the past and they've kind [25:23] of framed it differently, but they [25:25] framed it as guilds or or other things, [25:28] but yeah. So, the community guild that [25:30] we have or Yeah. is it's it's just it's [25:33] like an interest group that um people [25:35] that might not necessarily get to do [25:39] more of like the dev rail or like these [25:42] or for example in this case kind of like [25:43] some of like the OSPO work they can kind [25:46] of get their feet wet and they can go [25:48] experiment around and and pick up [25:50] projects uh for that. So, I think it's [25:54] and it's actually really great and I and [25:55] I really enjoy the fact that the company [25:58] um has these sorts of groups and and [26:00] makes that available and it and it [26:02] really it for me it feels like it [26:04] empowers everybody to be able to sort of [26:06] pursue their own interest um at the [26:08] company regardless of what role role [26:11] they're in. So, does that does that help [26:13] make make does that make a little bit [26:14] more sense now that I've explained it? [26:16] >> Um yeah. Yeah. I guess I'm I'm still [26:20] wondering a little bit um about how [26:23] whether whether community is Oh, I'm [26:26] sorry. I'm totally asking a second [26:27] question. Is that allowed? [26:28] >> All right. Yeah. No, no. Yeah, totally. [26:30] Yeah, please go for it. [26:31] >> Um [26:33] how how weather community is like um is [26:39] part of the Anaconda narrative super [26:41] much? like I I am more familiar with [26:44] like um Pippi than I am I mean I know of [26:47] Anaconda but I haven't I haven't used it [26:49] very much. Um and so I'm just curious to [26:52] know how much how important the [26:54] community is to Anaconda as a whole if [26:58] that makes sense. [26:59] >> Yeah. Yeah. I mean, I can unequivocally [27:01] state that the community is incredibly [27:04] important to Anaconda, right? Like [27:06] Anaconda is a is a company that is [27:10] founded by open-source and would not [27:12] exist without um all of the things that [27:15] like open source does. And so because [27:18] the company has such a huge vested [27:19] interest in open source, it it directly [27:23] invests like um in projects in [27:26] conferences. Um, for example, this very [27:29] strong relationship with Num Focus. Um [27:32] so it's it basically yeah I would say [27:35] that it's a it's a really really big [27:38] focus of of the company as a whole and [27:41] specifically you know the company has [27:43] has you know I think what you could [27:45] what's known in the industry as an open [27:47] core model where we we offer our core [27:50] offerings for free for and so free for [27:53] for community users and because we want [27:55] to support people we want to support the [27:57] language we want to make sure that these [27:58] these technologies are going to be [28:00] around for the long haul and that [28:02] they're sustained. But then obviously [28:04] and then like a lot of other open source [28:06] companies, we have, you know, added [28:07] added features that people can buy and [28:09] you know um use for subscriptions. So [28:12] yeah, that's that gives is that that [28:14] that's a bit of a long and short of how [28:17] um yeah, of how the the company engages [28:21] with community and stuff. So [28:22] >> cheers. [28:24] >> Thanks. Yeah. Uh Remy, can you Remy, can [28:27] you uh would you like to answer or ask [28:29] another question? Yeah. So the first uh [28:33] question is just a request to post the [28:35] link to your hackd page that you were [28:38] sharing earlier. [28:39] >> Totally. And then uh the second one is [28:45] so it sounded to me like um [28:49] the [28:51] community guild was not just like an [28:54] open-source uh interest group or like [28:58] community but like actually was like a a [29:01] focus on making upstream contributions [29:03] or making open source contributions type [29:05] thing. Did I get that right or did I not [29:07] understand it? maybe just tell tell me [29:09] more about the community guild. [29:11] >> Yeah, so the community guild because [29:12] because the thing is is that uh Anaconda [29:16] by by staffing um open-source projects [29:21] uh they they already do I mean they they [29:24] already do make substantial like very [29:26] like big contributions to the wider open [29:28] source community. um especially um all [29:31] the different sort of technologies that [29:32] Num Focus sponsors um like the Beware [29:35] project, PiScript and everything. Um, [29:37] the community guild is more of a it's [29:41] more of like a meta group that is trying [29:44] to [29:46] trying to make sure that we are being [29:50] attentive to the communities like the [29:53] different open source communities they [29:55] and it's also you could think about it [29:57] also as a place because because I I kind [29:59] I listed off like a bunch of different [30:01] open source projects at the company [30:02] right and so the thing is is that we we [30:05] all these different open source those [30:07] projects and and like I said with the [30:09] Honda like ecosystem right like that's [30:11] its own big ecosystem of of people and [30:14] stakeholders in at Anaconda outside [30:17] Anaconda and the thing is is that we we [30:20] wanted the community guild as a space to [30:22] for people working in these different [30:24] open source projects to be also to be [30:26] able to come together to kind of share [30:28] like hey you know I'm organizing a [30:31] community group this is working this [30:33] isn't working or hey there's a project [30:36] that um I feel is in danger of going [30:40] under because um you know the core [30:42] maintainers stepping down is this [30:44] perhaps something that we want to start [30:46] supporting. So it's it's it's very much [30:48] like it's kind of like yeah it's like [30:49] the meta group of all the different sort [30:52] of open source efforts at at in the [30:54] company. So that's that that's that's [30:56] kind of my my best way of of explaining [30:58] it. It's and it's also had a couple of [31:00] different iterations too as well and [31:02] we're at the company we're currently [31:04] with on our on our kind of second [31:05] iteration of it. So yeah. [31:09] >> Yeah. No, that structure sounds very [31:11] familiar. Um not in my not in my current [31:15] OSPO but in previous OSOs I've been a [31:17] part of in the private sector. Um a [31:20] common sort of configuration or [31:22] organizational pattern I've seen is like [31:25] a like an open source shepherds group. [31:28] So you know across the organization [31:30] there will be the different open source [31:31] leads or community leads from each of [31:33] the individual repositories or projects [31:36] and then you have some kind of like [31:38] regular convening where everybody can [31:40] come together at that meta layer and [31:43] talk about like project level stuff and [31:45] board level stuff. [31:47] >> Um as opposed to being just focused on [31:49] like your one vertical within your one [31:51] project. And [31:53] >> yeah, [31:53] >> you know, I think that pattern [31:57] uh [31:59] from like an OSBO evolution standpoint, [32:02] like it predates OSOs to a large extent [32:05] as well because they're, you know, it [32:07] also is like a a maturity where when you [32:11] have enough mature open source projects [32:13] and they start to each have their own [32:15] community like identity and they're each [32:17] community like leaders, then the need to [32:20] like coordinate across them. So I think [32:22] that's like a wonderful sign of like a [32:24] very mature uh open source community [32:26] within I mean I'm not surprised like you [32:28] folks have some of the biggest Python [32:30] projects and some of the you know OG [32:32] Python hackers on you know in the [32:34] community [32:34] >> in your stable. So [32:36] >> yeah, [32:36] >> this is fascinating like really you know [32:39] and you probably didn't come here to [32:40] talk today about like your your [32:41] organizational structure and how you do [32:44] things but like you know OSBO folks run [32:47] them and manage them like we you know [32:49] everybody's learning from each other's [32:50] models and how we structure and you know [32:53] operate our organizations and it's you [32:56] know both familiar and novel to to hear [32:58] some of the ways you folks are [32:59] approaching it. So thank you for sharing [33:01] today. [33:02] >> Yeah great. Yeah. Yeah. and and it's [33:04] it's been it's also, you know, it's a [33:06] two-way street. I was really I don't I'm [33:08] not sure if I I wasn't able to make the [33:10] the previous meeting, but the one before [33:12] that was really informative. Um I think [33:14] it was Cornelius from Deutschean came [33:17] and spoke. And yeah, um for me it's it's [33:22] I think that's the power with these [33:23] groups is that we get to sort of share [33:27] about kind of what's going on um you [33:30] know as internally, right? we don't [33:32] obviously we're sharing what we can [33:34] share. Um, but it's it's really cool [33:37] that and I think because I mean really [33:39] the for me the the whole the whole open [33:41] source ethos is that there it's it's [33:44] good to know when it's when it when we [33:46] can collaborate and kind of knowledge [33:48] share and and really kind of make sure [33:52] that you know as a world right or yeah [33:55] whatever that we're all sort of maybe [33:57] we're we're we're able to operate more [33:59] efficiently. But then obviously there's [34:01] um you know we live in sort of [34:03] capitalist mostly capitalist countries [34:05] where we then also need to sort of [34:08] maintain a competitive edge right and um [34:10] I I think yeah open source software and [34:13] that whole ethos kind of yeah it's uh [34:16] it's it's cool it's it's on one hand [34:19] yeah has this kind of sharing mentality [34:22] but then still leaves the door open for [34:25] yeah a bit of competition and a bit bit [34:27] of like differentiation amongst the [34:29] competitors in order to see who's the [34:31] best, right? And who can, you know, [34:33] well, yeah, who can offer the best [34:34] products and all that good stuff. But [34:36] yeah, just my own little philosophy on [34:38] that, though. But yeah, I I actually and [34:40] I assume that it's interesting because I [34:41] think um if I'm not mistaken, some of [34:44] you are working in the the public [34:46] sector, right? Yeah. Yeah. Okay. [34:50] Ah, cool. Yeah. Yeah. I've I've Yeah. At [34:53] one point, I did I actually worked for a [34:55] university. It was one of my first tech [34:57] jobs. It was Portland State University [35:00] in Portland, Oregon. And um yeah, that [35:03] was an amazing job. I really really like [35:05] that um job. We got to I worked for the [35:08] research compute department. So Oh, [35:11] nice. Thanks. [35:13] Yeah, research. No, wait, excuse me. It [35:15] was It was way more fancy sounding. It [35:17] was academic research computing. [35:22] We worked it was great because we worked [35:23] right next to the they they put us in [35:25] the basement office that was like right [35:28] next to the data center. So the offices [35:31] were always and we were and we were [35:33] right next to the biggest air [35:34] conditioner on the whole campus. So the [35:36] office was always just super cold and [35:38] dark. It felt like a cave. So [35:41] yeah, Remy. Yeah. Yeah. Go for it. [35:44] >> Yeah. I know we have some some academic [35:46] friends here and curious is uh is [35:48] another group that's you know closely [35:50] associated with with chaos and the [35:51] hospital working groups but you know a [35:53] quick comment is uh you know the the [35:55] academic research computing you know [35:58] sector or like you know wing of the open [36:01] source community has really gone through [36:04] such a transformation in the past couple [36:06] years like it used to be just for like [36:09] are you doing like high performance [36:10] computing and like big distributed time [36:13] sharing massive of like supercomputer [36:15] stuff [36:16] >> and then all of a sudden once the LLMs [36:19] hit it's like oh yeah where do we have [36:21] the gigantic computing resources that [36:23] have GPUs to spare and like those places [36:26] have become just like the epicenter for [36:29] a lot of like AI things now too. So I'm [36:32] really interested to revisit, you know, [36:34] I I also was at the the the open source [36:37] program office at Rochester Institute of [36:39] Technology for a while and did, you [36:40] know, my stint in the academia in the [36:43] like era of like the 2010s through [36:45] 2015s, 2016ish. And so, you know, we [36:49] often were were hanging out with our [36:50] friends in the in the research computing [36:52] department. They had some some really [36:54] prolific open source projects there. And [36:57] um now that is you know sort of [36:59] reentered again I think is is one of the [37:02] key places in open source that uh I [37:05] haven't revisited in a while. Does the [37:07] UC OS system do you folks have like big [37:09] research computing connections or like [37:11] big research computing uh you know [37:14] stakeholders within your network Laura? [37:17] >> Uh yeah we do. Um, so I think I'm not I [37:22] guess I'm not actually sure whether each [37:24] campus has one, but I know of them on [37:28] most of the campuses. Each one has their [37:31] own HPC. Um, there isn't a centralized [37:35] like a UC HPC. it it'll be located in on [37:40] the various campuses but we also have [37:42] like we have three national labs and uh [37:45] they for damn sure you know um and and [37:50] uh one of our you know flagship projects [37:54] is Jupiter and they're kind of kind of [37:56] tight with the data folks too. So [38:08] >> cool. Well, uh yeah, if um yeah, I think [38:12] if uh is there any um other sort of [38:15] meeting talks for today? Maybe we can [38:17] kind of close this this portion of the [38:19] meeting. Thank and thank and I thank you [38:22] so much everyone for the warm welcome [38:24] and yeah giving me a place to come share [38:26] some of the fun stuff I've been working [38:28] on over the last year or so. So really [38:30] appreciate it. [38:35] >> Okay, last call for questions. [38:44] Okay, cool. Otherwise, thank you so much [38:46] for being our main presenter, Travis. [38:49] Um, we always love hearing what other I [38:52] guess you're not an OSO, but what other [38:54] companies have been doing in the open [38:56] source space, especially with [38:58] outbounding. [38:59] I'm uh part of the CMS OSPO and that's [39:02] like um one of the main uh focuses that [39:05] I have in my work. I think one thing I [39:08] want to share actually is this is our [39:11] own um metrics website with our own set [39:14] of reports. Um yeah, feel free to check [39:17] out. We have reports on the organization [39:19] level and also on the project level. And [39:21] we also similar to you have community [39:24] activity metrics, but on top of that, we [39:27] have things like libeers and other [39:29] security metrics. And we also include [39:31] metadata from um in our reports too to [39:34] kind of help with our storytelling if [39:36] you're curious about how we're tackling [39:38] this. And um let's see, we have new [39:42] interns coming on this summer. So our [39:44] goal is to have them um implement more [39:47] chaos metrics into these reports and [39:50] hopefully tell work on our storytelling [39:53] um in these reports as well. So we'll [39:55] see how it goes. Um but yes, thank you [39:58] for sharing what's going on on your side [40:00] of the world. [40:01] >> Yeah, perfect. And I'll I'll stay on [40:03] Slack and I'll I'll give you guys let me [40:05] give everyone updates on uh yeah on how [40:08] all the the tools that I'm writing. I [40:09] intend to keep the tools open source so [40:12] we'll be able to share it with you all. [40:14] Okay, cool. For sure. We'll keep in [40:16] touch. Okay. Um, let's see. For our last [40:20] couple of agenda items, we have a couple [40:24] of events that are coming up. There's [40:26] one next week. It's called the OSPO [40:29] virtual conference hosted by Punch Tape. [40:32] I'm going to drop this in chat. Um, [40:36] let's see. This one I believe is a 2hour [40:38] free virtual event where they are [40:41] bringing um a bunch of leaders from opos [40:47] and practitioners to kind of show how [40:49] opos work in practice not just in [40:52] theory. Um the CMS OSPO will be giving a [40:55] presentation there but we also have lots [40:57] of friends in the space that are joining [40:59] us such as the Ford OSPO. [41:02] Let's see. [41:04] Ford OSO, there's open source cowboy [41:08] consulting that's going to be speaking [41:09] there too. Some AI talks and some OSO [41:13] tools as well. Um, yeah. So, that's next [41:15] week and that's free. So, feel free to [41:17] join us there. And we have two more [41:19] events. The ASF's community over code [41:23] happening in Glasgow later in October. [41:26] And then we have all things open [41:28] happening in October as well in North [41:30] Carolina. and the CFP was due earlier [41:33] this week. Cool. Otherwise, that's all [41:37] that I had for today's agenda. Does [41:40] anyone have any final thoughts? [41:42] Otherwise, we will close out this [41:44] meeting. [41:46] Going once, going twice. [41:49] Okay, cool. Awesome. Thank you for [41:51] joining us everyone. We will see you in [41:53] two weeks. Thank you again, Travis. Bye. [41:56] Bye. [41:57] >> Thank you. [41:57] >> Bye, everyone. [42:04] Good. Yes. [42:06] >> Nice job, Matt. [42:08] >> Thank you. [42:08] >> Thank you for sharing.  📄 **Full text:** ![[Full Document Contents/Articles/CHAOSS OSPO WG, April 2 2026.md|CHAOSS OSPO WG, April 2 2026 by YouTube]] ## Highlights - All right. Yep. So, um, today what I'm going to talk about is more it's going to be more of a story and a sort of a look into what, um, is currently going on at Anaconda um, to better um, measure and kind of assess the health of the open-source projects that we support. And yeah, so I'll be kind of using a combination of this HackMD document I made plus a couple. We're going to be ([View Highlight](https://read.readwise.io/read/01kpb5pcpx7khwjnermjtbb8g7)) - Tags: 👻 ai highlighted - being pulled from from the um yeah from GitHub's APIs directly. But yeah, we can kind of see the various projects here. Yeah. And um and flip there to see sort of these uh projects as well. But yeah um so after writing this and a little bit time some time comes later I I'm also introduced to things like the LFX uh dashboard which is is an amazing tool um ([View Highlight](https://read.readwise.io/read/01kpb5pcqn4aaf371yjjzt8f21)) - Tags: 👻 ai highlighted - to address um these these questions. And so right now um internally what what we're doing is I've decided that it would be really great for us to take the the metrics in the community activity and then um and then basically like uh most of this is coming from GitHub. A couple other things are coming from some other sources. I mean, for example, like ([View Highlight](https://read.readwise.io/read/01kpb5pcs0hvtfvcdenc9mrs5z)) - Tags: 👻 ai highlighted - projects and some of the you know OG Python hackers on you know in the community >> in your stable. So >> yeah, >> this is fascinating like really you know and you probably didn't come here to talk today about like your your organizational structure and how you do things but like you know OSBO folks run them and manage them like we you know everybody's learning from each other's models and how we structure and you know operate our organizations and it's you know both familiar and novel to to hear some of the ways you folks are approaching it. So thank you for sharing ([View Highlight](https://read.readwise.io/read/01kpb5pcwp8vc6yaphy5m3xa87)) - Tags: 👻 ai highlighted - appreciate it. >> Okay, last call for questions. Okay, cool. Otherwise, thank you so much for being our main presenter, Travis. Um, we always love hearing what other I guess you're not an OSO, but what other companies have been doing in the open source space, especially with outbounding. I'm uh part of the CMS OSPO and that's ([View Highlight](https://read.readwise.io/read/01kpb5pd4b888gdnyv0wrdmgbh)) - Tags: 👻 ai highlighted - of the world. >> Yeah, perfect. And I'll I'll stay on Slack and I'll I'll give you guys let me give everyone updates on uh yeah on how all the the tools that I'm writing. I intend to keep the tools open source so we'll be able to share it with you all. Okay, cool. For sure. We'll keep in touch. Okay. Um, let's see. For our last couple of agenda items, we have a couple of events that are coming up. There's one next week. It's called the OSPO virtual conference hosted by Punch Tape. ([View Highlight](https://read.readwise.io/read/01kpb5pdd58y0gxkmazym8wzwj)) - Tags: 👻 ai highlighted