From Kernel To CoPilot: Microsoft's Open Source by Lachlan Evenson 🔗 **Source:** [From Kernel To CoPilot: Microsoft's Open Source by Lachlan Evenson](https://www.youtube.com/watch?v=NjoOzOq9c38) > [!note]+ Document Notes > tags: [microsoft, open-source, kubernetes, azure, transformation] > author: Lachlan Evenson > source: YouTube > reference: https://www.youtube.com/watch?v=NjoOzOq9c38 > date created: 2024-12-19 14:30 > > # 🏷️ Title: From Kernel To CoPilot: Microsoft's Open Source Journey > <!-- deep-read:sections start --> ## 🧰 Action Inbox - [H1] Explore Azure Kubernetes Service for large-scale deployments — ChatGPT runs on AKS with 12 engineers managing 10M cores #kubernetes - [H2] Check out Copa tool for automated container patching — patches vulnerabilities automatically using OCI images #security - [H3] Take Linux on Azure challenge for skills development — free course for Linux and open source on Azure #learning ## 🧭 Table of Contents - [📝 Summary](#-summary) - [📖 Main Content](#-main-content) - [🗒️ Notes](#️-notes) - [📚 References](#-references) - [🗣️ Paraphrase](#️-paraphrase) - [📰 Newsletter](#-newsletter) - [✅ Action Items](#-action-items) - [🧪 Hypotheses](#-hypotheses) - [📊 Methodology](#-methodology) - [❓ Questions](#-questions) - [🔮 Future Research](#-future-research) - [💡 Implications](#-implications) - [🧠 Learning Styles Assessment](#-learning-styles-assessment) - [📊 Accessibility Assessment](#-accessibility-assessment) - [🌍 DEIB Assessment](#-deib-assessment) - [✅ Logical Validity](#-logical-validity) - [🛡️ DARVO Analysis](#-darvo-analysis) - [🎯 Conclusion](#-conclusion) ### 🔗 References - [Microsoft Open Source](https://opensource.microsoft.com/) - [Azure Kubernetes Service](https://azure.microsoft.com/en-us/products/kubernetes-service/) ## 📝 Summary - Microsoft's transformation from open source skeptic to major contributor over 15 years - Open source now powers Microsoft's biggest platforms including Office 365, Azure, GitHub, and OpenAI - Two-thirds of Azure VMs run Linux, up from 20% in 2017 - Microsoft operates some of the world's largest Kubernetes clusters with minimal operational overhead - Company follows "upstream first, downstream second" philosophy for open source contributions ## 📖 Main Content Lachlan Evenson, who leads Microsoft's cloud-native ecosystem team, chronicles Microsoft's dramatic shift from open source skepticism to becoming one of the world's largest contributors. The transformation began in 2009 with Microsoft's first Linux kernel contributions and accelerated through key milestones like Visual Studio Code (2015) and the GitHub acquisition (2018). The company now operates massive open source infrastructure, including the Cosmic platform that runs Office services on Kubernetes clusters with millions of cores. ChatGPT exemplifies this scale, serving 500 million weekly users on Azure Kubernetes Service with just 12 engineers managing the platform. Microsoft's approach involves contributing to upstream projects first, then incorporating innovations downstream into products. The company has created numerous open source tools including Dapr, Radius, and Copacetic to fill ecosystem gaps they discovered while building at scale. ## 🗒️ Notes #### ⏱️ [00:00](https://www.youtube.com/watch?v=NjoOzOq9c38?t=0) — [H1] Opening and Microsoft's transformation story - Key idea: Microsoft's journey from open source skeptic to advocate represents fundamental company change - Evidence: "We can change. And I think we can do this together" #### ⏱️ [02:30](https://www.youtube.com/watch?v=NjoOzOq9c38?t=150) — [H2] The famous dinner photo - Key idea: Image of Bill Gates, Linus Torvalds, and Microsoft executives dining together symbolizes transformation - Evidence: "Testament to the full 360 journey that Microsoft has taken" #### ⏱️ [03:45](https://www.youtube.com/watch?v=NjoOzOq9c38?t=225) — [H1] Timeline of major milestones - Key idea: Microsoft's open source evolution began 15 years ago with concrete contributions - Evidence: "2009 with the contribution of about 20,000 lines to the Linux kernel" #### ⏱️ [05:20](https://www.youtube.com/watch?v=NjoOzOq9c38?t=320) — [H2] Current scale and impact - Key idea: Microsoft became largest cloud contributor to CNCF projects - Evidence: "Azure became the largest cloud contributor to open-source cloudnative compute foundation projects" #### ⏱️ [06:15](https://www.youtube.com/watch?v=NjoOzOq9c38?t=375) — [H1] Linux dominance on Azure - Key idea: Dramatic shift to Linux workloads demonstrates platform neutrality - Evidence: "Over two-thirds of all virtual machines running on Azure are running on Linux" #### ⏱️ [07:30](https://www.youtube.com/watch?v=NjoOzOq9c38?t=450) — [H2] Upstream first philosophy - Key idea: Microsoft prioritizes contributing to open source before product integration - Evidence: "We contribute upstream first and then we take those innovations and bring them downstream" #### ⏱️ [08:45](https://www.youtube.com/watch?v=NjoOzOq9c38?t=525) — [H1] Cosmic platform architecture - Key idea: Microsoft's container platform runs Office services on massive Kubernetes clusters - Evidence: "Runs millions of cores. All of it on Kubernetes. Makes up one of the largest Kubernetes clusters in the world" #### ⏱️ [10:20](https://www.youtube.com/watch?v=NjoOzOq9c38?t=620) — [H1] ChatGPT infrastructure scale - Key idea: OpenAI's ChatGPT demonstrates efficient scaling with minimal operational overhead - Evidence: "500 million weekly users... only 12 people at OpenAI actually administering this Kubernetes platform" #### ⏱️ [12:15](https://www.youtube.com/watch?v=NjoOzOq9c38?t=735) — [H2] Azure Kubernetes Service capabilities - Key idea: AKS provides managed Kubernetes with automated operations and security - Evidence: "Fully managed service... automated deployments, automated remediation and automated healing" #### ⏱️ [13:40](https://www.youtube.com/watch?v=NjoOzOq9c38?t=820) — [H2] Fleet Manager for global scale - Key idea: Open source fleet management enables worldwide Kubernetes cluster coordination - Evidence: "CubeFleet if you'd like to run the open-source equivalent of our fleet manager" #### ⏱️ [14:55](https://www.youtube.com/watch?v=NjoOzOq9c38?t=895) — [H1] Microsoft's open source projects - Key idea: Company creates new projects to fill gaps discovered while building at scale - Evidence: "As we're finding gaps we've taken those gaps and turn them into open source projects" #### ⏱️ [16:20](https://www.youtube.com/watch?v=NjoOzOq9c38?t=980) — [H2] Copacetic container patching tool - Key idea: Copa automates vulnerability patching in containers with minimal effort - Evidence: "Run it. It'll identify the vulnerabilities, patch your containers, and store that patch container" #### ⏱️ [17:45](https://www.youtube.com/watch?v=NjoOzOq9c38?t=1065) — [H2] Security and SBOM tools - Key idea: Microsoft provides tools for software bill of materials and vulnerability management - Evidence: "SBOM tools... so you can generate SBOMs and help make your open-source projects testable" #### ⏱️ [19:30](https://www.youtube.com/watch?v=NjoOzOq9c38?t=1170) — [H3] Community engagement and learning - Key idea: Microsoft offers educational resources and seeks community feedback - Evidence: "Linux on Azure challenge... free course online to give you the skills you need" ## 📚 References - [Microsoft Open Source Portal](https://opensource.microsoft.com/) - [Azure Kubernetes Service Documentation](https://docs.microsoft.com/en-us/azure/aks/) - [CNCF Graduated Projects](https://www.cncf.io/projects/) - [Copacetic Project](https://github.com/project-copacetic/copacetic) ## 🗣️ Paraphrase - Microsoft transformed from open source skeptic to major contributor over 15 years, starting with Linux kernel contributions in 2009 - The company now runs most of its major services on open source, with Linux powering two-thirds of Azure VMs - Microsoft operates massive Kubernetes deployments like Cosmic for Office services and supports ChatGPT's 500 million users with minimal staff - The company follows an "upstream first" approach, contributing to open source projects before incorporating them into products ## 📰 Newsletter Microsoft's remarkable transformation from open source skeptic to champion offers valuable lessons for the WordPress community. In just 15 years, the company went from contributing zero lines to Linux to running two-thirds of Azure on open source software. The key insight is Microsoft's "upstream first, downstream second" philosophy - they contribute improvements to open source projects before incorporating them into their commercial offerings. This approach benefits everyone: the community gets better software, Microsoft gets proven technology, and customers receive more reliable products. For WordPress developers and agencies, this demonstrates how contributing to core WordPress, plugins, and themes strengthens the entire ecosystem. When we improve the foundation, everyone benefits from enhanced performance, security, and capabilities. Resources: [WordPress Core Contributor Handbook](https://make.wordpress.org/core/handbook/), [GitHub Skills](https://skills.github.com/), [Open Source Friday](https://opensourcefriday.com/) ## ✅ Action Items #### DevRel - Developer advocacy: Share Microsoft's transformation story at WordCamps, emphasize upstream-first contribution model - Education: Create workshops on Kubernetes for WordPress, container orchestration patterns - Developer experience: Document AKS integration patterns for WordPress hosting - Developer marketing: Highlight successful open source transformation stories in campaigns - Community management: Foster relationships between WordPress and cloud-native communities #### OSPO - Policy & Governance: Study Microsoft's upstream-first policy model for contribution guidelines - License Compliance: Review Microsoft's open source project licensing strategies - Security Management: Evaluate Microsoft's security tools like Copacetic for container scanning - Software Inventory: Implement SBOM generation using Microsoft's open source tools - Developer Support: Provide training on cloud-native WordPress deployment patterns - Community Engagement: Establish relationships with CNCF projects relevant to WordPress - Strategic Alignment: Track contributions to upstream projects as KPI #### Community - Maintainers: Learn from Microsoft's massive-scale operational models for WordPress.org infrastructure - Contributors: Promote upstream-first contribution philosophy in WordPress community - Sponsors: Develop case studies showing business value of open source contributions - Ecosystem: Foster collaboration between WordPress and cloud-native ecosystems #### WordPress - Core: Evaluate container-first architecture patterns from Microsoft's experience - Core Performance: Study Microsoft's approach to scaling Kubernetes workloads efficiently - Community: Share transformation stories to encourage enterprise contribution - Hosting: Document best practices for WordPress on Kubernetes using AKS patterns - Meta: Consider Microsoft's fleet management approach for WordPress.org infrastructure #### The WPCC - Programs: Create fellowships focused on cloud-native WordPress development - Projects: Fund integration work between WordPress and CNCF projects - Funding: Support developers working on WordPress containerization and orchestration #### FAIR - Technical independence: Study how Microsoft maintains upstream-first contribution model - Policies: Develop guidelines for corporate open source transformation - Governance artifacts: Create frameworks for measuring open source contribution impact ## 🧪 Hypotheses - Companies can successfully transform from proprietary to open source models within 10-15 years through consistent contribution and culture change - Upstream-first contribution strategies provide better long-term business value than downstream-only approaches - Massive scale operations (millions of cores) can be managed efficiently with small teams using proper automation and open source tools - Corporate open source transformation requires visible leadership commitment and measurable milestones ### 📊 Methodology - Longitudinal study tracking Microsoft's open source contributions from 2009-2024 - Comparative analysis of operational efficiency metrics between traditional and cloud-native architectures - Survey of enterprise attitudes toward open source before and after major vendor transformations ### ❓ Questions - What specific internal incentive structures enabled Microsoft's cultural transformation toward open source? - How can other traditional software companies replicate Microsoft's upstream-first contribution model? - What operational patterns from Microsoft's Kubernetes deployments could benefit WordPress infrastructure? ### 🔮 Future Research - Analysis of Microsoft's open source sustainability funding models - Study of developer productivity gains from upstream-first contribution policies - Investigation of security benefits from transparent open source development processes ### 💡 Implications - Practice: Enterprises should adopt upstream-first contribution strategies for better software quality and community relationships - Policy: Open source transformation requires clear governance frameworks and measurable contribution targets - Security: Transparent development processes and automated tooling improve vulnerability management - Education: Corporate transformation stories provide powerful advocacy tools for open source adoption - Funding: Upstream contributions generate downstream business value through improved software quality and ecosystem health ### 🧠 Learning Styles Assessment Reading level estimate: 10th Grade (Flesch-Kincaid) #### 👀 Visual learners - Present: Timeline graphics, architecture diagrams, scale statistics - Improve: Interactive transformation timeline, infrastructure architecture diagrams #### 🔊 Aural learners - Present: Conference presentation format with clear narration - Improve: Podcast-style discussion of key transformation moments #### 🗣️ Verbal learners - Present: Storytelling approach, clear analogies, technical definitions - Improve: Glossary of cloud-native terms, comparison tables #### 🧩 Kinesthetic learners - Present: Hands-on Azure challenge mentioned, practical tool usage - Improve: Step-by-step tutorials for Microsoft's open source tools #### 🧠 Logical learners - Present: Clear timeline progression, metrics, cause-and-effect relationships - Improve: Decision framework for open source adoption, ROI calculations #### 🧑‍🤝‍🧑 Social learners - Present: Community conference setting, emphasis on collaboration - Improve: Discussion guides for team transformation planning #### 🧘‍♀️ Solitary learners - Present: Self-directed learning resources mentioned - Improve: Individual reflection prompts on organizational change #### 🌿 Naturalistic learners - Present: Ecosystem thinking, organic community growth - Improve: Metaphors connecting software ecosystems to natural systems ### 📊 Accessibility Assessment #### Screen reader semantics - Present: Clear heading structure, descriptive timestamps - Improve: Alt text for slides, ARIA labels for interactive elements #### Keyboard navigation - Present: Standard video controls accessible - Improve: Keyboard shortcuts for timestamp navigation #### Contrast and readability - Present: High contrast presentation slides - Improve: Ensure transcript meets WCAG contrast ratios #### Text resizing and reflow - Present: Responsive video player - Improve: Scalable transcript text without horizontal scrolling #### Media and captions - Present: Video content available - Improve: Closed captions, downloadable transcript #### Forms and interactions - Present: Basic video interaction - Improve: Accessible chapter navigation, bookmark features ### 🌍 DEIB Assessment #### Cultural breadth - Present: Global technology impact discussion - Improve: Include perspectives from diverse geographic regions #### Inclusive language - Present: Professional technical language - Improve: Define jargon terms, avoid military metaphors #### Neurodiversity support - Present: Clear structure with timestamps - Improve: Content warnings for rapid topic changes #### Representation - Present: Single speaker from major corporation - Improve: Include diverse voices in transformation stories #### Psychological safety - Present: Respectful discussion of past corporate positions - Improve: Acknowledge impacts of previous proprietary approaches ### ✅ Logical Validity #### Strong reasoning - Deductive strength: Clear premises about open source benefits leading to transformation conclusion - Inductive strength: Multiple examples (Linux kernel, VS Code, GitHub) supporting transformation pattern - Causal reasoning: Timeline establishes clear cause-effect relationship between contributions and business success - Analogical strength: Dinner photo analogy effectively illustrates relationship transformation #### Fallacies - Appeal to authority: Relies heavily on Microsoft's size and success without addressing potential <!-- deep-read:sections end --> ## 📺 Video  --- # 📄 Full Transcript *Auto-generated from YouTube captions · timestamps roughly every 30 seconds · [watch on YouTube](https://www.youtube.com/watch?v=NjoOzOq9c38)* **[00:05]** Hello. Does everyone hear me? Hello. Hello. All right. It's all I love when a plan comes together. It's almost serendipitous. All the people that spoke before me led me to this talk that I'm about to give right now. And uh for anybody that knows me, I'm a storyteller. I come from a long line of storytellers. So, buckle up. We're going to tell a few stories. But I want you to think back to the beginning of today when we had Jim up on stage. Jim, uh, it was great to have you up on stage because Jim said a couple of things that **[00:36]** stuck with me. Uh, one was there were plenty in the room back in the day that didn't believe in open source and, uh, we were right in the end and many had to eat their words and change their ways. So to have to be here representing Microsoft on the stage talking about open source is a great honor of mine because I've been able to work in that. But I think above all that, you know, we can change. And I want to tell a story about change. And I hope, you know, through Jim's advocacy and everybody in this room, we can cause people to change. So if you look at me and take **[01:08]** anything away from this talk, it's that, you know, we can change. And I think we can do this together. So I'm excited to tell a story. Come with me on a journey. Uh, and we'll get going. Sound good? All right, let's do this. So my name is uh Lachlan Evenson and as Jonno said I lead the uh cloudnative ecosystem team at Azure and I'm going to tell a story today and share how our open-source journey has shaped the core and become **[01:38]** our strategy at Microsoft and how it's powering the biggest platforms in the company. This story is quite special to me because I've been working at Microsoft for eight years almost exclusively on open source and as you know folks like Jim said I don't think if he had a thought to 20 years in the future you know there would be folks at Microsoft actually working on open source as a as a full-time job. So I'm going to take you through our journey. **[02:11]** So, you know, we've all been around for many years. Open source is core to Microsoft strategy, powering our biggest products, for example, Microsoft 365, Azure, GitHub, and Open AI. And those platforms are enabling us to give back to the community. But how did we get here? Because as you all know, uh this certainly hasn't been the ways of this company. And if you look back to 15 or 20 years ago, if you were polite **[02:43]** and maybe you know Jim would indulge me, uh you would say that we were very skeptical about open source. So let's go for the next 20 years to me up on stage here. There's been a massive change that's occurred and I would like to share how that change has happened and how we got here. So like many of the photos on screen today, this was not generated by AI. We talked about Helsinki earlier with **[03:14]** Nithia. But has anybody seen this image? Could you believe that this is real? The first time I had to double take. So here, let me just explain it to you from left to right. We have uh Mark Rasinovich who's the CTO of Azure and uh cloud and AI. We have Bill Gates obviously no explanation needed. Lenus, right? No explanation needed or introduction. Dave Cutler, who is actually the architect behind Windows NT. Now, I think this is just such a, you know, a testament to the full 360 journey that Microsoft has taken. **[03:46]** They're actually having dinner at Mark's house and they're all chummy now. You know, hell hath frozen over to think about, you know, what happened to get to this point because I think that's just a really great testament to how we building community and how we're changing at Microsoft. The fact is our journey didn't start today and it didn't start at that dinner. It started almost 15 years ago. The first big change for us was in 2009 **[04:18]** with the contribution of about 20,000 lines to the Linux kernel. I think this was the start of a journey and a moment that marked at Microsoft to make us think that we are possible to change and make us think that hey we can contribute to open source as well. By the time we got to 2015 we had released Visual Studio Code which is a free open-source cloud editor for your code which turned out to be one one of the most popular **[04:48]** code editors uh out there. And in 2018, we acquired GitHub, which at that point had 28 million users. And if I've got my numbers correctly now up on on screen, we have about 85 million repositories. And it has continued to grow handover fist like crazy. And we're up to now 518 million projects that are hosted on GitHub, which is absolutely phenomenal. And I think we're just getting started. **[05:23]** We're doing all this innovation with our services as well as contributing to open source projects upstream. We're proud to say that last year Azure became the largest cloud contributor to open-source cloudnative compute foundation projects in the world. This slide shows some examples of the projects that we've made significant contributions to. We contribute to a variety of different projects. These projects are critical not only to how we operate our services, **[05:53]** but also to our customers and the community. We're going to go into a few of them in a few moments. Now, if we fast forward to today, open source now isn't just an afterthought to us. It's ingrained in how we build products and partner with the community, all of you. Today at Microsoft over twothirds of all virtual machines that are running on Azure are running on Linux. Can you believe that? **[06:24]** When I started back in 2017, it was about 20%. So that's been the growth uh over seven, eight years. Linux operating system powers some of the biggest workloads, workloads for Microsoft 365 and Open AI. We continue to be even more committed to open source than ever before. And we have the simple philosophy that we contribute upstream first and then we take those innovations and bring them downstream into our product. So upstream first then **[06:55]** downstream. Security continues to be a really critical part of how we think about open source going forward. And we have a number of projects that we've created. for example, Copathetic, Dalek, and even an ESBOM tool. These are all open- source projects that we're collaborating with the community on. **[07:25]** Now, let's take a look at some of our biggest platforms so I can demonstrate where we're using open source inhouse. The first workload I want to talk about is something called Cosmic. Think about Cosmic as our entire container platform and it's our developer platform for everybody who builds on office at Microsoft. There are thousands upon thousands of developers working upon this platform. It's the foundational system for Exchange, Teams, and Office, **[07:58]** and runs millions of cores. All of it on Kubernetes. And it makes up one of the largest Kubernetes clusters in the world. That system bakes in specific guard rails for software software developers making it so that we can operate one gigantic cluster at scale without being incredibly expensive or having a massive operational burden. It builds upon many other open source **[08:28]** projects as demonstrated up there like KADA which is a Kubernetes uh event-driven autoscaler, Prometheus and Metric uh Prometheus and Grafana for monitoring. So, Cosmic is a massive platform inside Microsoft powering many of our workloads. Now, I know we've talked about uh AI enough. Today, we're going to talk about it in the context of it being run on open source as well. So, chat GPT is another service that runs on top of AKS. It's a gigantic workload. There are over **[09:00]** 500 week uh million weekly users to chat GPT. And this entire service is built on the same kinds of Azure services that anybody can use. GPUs, Cosmos DB, AKS, Postgress, Azure storage. Many of these services are built upon open-source and as Jim said, open protocols and open standards as well. But what is really interesting about the platform that runs uh chatp is there's only uh 12 people at uh open AAI **[09:33]** actually administering this Kubernetes platform. So it's running at incredible scale but not incredible scale at the human side. So let's take a look at how you scale an app to 10 million cores with only a dozen engineers. Take a look at some of the platforms behind it. So this is where Azure Kubernetes service comes into play. Chat GBT is built upon AKS which provides a highly scalable Kubernetes service for cloudnative applications. Now as the **[10:05]** name says Kubernetes it's built on Kubernetes. It's upstream conformant and is open source first. So we build open source and then we bring downstream into product as I mentioned. And AKS is a fully managed service that's available in every Azure region that streams operations at any scale. It offers automated deployments, automated remediation and automated healing and built-in security standards **[10:36]** to make your Kubernetes journey easy. So platforms like chat GPT can scale without any significant operational overhead and of course being AKS it's Kubernetes first AKS provides so many additional services to help make your Kubernetes journey easy and the really great thing is you can take it further with things like fleet manager enabling global scale for your clusters anywhere around the **[11:07]** world fleet manager manager itself is built in open source and you can pick it up today in the CNCF. The project is called CubeFleet if you'd like to run the open-source equivalent of our fleet manager. We're using standard policy and standard operational uh models so that you can service and scale to many different clusters across the globe. As we've been building these massive services, we've noticed there are a bunch of gaps in the ecosystem. So, our **[11:39]** journey hasn't stopped by only using and consuming. We make sure that as we're finding gaps that we've taken those gaps and turn them into open source projects themselves. If I was to go through a few up here on screen, we have the first one up on the top right there. It's called Dapper with a beautiful tie. It's a CNCF project under the Linux Foundation and it helps application developers build complex distributed applications without having to understand the nuances of all the individual components. **[12:10]** Take a look at Radius up there in the middle. Radius was designed with the platform engineers in mind to help them enable their developer fleet to have access to the best of breed infrastructure to power their platforms. We have copacetic there down in the middle. Copa for short. Copa is a great tool. You can run it to patch your containers. You just run it. It'll identify the vulnerabilities, patch your containers, and store that patch container using um open source images and OCI images under the hood and it **[12:42]** leverages tools like Trivy to do scanning and vulnerability management. Couple more if I may. Dalek is another one that we've been working on recently in the CNCF. It is using declarative builds under the hood so that you can have reproducible builds for your open- source platforms and tooling. We actually use it internally ourselves to build many of these projects including things like Kubernetes. And I talked about security before. We have sbomb tools as well that we've given out so you can generate sbombs and **[13:13]** help make your um open- source projects are testable and identifiable for vulnerability management and uh a testability over time. So all these ways we've demonstrated how we've given back and hopefully I've shared a glimpse into how open source is something that that's at the core of Microsoft at the core of all we do. It's at the core of everything that I do and that we're running critical workloads on top of open source. We're **[13:44]** not just generating open source or contributing out there. We're actually running it ourselves and giving back. We continue to contribute to not only established projects like Linux and Kubernetes, but building new projects like Copa, Dalek, and our ESBOM tool as I mentioned. And there's some great contributions across the uh ecosystem. Finally, I just want to say that this is a journey and even though I'm up here on stage, I hope and there's been many talks of the next generation of open source engineers **[14:16]** building open source solutions. I hope that I inspire and the folks in this room can inspire you to continue to contribute to open source because I know that it is part of my job at Microsoft to make sure that these open source projects and communities are built in sustainable ways for the future. If you see me around the conference, I'm the one that loves to have a conversation. So please come up to me and talk about anything that you saw today. Also want to give uh a quick shout out for those who may be early to the open source journey. We actually have a Linux **[14:46]** on Azure challenge going on at the moment. You can come speak to us about that. The goal of that challenge is to give people skills working with Linux um and being successful with open source projects on Azure. So you can open that up. It's a free course online to give you the skills you need to be successful with Linux and open source. So, that's a great way to get started if you're looking to kickstart your journey. And then finally, we're really just excited to share with you about what we're building uh in open source at Microsoft. Obviously, it's not something that we've been big on in the past, but we're **[15:16]** looking to change that. And we've been really excited to get out here and take your feedback and learn from you all and hear what problems you're trying to solve. So, please come. I'll be at the booth after this and for the rest of the conference. Love to meet with you. We'll be at uh booth 21 and 22. And thank you very much and thank youto. Thanks Jonah.