 Okay, welcome back everyone, the Cube's live coverage here, day one of three days of coverage of Open Source Summit 2023. I'm John Furrier, your host. We've got great guests here, here in Vancouver, Rob Streche is analyzing and getting all the notes, bringing down all the action. One of the big conversations we're having is platform engineering, DevOps, software supply chain, and the role of communities as an advantage, as AI comes in, we're going to break down all the open source opportunities. Is it a challenge, is it an opportunity? We'll see, Diane Mule is here. Cube alumni, going way back, OG for the Cube, great to have you. Managing director, research and advisory services, Bittersierra. Bittersia, Bittersia, Bittersia, the new company name. You left Red Hat, I saw that announcement, welcome back to the Cube. Well, it's a pleasure to be here, Bittersia, and welcome to my city, we turned on the sunlight for you all, it was really wonderful to have everybody here at Open Source Summit in my hometown, so I didn't have to go through an airport to get here for a change, which is wonderful. Great to see you, just for the folks watching, you've been a really big part of the Cube, we've had conversations during the OpenStack days, the transition to cloud native with CNCF, at Red Hat you were involved in OpenShift, when it was a critical juncture, and it kind of, I won't say pivot, because that would make Red Hat roll in their grave, but they got on the right track and they had been very successful with it. Graduates at Red Hat, we thank them for all their support, so thanks Red Hat. But now, you've always been a very community-focused person, you understand communities, the role of ecosystems has been a huge discussion point as the industry has completely transformed to Open Source. It's not just standing on the shoulders of giants, it is all giants now, it's all the industry. The role of the power dynamics are shifting. What is your view right now of how Open Source is positioned, looking at the future of big wave of Open Source coming at AI, and then you got security, you got cloud growth, what's your perspective? So, I'm glad you said the word ecosystem right off the bat, because I really think one of the dynamics that has changed, back in the good old days, you could focus on a single individual project and you try and build a community around that, get people to work and contribute to it, and those days are gone. To paraphrase Steve Ballmer, it's ecosystem, ecosystem, ecosystem, and there's lots of perspectives to it. From a vendor perspective, it may be everything from the kernel, to the container, to the cloud, and that stack of things, from an enterprise that's using and consuming there, it may be everything that's in the supply chain or in the S-bomb that they've created for their product, and figuring out all the things that they consume that's Open Source, and their end users, and how they incorporate that into their services, so it has completely changed my perspective over the, I'd say, the past 10 years of working in Open Source, and I think the reason that that perspective change is important to recognize and acknowledge is because of the interdependencies of all these projects on each other. If you look at any of the foundations, and we can talk about big tents at Open Stack and the move to Open Infra for days, because we've been there, we watched the CNCF bringing in new projects in the sandbox at some exorbitant rates and incubating them and graduating them, but they're all interlaced. They have release cycles, they have library dependencies, dependencies on each other's resources, and if you don't pay attention to that, you're hooped. The other piece of the puzzle, I think that's really interesting that, Open Source has won, right? We've won, all right. We've got minutes. You can't bet against open, that's the theme here. Yeah, you just absolutely can't, but I think what we're also seeing is the dynamics about who's participating in Open Source projects and ecosystems change. Whereas before you'd have large vendors benevolently creating Open Source projects, incorporating your feedback and creating, wonderful things and Linux and all the goodness there. Now what you're seeing is projects like Backstage, like a Spotify, donating that. You're seeing enterprises like Amadeus or BMW or Audi or whomever, they're contributing. When you look at the statistics on Open Source projects and in the ecosystems, you see a lot of names that are not software vendors anymore, and so that changes the dynamics in the community around collaboration. End users. And end users, yeah, as we call them sometimes, the virtuous end users that are. Lyft is a good example. Envoy came from Lyft, Uber contributed stuff, so yeah, the hyperscale attack. Yeah, and I mean Meta's contributed stuff as well. Oh, absolutely, Meta's done amazing things. I think the interesting thing was, and going just back a couple weeks to CNCF when they were talking about that contributors were 25%. My coming out of that, my question was, is that healthy or not healthy? Well, community health is an interesting living organism and there are lots of perspectives to it. There are some really basic metrics, like how many contributors, how many deployments of your project, all those kinds of things, and those are things that historically we've looked at. I think when you start looking at community health from a broader perspective, are the missions of these projects, are the feedback mechanisms, are they in place, and is there a social network of people working together to bring these projects and keep these ecosystems healthy is really the change. And the other piece, I mean, we talk about the virtuous end users and we talk a lot about contributors and maintainers. I am of a new opinion or a loosely held belief, shall we say, that we need to switch the dynamic a little bit to participants, right? Because code is essential, people, but there are everybody from the translators to the documentarians who are doing this, to the end users who are giving us feedback, to the CICD, the people who are doing the build processes, all of the legal departments, all of those people are participants. And if you don't have a healthy engagement from all of those folks in your project, you don't have a healthy project. I want to ask you on that note, because we were talking earlier about how AI is coming in structurally, you got how things are organized, how the projects are organized and the team dynamics. But if you look at projects and the power dynamics and open source, you see the common thread. You have entrepreneurs who want to create business models out of it. They do a project and they try to be a red hat or try to be a support model. Everybody wants to be a red hatter. So which by the way, we confirm that no one can do that yet since it's never going to happen. You'll never see another red hat. But you'll see businesses come out of it. There are business models. So let's say entrepreneurs, entrepreneurs with some investment back in thinking. Then you've got companies, like the big companies, which could be like a red hat, a Cisco, and then you've got the lifts, contributors, all of this is called those corporations. And then public interest could be nonprofits and just general people. How do you look at those three pillars? Because entrepreneurial side has been very active over the past decade and so, so that's been very key. You're seeing a lot of entrepreneur activity now with AI surge coming in, a lot of open source being donated on open foundation models and large language models. That's booming up. What's the, how do you view the act like the core knobs? What are the core drivers of value? When you see an entrepreneur or someone come out with a new project, whether it's small one like MicroRocks inside the postman community or another, something like that, it's really, I think one of the things that almost everybody has drunk the Kool-Aid is that your project should be in a neutral repository. So things that are single vendor driven open source projects, you look to them if it's a trusted, wonderful vendor or something like that, then great. But for entrepreneurs now, what you're seeing I think is a lot of them looking to the foundations to be the Switzerland, to offer them governance and legal and indemnification if something goes south, those kinds of things. I think foundations play a big role. They get over their skis basically and they can't handle themselves. They need help. Yeah, they need help and it's hard to get brand recognition. So in a way that's really the role that foundations play in that. I think there is a role though. There's a thing that's, we talked a little, I mentioned the big tent effect in OpenStack and stuff. And there's a sort of a tipping point too where other foundations, I'm trying to be polite about this. I'm not going to be too polite. There's a duplication of efforts. There's only, there's a limited set of developer resources out there working on projects. So if you have an AI project in one foundation or automotive or Edge or whatever it is, and then the next foundation or the next group does another one, you're spreading those resources thin, you know? And so you're competing for those. So there's a level of competition between vendors trying to hire all the engineers, but there's also a level of competition between foundations for members and contributors. This is why the health thing, I want to loop back to what you said about health. Because what we're seeing is that there's a lot of people who want to jump in to make money or promote a project. Whatever their motivation is, it's putting a strain on the system. How do you know when something's rising and falling like legitimately? Like, and I bring some to context, some projects just deserve to die, right? Okay, no traction, whatever, old, not a lot of momentum. And then some that should be doing better, but aren't. And so how do people figure out where they're, where they can jump in and help or nurture? This is an open question. We don't know the answer. Well, sometimes the data shows you, okay? And it's a lot of it's the data plus the context. So you can analyze participation at the contributor maintainer, at the end user level, the deployments of projects, and to measure the health of an ecosystem. But you also have to really look at the connectivity of these projects to other projects in context. And I think that's really where you start to see true health if they're doing. So one mistake people often make is that a project will look like it's mature and static, but it may not need more than one or two maintainers. It may be just an essential small code base that we do it. Or if you don't dive into it and have the context to understand it, it, you know, all of its maintainers may have walked away and gone onto a sexier and new open AI GPT chat. They're just showing you toy dynamic. Oh, that's the elephant in the room is like, developers are just, they're human. You know, we're human. We like the new toys. And, but I think if you, you know, we talk all about AI and, you know, edge stuff and all that, but it all is reliant on this huge stack from the kernel to the containers to the cloud. All of those pieces still need to be in play in order for the AI and the algorithms to run and the Jupiter notebooks to launch and all of that goodness. So. And I think just building off of that and something you said earlier, I think it is a stack and there's dependencies up and down the stack. And I think, you know, again, two, you know, the two times I've been on the cube recently with CNCF and then here, I think one of the things that's striking is that there seems to be a lot of people bringing projects forward that overlap or. There's a lot of overlap. And to your point about competition, it's got to be hard. And I think it's, and some of it, it's up to the foundations that are accepting these projects to collaborate more and not to see each other as threats to it. You know, one, you know, land grabs for different projects and to allow, you know, maybe an automotive, you know, edgy thing to live in one foundation and be used with another and to collaborate across ecosystems. And there's not a whole lot of that, you know. I remember back, I actually ran a research group in the global grid forum before OpenStack. And I actually brought somebody in from DMTF because we were doing stuff and I'm like, why are we reinventing the wheel here? And I think it's, it didn't have, I mean, that's part of the reason I still believe they went away was that they lost their major sponsor. But at the same time, it seems like that, like you said, the foundations looking at their stack, very, I would say, in a silo. Yeah, it's interesting because in some ways, we absolutely need the foundations. We need them for governance, guidance, strategic visions and things along that line. And they serve a huge purpose. But in some ways, they are also causing some fragmentation and the spread of projects and I'll keep saying, there are a limited number of developers with expertise. And the one thing you hear from vendors and enterprises is we need more people trained on whether it's Kubernetes or something else. We can't find them fast enough. And so if you think about that in terms of the deployments at enterprises, that's a concern, but it's also a concern for these, the code bases of these projects, we need to find the resources and to continue to encourage and engage them. And to your point about that skill gap, but we're also having a conversation around platforms versus tools. So you're seeing a platform where everybody wants to be a platform. Where are those platform engineers? But then again, is it really a platform? Is it just an app that has systems thinking into it? We were talking about this before we came on camera around dependencies that the holistic view is now the new normal. Yeah, I think we need to apply more systems thinking to our ideas around ecosystems in the open source and all of our technologies. Because they cross industries now, they cross technologies, they cross all kinds of barriers from bare metal to hypervisors. There's just so many interlaced pieces to this that if we don't step back and reevaluate how we manage these ecosystems, there's a lot of risks involved. And what's your advice right now for the folks watching? As we are at the front range now of the next generation of open source, which is maybe a step function higher than, I mean, we were riffing earlier with some other folks on the Cube, us OGers, where we're doing, there was no open source. We have steel software. Oh, and I did. Back in the day, everyone will admit it now, looking back. Statue of Limitations is past, Rob, we can admit. Yeah, the cut and paste. But now it's been a great run. We've won, it's an industry. The software industry is open source, period, full stop. As we look at the next generation of talent coming in, opportunities in the landscape and the marketplace, what's your advice for how to lean into the ecosystem conversation, out of the mindset, set up the approach? So there's lots of ways to look at it. If you're looking at it from an enterprise that's trying to figure out your open source strategy, how are you going to look at the bill of materials of the products you're delivering and analyze them so that they're compliant, that you know what's actually embedded in there? That is like one of the oldest tasks I ever had. Like years ago, I worked at ActiveState doing Python library packaging and analyzing all the licenses. That was like 15 years ago and we're back to having bill of materials, supply chain conversations. It's really one of the things you really need to know if you're an enterprise is what are you building with? What are your developers using? What libraries? What dependencies do they have? So my first thing is know yourself, right? So really to look in deeply and gaze in to what you're using, what your developers are using. We at Bitersia have lots of tools to do software developer analytics and dive into that and help you give a bird's eye view of where your in-house developers are actually contributing and then maybe where you can find vendors that will help you to do that, to work on those projects as well. So there's that side of the house which is an old side of the house and I'm always shocked and surprised at how many people aren't doing it, right? And then there's the benevolent software vendors and then the virtuous end users and so trying to figure out how we're going to get all of these people to play with. And virtuous in the same sense, I love that, well done. Yeah, well we had our run of BDL, the Benevolent Dictators for Life and there are still a few of them out there floating and some of them I'm incredibly grateful for because they've done amazing things. But I think that the people that I'm most grateful for are these end users and the role that they're playing now because now they are contributing back into open-source ecosystems and that just really changed the dynamic. Yeah, and I think the hyperscalers, like the big ones who had to build on their own because the general purpose old proprietary market couldn't serve them, is the tell sign for how I see AI shaping out. You already see Meta just released, who called it, leaked. They leaked their kind of software, I think that was by design, but that just gave the open-source community a shot at the adrenaline. They created a lot of vibe and again, because it's shiny and it's white, but it's relevant. Cool and relevant. So cool and relevant is a good equation. Well, I mean, I've been looking at chat TPP and documentation, creating good documentation for an open-source project or basic information, those tasks, if we can automate some of that, that would be amazing. So applying some of those toolings, also I think the predictive analysis. So we have huge data lakes of developer participation and open-source projects. If we can take that, tie it to predictive analysis and watch as I'm in Canada, I can say this, where the hockey puck is going, then we have a better sense of what our next dependencies are going to be, what the next risks are. And a lot of it is just risk analysis, just like you would for a product. You should be doing risk analysis on your open-source consumption. Yeah, amazing industry power down there, product market fit, you got to promote the project, you got to recruit a team, it's all the same kind of venture dynamics. Diane, great to have you on theCUBE. And I know you're now at a new, you left Red Hat, you're at the new gate. Take a minute to explain what you're working on. What are you excited about? Well, I am incredibly grateful for the run that I had at Red Hat. My colleagues, I got to work with some of the best and brightest people, so big shout out to Red Hat, love you all. I know I said I was retiring, but I couldn't get away from having the data lake and the information. And I love what you guys do, and I always wanted to be a research analyst. I wanted to be the person that said, oh yeah, that's the next big thing. And in order to do that, I needed to have continued access to the data sets that I was mining to manage the open-source community. You should do what the Red Hat, you can come join theCUBE. Oh well, let's see. You hear that? I do. No, no, no, no, no, no, no, no, no, no, no, no, no. Yeah, travel is not really my gig anymore. But I'm looking at booting up a research and advisory services that leverages the data that is in this biturgy data lake. So if you're interested in finding out more, give me a shout. Awesome. Diane, thank you for coming on theCUBE. Great insight, thanks for coming on. Okay, day one here at the open-source summit, I'm John Furrier. I'll be back with more after this short break.