The Many Faces of Cloud Native | Cloud Native Now 2023

Hey, everyone. Welcome to Cloud Native. Now, you know, uh, the, we've been doing virtual events around the cloud native ecosystem for probably five years here at Techstrong.

Um, and I think this year's a really pivotal year. The theme this year for Cloud Native now is the many faces of cloud native. Um, you know, the cloud native marketplace, the cloud native community has, has really just taken off over these last five, six years.

Covid couldn't stop it. It's back stronger than ever. I couldn't think of a better person to talk about the evolution of this community than our, our guest today, our guest of Art, a keynote speaker.

Chris, I'm gonna try to do the real Pronunciation. Yeah. An anesthesia, but I probably messed it up.

Well. We'll, we, we'll, we'll, we'll take it. We'll take it.

Hey, Chris. Welcome. Welcome to Cloud Native now, and thanks for coming on today and sharing with our audience.

Definitely happy to be here. So, for those who don't know, Chris is the long, he's been CT O of C N C F for as long as I remember. C N C F.

How long has it been Chris? Uh, technically it's probably been a little over six years, but I was kind of the original executive director also when the organization started about eight years ago. So eight years in December will be our, uh, or in December, eight years, it'll be the year anniversary since we held all our first board meeting for, for cncf.

Yeah. And it's, it's, and what a, what a Australian, what a long and successful trip. It's been Not even strange, Chris.

So I wanna jump into it. Yeah. Our audience is, is eager to hear all your views on this.

Look, the, the point of this is when I first started covering, I think my very first cube con, cloud native con was up in Seattle, I'm sure. Mm-hmm. And, um, yep.

It was all Kubernetes all the time, basically back then. Right. 8, something like that, had just come out.

Yeah. And, And a lot of people thought Kubernetes was synonymous with C N C F. And C N C F was synonymous with Kubernetes, kind of like all spaghetti macaroni.

But I think what we found out is that all, not all macaroni is spaghetti. Right. And there's more, There's more to cloud native than Kubernetes.

Um, yep. One of the big, I think, kind of, you know, red letter proof of this is this whole observability market that sprung up and it's really sprung up on the backs of, you know, two or three C N C F, uh, projects that are just, I mean, they're monsters. They rival Kubernetes in, in size.

Yep. What's your views on, on, on this? Yeah, it's, uh, it's a little bit fascinating.

You know, you know, kind of to your, your point where in Seattle, uh, in kind of the early cube con days, it was very, very kind of heavily focused on kind of Kubernetes. And then obviously there was a lot of discussions we had of, you know, do we want to, you know, double down on Kubernetes? Do we want to support a wider ecosystem of projects that kind of help, you know, get people closer to Cloud native?

And, you know, obviously we went for the ladder approach and started to kind of grow ecosystems and networking storage solutions and, and observability. Uh, now, and if you kind of look, you know, at this, um, you know, you know, maybe you can kinda link in, in, uh, your, your, your notes here is, you know, we do this kind of open source velocity diagram that kind of shows the breadth and size of communities, right? You know, Kubernetes obviously, uh, the, the largest in terms of kind of unique contributors on a annual basis, but closely behind it, you have projects like, uh, open Telemetry, you know, PR Prometheus and so on, which are obviously, you know, in the, this observability space.

Uh, and their communities are, you know, pretty not, not too far off on the size of, um, you know, Kubernetes. I think what we're essentially seeing is, you know, if you go to Q Con and you kind of go poll people, why are you here? Right?

What's going, you know, and a lot of people are already kind of bought in on the Kubernetes, you know, aspect or their migration to Cloud Native. They're looking for kind of the next step, right? And the next step generally is improve my observability, right?

Improve my security, improve, uh, developer experience. And so, uh, we're seeing a ton of interest in, in, you know, the observability, uh, market in general, you know, a, a big power of C N C F as generally bringing people together and trying to either build a defacto standard or try to like solve, you know, uh, let's solve, it's called problems across maybe different vendors. And so, if you looked at Open Telemetry, for example, um, our probably second largest project now in C n ccf, that actually started out as a multiple set of projects in, in, uh, in the ecosystem.

We had something called open tracing, which really just cared about the tracing, uh, aspects of, of the observability problem. And then there was an external project called Open Census, which kind of came from Google that was mostly focused on kind of metrics, uh, and a little bit of logs. And obviously we have things like Fluent Bit and so on that are, are, you know, kind of very, very widely used logging, you know, aggregator, uh, solutions out there.

And so Open Telemetry was birthed as, you know, uh, you know, trying to unify a lot of these things and kind of having one way to do metrics logs and trace traces to kind of simplify, simplify things because our end user community was super interested in helping, you know, solve this problem. So they just have a little bit more leverage with some of the vendors they're using and are just able to kind of modernize their observability, uh, stack. And, and we've seen Open Telemetry become very, very successful in, you know, a few years.

It literally kind of came out of almost nowhere, very, very small community. And now has, uh, grown to be our second most active project in SO'S bigger than Prometheus. Prometheus Already.

Oh, it over to Prometheus. So, so, so it depends how you define, uh, you know, it's like how, how you measure things, right? And so, you know, if you look at the open source velocity, uh, diagram in terms of unique, uh, contributors, uh, it is already larger than, uh, Prometheus.

It has a lot more kind of unique authors and organizations contributing. I'm not bought in on, um, adoption is always tricky to measure, right? Whether it's download or what people run in productions, Prometheus for sure is, is widely, widely, uh, you know, de deployed.

There's no question. Open Telemetry is a little bit more kind of like plumbing in many ways. So it, it probably has a different, um, you know, deployment profile.

And, you know, obviously a lot of solutions out there, like your Datadog and Splunks or whatever, they'll support some aspect of, you know, being open tele elementary compatible. So it's a little bit more difficult to mentor. But in terms of like unique contributions, um, it is, is second largest.

Yeah. You know, you know, by part with The success of, of Open Telemetry, actually the success of all of the observability related Yeager, uh, yep. Uh, projects from C N C F, it speaks to the model, which is you can get a Datadog, you can get a New Relic, you can get blogs io, you can get all of these companies collaborating, even though they may compete in the market, that base level, because they all recognize that these, these projects are the, the guts, if you will, of, of, of so many of these commercial applications today.

Totally. And, and it's kind of amazing to see the initial focus, you know, in Open Telemetry was very much on kind of logs, traces, metrics. And now everyone's kind of together in this, in this, let's call it collaborative room of trying to improve, uh, you know, observability.

And they're like, why not something new? Right? And so, you know, one, well, you know, one other problem that a lot of people are facing is, you know, large enterprises generally have not just one observability solution, right?

They tend to have multiple, right? Maybe they have a sim from back in the day, and they have some maybe new stuff, maybe a little bit Datadog here, maybe some new stuff from like a startup like Logs audio, you know, it's a mix of everything, right? And what, and what ends up happening.

They're like, why do these all have like different crazy query languages, right? Like, why is there Splunk QL versus like what New Relic has? And so the folks at Open Telemetry and the CNC of community are kind of getting together with this, uh, what they're calling like a query, query language standardization effort, right?

Working group that they're kind of coming together first where they're documenting all the different ones. They're seeing the compatibility. And eventually that will most likely lead to like a spec or some type of project out there, probably part of Open Telemetry to help standardize, you know, query languages.

So it's easier to build tools across all these different observability solutions, so on. So to me, it's like the, the beauty of, you know, once a community's together and fairly working out, well, they start, you know, coming up with cool things like a query language standardization or, you know, another, uh, you know, big thing that's going on in this crazy economy is people care about cost management, right? So we have a, yeah, open cost project in C ncf, but like that, like, what do you need to do cost management?

You need a solid observability solution, right? So there's kind of this blending of cost and observability and, and, you know, you're, you're seeing this in the marketplace where traditional observative vendors now are also offering cost management or even dipping their toes in the security, uh, or traditional sim space, because how do you do good, good security without good observability? So it's kinda a fun little, um, well, I, I think that thing that we're seeing, Yeah, I think it's speaks to the, the utility of observability.

I mean, yeah, look, a lot of people use it like Datadog, you know, for man monitoring it, it's obviously a monitoring tool. To me, security is a huge driver as, as security is a huge driver of so many things today of observability replacing sort of my grandma sim and what have you. But yeah.

Um, It, but it's such a, a, a u, you know, Swiss Army knife, a utility that, you know, anything you want to grab with that. Um, yep. The other thing about it, though, but by doing it this way, Chris Yep.

He's accelerated the development, right? How long would it take, like the various pieces, the bits, if you will, of open telemetry to, to have coalesced into a commercial product and, and, and to be so widely, you know, recognized would it happen without this kind of model? Right?

Yeah, I know. It's super fascinating. Yeah.

I mean, it's a great, it's a business school, you know, kind of, uh, Study, but even this query language thing, right? Yeah. How long did it take for sql?

And even today, everybody's SQL's a little bit different, you know? Yep. And it's, Yeah.

It, you know, the Unix wars all over again, right? Um, so it, it, it really is, it's, it's an amazing story when we look at observability. You gave us a great example here of now it's being used perhaps for, you know, what we call finops and cost management and all that.

What a, what a great use it is for that. Yep. What else?

I mean, do you cons, do you still think you were going to, like, is the, is it still hockey sticking kind of the growth of Open Telemetry, Prometheus, et cetera? So Your point about what, you know, what, what's, you know, is it still hockey growth on the observability side and Open Telemetry and some of these projects? So, you know, I think Open Telemetry for sure, because they're, they're entering new spaces.

So I mentioned, you know, the query language standardization effort. There's also, you know, we had metrics logs and traces, but, um, you know, as you kind of do deep profiling of applications, you kind of profiling data, right? That's also not very kind of standardized.

And there's discussions happening there. So like, once you kind of have, you know, new pillars added to open telemetry where things like profiling data or query language standardization and efforts, that causes a lot more growth and interest, uh, in, in the project. So I think you'll definitely see growth there.

And I, and I do think there'll probably this be this kind of delicate dance between, um, some of the cost management standardization efforts and some of the things like open telemetry, because, you know, there is rough kind of overlap, you know, um, you know, here in some ways. So you'll probably see some collaboration between, uh, you know, projects like open costs, uh, the focus effort outta the finops foundation and open, um, you know, telemetry to kind of make this stuff easy, um, you know, to use. And, you know, q uh, you know, if I just look at straight up Q con, you know, data from like, you know, people we survey observability, uh, and security are the two hottest, um, you know, topics, you know, and, and they usually rank one or two and they kind of switch, uh, sometimes between, uh, between spots depending on the, you know, the, the, the weather and the mood of, of the attendees.

So I was, you know, I was out in Amsterdam, obviously. Yeah. And I would say that absolutely was the case.

Yeah. I'm looking forward to Chicago and see maybe sometimes Europe can be a little different than North America. Totally.

Totally. Yeah. We'll, we'll see there.

Hey, let's jump into our next topic, Chris, and that is cloud native, beyond the cloud. So it's kind of funny to think about cloud native and cloud native computing foundation. Yeah.

Beyond the cloud, right? Without the cloud. But, you know, you've got the edge is, is really exploding.

And whether you want to say the edge is part of the cloud or the edge of the cloud or whatever, it's not, it's not the cloud, it's not AWS or Google or Microsoft or public Cloud in any event. Um, and then you have these C N C F projects that really are running on the edge on IOT devices, sometimes on endpoints on, in private data centers, right? I don't know if anyone has server closets anymore, but I'm sure there's probably some stuff running in server closets somewhere.

Totally. Um, yeah, Wasm is a great example of this, right? Web assembly.

Yep. How, you know, what do you view on this? Is there, is it still cloud?

If there's not, is it still cloud native if there's not a cloud? Yeah, it, it, it's, it's a funny term, right? Cuz you know, I think if you kind of look at the history, I think it would, the cloud native was kind of coined by the pivotal, you know, cloud Foundry community back in the day.

But, you know, it's kind of taken on its own life. If you look at the actual, you know, D N C F technical board, you know, stewarded definition of what cloud native is, it kind of comes that, you know, cloud native is all about, you know, empowering organizations to build scalable, you know, applications and dynamic environments, whether it's public cloud, private cloud edge, and you know, things like container service methods, you know, things like open telemetry, they're all kind of implementations of that. But it doesn't necessarily mean you run on like one specific cloud, whether it's public or private.

It's all about running across these different things that enable you to kind of deploy, uh, and ship scalable, um, you know, software. So we kind of look at the stack inside of C N C F. We have a lot of projects that, you know, I think enable cloud native principles, you know, that, that I talked about that necessarily don't have anything to do with Kubernetes, right?

You look at things, um, you know, for example, we talked open telemetry. A lot of people use Open Telemetry without even having Kubernetes in, in, in, in the mix at all, right? It's just a, it's a generic kind of way to do, uh, you know, modern observability.

You have things like, uh, Argo right? You know, Argo's becoming very popular and kind of the get offs and, uh, inversion, you know, AIC style of community, uh, out there. And, you know, a lot of people are just doing their workflow and then management of apps without, you know, even without even Kubernetes in, in, in, in the picture.

It's just like, it's, it's just part, it's part of the cloud native journey of getting there. I think another good example is, is backstage, you know, backstage mm-hmm. Has kind of popped up recently as just a, you know, a lot of companies, when they get to a certain size, they need a central place to go to to look at what services are running in production, who owns them, SLOs, and kind of like integrated developer experience, you know, options and kinda write a new service and kind of backstage solves, uh, that problem really well.

And like Backstage is like a funny project because we've, uh, you know, it, it's the one that kind of constantly comes up where if companies that reach out that they're not even using Kubernetes yet, but maybe they're like an airline that just wants to improve their internal developer experience, and they're already like, full backstage and they're just starting their maybe Kubernetes journey. So it, it is, um, you know, kind of an interesting thing that we're seeing, you know, we're seeing CFS this home of, you know, collaboration, vendor neutral cross cloud collaboration, where it's kind of, you know, burgeoning projects that, you know, are building, um, and enabling cloud native that just purely outside of Kubernetes. You mentioned, um, you know, web assembly, right?

I, I view, you know, wam or Web Assembly as essentially just like a new form of packaging compute, right? You know, we, we have VMs, we have containers. Wam is essentially kind of new form the package, you know, com, you know, compute almost similar to kind of a, you know, Java virtual machine where it could kind of run on different architectures, oss, and so on.

It kind of has its own a little optimized environment, uh, still early days technology, but we've seen Wam pick up, um, and be adopted in c ncf f projects in, in, in many different ways. And even, you know, uh, multiple years ago, like in, in a simple case, we had projects like Envoy mean, Istio replaced their, uh, existing extension mechanism that was powered by Lua, which is like a lovely little language and, and kind of runtime environment, but also not very, you know, super efficient, right? You know, there's a hit every time you have to kind of start a Lua instance web assembly.

Great. Perfect. You know, you could write plugins now in different languages and are all kind of, you know, cross compiles to this lovely web assembly bike code that could run everywhere, very efficient run times.

And so things like Envoy now support, you know, WAM, you know, as kind of its extension, you know, uh, you know, mechanism. But we're seeing a lot of new projects popping up in C N C F. We have things like Wam Edge, WAM Cloud, uh, there's even run part of container D where the whole, like plumbing of Kubernetes is starting to be rewritten to support running, uh, you know, WAM a little bit more natively than, um, kind of the current situation, which is a bit of a, uh, let's call it a, a call clu, uh, compared to, uh, maybe how simple it could be.

But that's like the evolution of technology. You look at Linux, Linux and Linux originally was never really intended to like, run in small embedded crazy environments, right? And then eventually people showed up and enabled it.

So you now have Linux in your car, your tv, your, you know, small little devices, right? So I think people show up to do the work and the technology evolves to meet new use cases. Absolutely.

And, and look, I, I think Wam, you know, it's funny, I was talking to, uh, was it Matt Cutler from Fairmont? Yeah. He gave Us The whole story about how he Oh, yes.

He Was the inventor of the, of the C NCF animals, right? Yeah, Yeah, yeah, yeah. D for his daughter.

Yeah. Yeah, yeah. But I mean, there is some, you know, look, he's a great ambassador for this whole WAM thing and, and everything going on with it.

It, yeah, it has a bright future, and it, and it does, I think it, but not just it alone. A lot of the other ones you mentioned as well, really, you know, to go where no C N C F has gone before, right? And, and, and Yep.

And where computing goes, that's where I think Cloud native goes. Yeah. And, and web assembly, just a very interesting to, it's still very early days.

You know, I try to caution folks, you know, if you kind of look back at the history of containers and Kubernetes, you know, in the early days, people sometimes forget that we had, you know, obviously we had Docker, Docker engine, but there was kind of a bunch of other upstarts. There was, core OS had a whole Rocket app, c spec, like different way of doing container images. So there was kind of this compatibility and multiple runtime issue.

Same things kind of happening in the web assembly space. They're, they're still kind of in that like, early standardization effort, um, of like getting everything fully standardized. You have too many runtimes out there, you know, naturally market forces usually end up with like one dominant solution or two, three dominant solutions.

It's still, those things are still happening. So I kind of caution folks, like definitely check it out, explore it. I think your team should definitely look at it, but it is very kind of early days, but technology moves very fast these days.

Matt and his company is doing amazing things to improve the developer experience, but there is this whole kind of standardization, you know, effort. And I, I think in the next year or so, the Kubernetes and C F C F communities gonna be, you know, really kind of, you know, almost a m native. Like, it's just gonna be in a good position to kind of support Wam as a, as kind of a runtime, uh, en environment.

Got it. You know, I think that's a great segue to our next topic, Chris. And that is the big dog Kubernetes itself.

Look, I remember a time, I guess it was a Docker con in Austin, and I go to DockerCon in Austin, and, you know, it was DockerCon was taken over by Kubernetes. Everybody There was just talking about Kubernetes. I came back to my office here.

Yeah. I was originally, when we did Container Journal, we were, we had already had the name Container Journal. I was actually thinking of trying to use the word daca, but that would've been trouble.

And I went with containers. Yeah, yeah, exactly. And I Came back, and that was the first time it dawned on me, maybe that wasn't such a great name for what we were gonna be writing about, you know?

And then, um, but the fact of the matter was, for those who, remember, Kubernetes was not the first orchestrator. And as a matter of fact, if you would've had a Las Vegas betting board of what was gonna become the dominant orchestrator for this stack Yeah. Kubernetes wasn't the favorite.

It would probably have been Docker themselves. Yeah, right. Swarm and and stuff.

Or Mesos, or for that matter, rancher os Yeah. And I mean, how many there was, there were a half a dozen easily. Yep, yep.

But the, the co the, the way the community coalesced around Kubernetes, and it became the defacto standard Yep. Is a great example of what you're talking about, which is, look, the, when it moves, it moves fast sometimes, right? It moved really fast.

Right. And I, me personally, I came back and told my team, you know, this Kubernetes, everyone's talking about it. It's hard as heck.

It's not easy. I don't know how people are gonna wide, how widespread people are gonna be able to use this. It's so hard.

But nevertheless, that's why I still work for a living. And, you know, I'm not off in some island. Um, but, you know, but every dog does have its day, and these things are life cycles, right?

Kubernetes is not the Kubernetes, it was in, I don't know, seven, eight years ago now, six, seven years ago. What do you think, Chris, where, where's Kubernetes on its journey, on its life cycle? What can we expect there?

Yeah, I know, I, you know, I think we're still at a point where, you know, sometimes you get biased based on kind of the, you know, folks that you interact with, right. You know, if you're talking to a lot of leading age companies, you know, you kind of see like, yeah, Kubernetes is everywhere, but, you know, know sometimes we'll, you know, host an event at the Linux Foundation where we talk to the automotive sector, or let's call it the industrial iot, you know, you know, let's call it very, very traditional enterprise. And then, and sometimes we'll talk to folks and they're like, yeah, we're just kind of starting our journey with like, you know, Linux open source, and we're trying to Kubernetes cloud.

So, so I think that the, the market is much larger than, you know, I think we, you know, kind of perceive sometimes. And so, you know, I think KU Kubernetes has kind of really cemented itself as something that's available in every major public, private cloud, fairly safe choice. And the technology kind of continues to evolve to meet use cases just like Linux did.

And, you know, are there other, you know, options out there? Sure. Just like, there are other options out in, in Linux, right?

Free BSD is a, is a lovely and great, you know, operating system, but it's a fairly niche, you know, you know, community, right? The, the most of the world has decided to back, you know, Linux as kind of the defacto system. But it doesn't mean there's other options that people, um, you know, could use.

Like for example, in the Wam case, uh, you know, there's, uh, a lot of initial, uh, uptick happened with kind of people using, uh, Hashi corpse, uh, you know, nomad Orchestrator, because, you know, nomads great, nomads great. It's a pretty great technology, and it's very easy to run a binary and orchestrate a binary where you don't have to do anything container related. And Wam at the end of the day, is producing binary.

So that's what people did. But what's happening now is people get together and like, you know what? We want to use Kubernetes, so how do we kind of make it better to support these use cases?

And that evolution is starting to happen with the run w Z project, product container D. So I think, you know, at the end of the day, like people will come together and make, you know, things, things work. So, um, you know, I think Kubernetes itself is kind of going through, if you kind of look at the releases every release, you know, it's generally APIs are kind of maturing.

Things are moving from like alpha to beta, to, to, to kind of, you know, stable sometimes pieces of Kubernetes, um, you know, are being simplified where I see, um, you know, sometimes we spin things out, right? You know, traditionally Kubernetes used to ship with a lot of, um, you know, uh, you know, uh, in class storage drivers or, you know, that kind of shift with part of Kubernetes that was kind of, you know, uh, essentially they did the work to kind of break that apart. And, you know, all the storage, you know, stuff exists outside out of tree, and now you have to kind of choose what you want.

Et c d was initially kind of spun off too, as a separate thing. So I think Kubernetes is cognizant that, you know, tra traditional software, any type of software development, the, uh, traditional thing is let's keep building features, adding features, adding more stuff, you know, to the core, just the way software kind of works, uh, Kubernetes and things like Linux have, I think, have done a pretty good job of historically of having some type of either plugin or module, uh, extension mechanism or created something new and kind of allow the innovation to happen outside the core. And I, I think you're kind of seeing that happen with Kubernetes, you know, over, over time.

And, and they're kind of mindful about that. So I think they're gonna have a healthy, you know, next, uh, let's call it decade, uh, for, for sure. I think the challenge will be, is like when the core, uh, languages potentially cause you maybe to rewrite certain parts of the stacks.

Like Linux is kind of going through this interesting evolution now where they're looking at how to support Rust as a popular memory, safe language, right? Because, you know, love c c plus plus, but definitely comes with its downside, for sure. Look, you, you're preaching to the choir, but, you know, flip side of it is varieties the spice of life.

Totally. And so the same way we see with Linux Distros, right? What packages come with a particular Linux distro, but you could still add, you know, optional packages.

It'll be the same thing with Cobe, right? And, and yeah, there are companies, all those companies that we were gonna bet on as an orchestrator that now adopted Yep. Kubernetes, they all have their kind of unique distro, if you will, with, with their own, you know, package Yeah.

Recipe. Um, yeah, totally. So, I, I think we'll see more of that.

You're, you're, you're correct. But I think, you know, if the, I think the, the, the, the core kind of stack of, of the cloud native stack with Kubernetes as its orchestrator, if, if that doesn't get upended, Kubernetes doesn't, right? Yeah.

You know, I think the technology, the underpinnings will probably change the api. If you kind of look at what Kubernetes is, at the end of the day, it's like, you know, there's a nice API server that makes it very easy to write extensible things that are kind of managed by Central or Orchestrator, that API server. Um, you know, it could be completely, you could have a new completely core that is built maybe in Rust or something else that kind of supports, you know, that api and you, we, we can move to a new technology set, but we'll see.

There's a lot of innovation. There's kinda like these fun projects called, like, there's kcp out there. Yes.

That kind of came outta Red Hat. There's, um, you know, even the folks at Cross Plaine and, uh, there's, there's others that are basically trying to use just like the Kubernetes API server to manage different types of resources. So I think we'll see some innovation, um, you know, uh, probably in the next couple years in, in that space where Kubernetes may actually have even a smaller lighter, uh, control plan.

But the API will be, will be, we will be consistent. Like I think we've all kind of agreed that having, you know, the, the Kubernetes API APIs is, is not, not too shabby. The underpinning underlying technology may, may, may, may change, uh, in, in, in the future, just like Linux is going with the whole supporting Russ modules, which is, you know, a bit of a crazy effort on, on their, on, on their end to kind of make that work.

I think it's meeting the network. You know why Yeah. We, we made it over a half hour here without mentioning it, but this is a good time.

Yeah. Everyone's talking ai Crazy. But when we talk about APIs is a great example, right?

Where, you know, you could start having Yep. Modules talking to each other, doing APIs to each other via ai. What, what, if any kind of impact are you seeing already or we might see in Kubernetes?

Yeah, I mean, if you, if you were in Amsterdam, you know, you, you probably saw, you know, there's a, you know, maybe a handful or two, you know, kind of startups that have, have, you know, either popped up or pivoted to kind of support, you know, ai, DevOps or ai, you know, management workloads. Yeah. No, and, and I think there's, there's, there's many different approaches here.

I think it's still very early, um, you know, days whether it's like, Hey, maybe I could use AI to make it easier to write like an OPA or security policy. Maybe I could use AI to help with remediation, maybe AI as like a, a co-pilot or buddy of mine help me debug, uh, you know, an issue, right? So I think we're still kind of in that innovation phase, you know, the models and infrastructure is changing a ton.

You know, we have a lot of folks at the Linux Foundation this week at an AI conference in the Bay Area, hosted by, I think Databricks and, um, uh, couple others kind of in the space. So it's like, this stuff is quickly moving and evolving. There's, there's all sorts of legal issues that are kind of being ironed out in the, in the court system, but at the end of the day, developers are gonna use stuff that makes their lives easier.

So, like, if, if the developer experience is improved, if developer is able to save time, they're gonna use these, um, you know, tools and kind of build their own, uh, you know, variations, the productization of these things, uh, is a whole interesting, you know, thing. Like, how do you charge for it? Are people gonna be willing to pay for it?

I, I, I don't, you know, I don't know yet. Like I, I'm, I'm, I'm still like a casual observer. Um, you know, we're using a lot of kind of AI tools internally, uh, at, you know, C N C F and a Linux Foundation to kind of help, you know, help with our marketing copy, help with like, you know, kind of like a more, more, I think, uh, you know, use cases at the end of the day.

Yeah. So It's the same here. Hey, you know, this is a great transition to our last topic, Chris, cuz I want to be mindful Yep.

Of, of your time as well. Sure. And that's the, the future of the C N C F itself.

Man, I was at the, in Amsterdam, I went over to the C N C F store to pick up, I always look to see what's marked down. I bring my kids home, t-shirts, sweatshirts, they love them. Um, yeah.

But, you know, the, the amount of shirt, you know, they got a shirt for every project. You don't have enough space in the store for every project anymore. Yeah.

And, you know, how many projects are there in C N C F now? Is it unruly? You know, you could group for my families.

How do you organize this? Yeah, it's, it's interesting, right? You know, I think, you know, uh, you know, we have a little over, I think it was like 162 projects now as of today.

And, you know, we have over 200,000 contributors spanning 108 countries. And if you kind of look at, what's interesting is for historically a long time, you know, Kubernetes made up the majority of the, the contribution ecosystem, which I think right now there's about 80,000 or so contributors to Kubernetes, cumul, you know, cumulatively, if you look at C N CF now, it's 200,000 total. So Kubernetes now is actually a, still a huge part, but it's no longer the, the, the majority, which kind of has shown the growth of the, the ecosystem.

You know, how do you manage it all? I, I mean, I think it's just our philosophy of, you know, we, we enable, you know, uh, competition across our projects. And sometimes there's, you know, good competition where people get together and innovation happens.

You get better solutions. Sometimes competition leads to the convergence, you know, of projects like Open Telemetry, which I think benefited everyone at the end of the day. So as long as we kind of basically enable market forces, you know, and, you know, typical market competition, which bubbles up the best, you know, solutions, um, you know, I think we'll kind of continue to see innovation happen that does come with a downside.

You know, I think a lot of times, you know, we have companies that come to us. It's like, I would just love to download a C N C F distro or cloud native distro that everything is just like batteries included in work. Then I'm like, that's not the purpose of our organization.

I think if you do that, yes, you have something that's a little bit easier. So something like a Cloud foundry, uh, or an OpenStack for example, right? Mm-hmm.

OpenStack is mm-hmm. Free technology. But OpenStack was like one way to do networking.

You use neutron, right? You have swift, you know, do storage. And when, when you only have like one way and you, and you don't allow for other competitive things, you kind of potentially lose out on, you know, technology shifts.

Like, you know, for example, networking in Kubernetes, you know, uh, didn't really kind of take advantage of like E B P F technology, you know, back, back, you know, back then Eeb p f was something that existed for a while, but never really, you know, got a lot of steam until recently. And now the majority of kind of, I would say networking or c n i compatible technology is all using, you know, EEB p f Like, if we kind of locked ourselves in with just one way, uh, that evolution may have taken much, much, much longer. So, I, so I think, uh, you know, it comes at the cost of confusion, uh, sometimes of like, there's all these different, you know, things and they're like, oh, you call like Cube con and Cloud Native con, why do you do that?

You know, or, uh, yeah. I was just in, uh, you know, Israel, uh, you know, last week for our, our first Cube day in Tel Aviv, and it was actually funny talking to folks where they're like, I thought Cube Day was only about Kubernetes. And I'm like, no, no, no.

Like, you know, people use Cube and Cube Con as like almost a shorthand to mean, you know, everything cloud native. But I, but I could see why it's confusing to, to folks, you know, on, on the outside. And I think that's, that's the, that's the bitter pill that we take to support this kind of innovative future tech ecosystem of, you know, competing and collaborating projects.

It gets a little bit confusing cuz you have too many, too many choices. Um, you have a Lot of rockstar. I mean, you mentioned dopa, Oprah itself is a, yeah, there's a whole world built around Oprah these days, man.

It's crazy. Oh, To totally. And, and so our stance is like, enable this and, you know, you know, experiments, sometimes things work, sometimes things don't.

Like if you look at CubeCon, one thing we kind of do is we always rotate program chairs. Right. You know, and we always kind of have, they, they, they, they leave their unique take on the conference.

We had, um, Kelsey Hightower, if you saw recently, uh, graciously retired from Google. I'm so happy. So happy for the So learn To live as he says.

Exactly. So he was one of our early program chairs and he left, you know, his unique spin on the program. We had Liz Rice, you know, she left her unique spin.

And we have all these folks that kinda in, in different little areas that kind of le leave their own take and, you know, uh, thing. And I think that that's healthy and that's why kind of Cup Con has kind of, you know, evolved. And like the other thing that we do, we sometimes experiment with new events.

Like we have something called App Developer Con, uh, that was, uh, pushed by a certain segment of our community that was like, we love CubeCon, but there's a lot of infrastructure folks here. Like, we want to reach app developers cuz we have technology like, you know, dapper or G R P C or other things that, you know, hit that application developer more. And sort of like, alright, well do app developer con, see what the hell you know happens there.

Maybe it works out, maybe it's a failure. And well of Course you got the, the other shoe of that is, well now you're gonna have to do OpCon or something like that. And, and, and to be fair, I saw it in, in Amsterdam.

This Amsterdam was the first Q com CubeCon where I felt it was, it was more balanced mm-hmm. Between op and dev. Yeah.

Right. com, you know, there's a constant battle for us here cuz we try to Yeah. We try not to play favorites.

Totally. But I, and I thought, I thought there was a really good balance where I think previously it was very app-centric. Totally.

Totally. And I think it's, you know, it over time it improves and, and like we share the data with everyone and I think to your echo your point, like the data definitely shows that there's a kind of a lot more balance between like infra op upside and kind of app, you know, app dev side than, than previously. And I think that's just reflective of, of, of kind of the community and the technology that kind of continues to, uh, you know, evolve.

But like we experiment, like we used to do serverless days or we had like serverless things and we kind of don't do that anymore because the community kind of lost interest and kind of moved on to the next thing. So like, we kind of pop these things up and, and see what works and, you know, understand that it's a lab. Yeah.

Sometimes it won't, it's a lab, sometimes it won't. Yeah. Sometimes they don't.

But you, you know, you fail fast. Yep. And you learn and you iterate and, and that's, yeah.

That's the whole point of this, right? That's live and eat your own dog food. And, and, and to, and, and to be honest, you know, we, we look forward to like folks like yourselves and like our, our wider community that does events.

Things like cloud native, you know, now, and we have these little Kubernetes community days that are kind of like, like DevOps days that have sprinkled throughout the Yeah. The community. So seeing folks kind of enabling them and, and you know, helping them, connecting them with speakers or projects, something we kind of love to, you know, you know, see cuz this community is truly global and in, you know, many, many places.

So it's kind of great to kind of support folks like yourself, uh, with Appreciate It, Chris. That's a great way to end this. Can't thank you enough for, for coming on and sharing with our audience today.