Sovereign Cloud: Who Really Owns Your Infrastructure? • Jake Warner & Charles Humble

Speaker A: Welcome to the Goto Podcast. Each episode covers the brightest and boldest ideas from the world's leading experts in software development. Tune in for practical lessons, compelling theories and plenty of inspiration. GOTO gathers the brightest minds in the software community to help developers tackle projects today, plan for tomorrow, and create a better future. Stay up to date with the latest in tech through goto's top rated events in cities like Amsterdam, London, Copenhagen and Chicago. And by subscribing to the Goto Conferences YouTube channel where you can find thousands more high quality dev talks. Learn more at Gotopia Tech.

Speaker B: Hello and welcome to this new miniseries for Goto called the State of the Art. This series will explore generative AI, but also other emerging trends in things like platform engineering, new practices, new programming languages, and in this particular episode, uh, sovereign Cloud. I'm Charles humble. I have 30 years experience as a programmer, an architect and a CTO, and I'm currently working mainly as a freelance consultant and podcaster. I'm joined today by Jake Werner. Jake is co founder and CEO of Cycle. Jake has nearly two decades working in data centers, infrastructure automation and DevOps. From contributing to early OpenStack projects to leading development and one of the first bare metal infrastructure as a service providers, his background is rooted in solving complex infrastructure challenges. Jake, welcome to the show.

Speaker C: Thanks Charles. Happy to be here.

Speaker B: It's lovely to have you on. So let's start at the top. What is Cycle and what problem are you trying to solve in the infrastructure space?

Speaker C: Yeah, so, uh, cycle is. Well, so Cycle originally started as a alternative to Kubernetes. It was just mainly focused on how do we allow companies to take containers, be able to piece them together. Kubernetes, but without all of the extra complexity that typically comes with kubernetes. Now, over time, Cycle has continued to adopt or Cycle has continued to grow. Now you can run, you know, virtual machines, functions, et cetera, you know, on top of the platform, but from a high level, it's. It's a distributed control plane that allows companies to bring their own infrastructure, bring their own storage, bring their own compute, et cetera, own all of that, and then be able to run containers, virtual machines, you know, with, with ease.

Speaker B: Okay, so pulling on that a bit more then. If Cycle started as an alternative to Kubernetes, what were the kind of limitations of kubernetes that pushed you to build something different?

Speaker C: Yeah, so this kind of goes back to maybe my OpenStack days. So I was involved with OpenStack when OpenStack went from no one really knew what it was to everyone wanted to use it to where it became so complex that people kind of didn't want to touch it anymore. And so with Google releasing Kubernetes about a decade ago, I was kind of this, this long term bet that we would see the same thing happen eventually where companies would adopt Kubernetes. They would, you know, again, I mean we all know developers, they love to test things, they like to uh, change, you know, all the switches and levers. They, they, they enjoy breaking things, putting them back together, et cetera. Like, I completely get it. And so the bet there was that organizations, uh, would adopt Kubernetes because it was, you know, the, the next great hot thing. But then at some point these same organizations would kind of take a step back and say, like, is this really what we should be spending our time managing? Like, is this what we wanted to really focus as one of our core competencies, or should we spend more time building our products, et cetera. And so Cycle was built kind of in that vein of how do you allow companies to still have the same ownership and control that they would typically get with Kubernetes? Um, again, meaning that they own their infrastructure, their networks, et cetera, but still have like a platform like experience on top of that infrastructure. So you know, I guess we can kind of, I mean, I know Heroku is not kind of the, it doesn't have the appeal that it once does. But if we wanted to kind of overly generalize, you know, you can think of Cycle as Heroku, but on your own infrastructure.

Speaker B: Right. And that's, I think one of the very interesting is one of the things that I found really interesting about Cycle is that it kind of goes beyond containers, right? So you're supporting VMs, but you're also supporting bare metal and functions as a service and so on. So can you maybe talk a bit about what the philosophy behind that broader approach is?

Speaker C: Yeah, so when we started building out the platform, it was. Well, originally our goal was to build a containers as a service platform. So where we owned all the infrastructure. There's a number of other companies that are doing this now. But you can also think of products like ecs, right? Uh, so originally our goal was to own all of the infrastructure, make it super easy, and then companies would just pay us per minute of runtime for these containers. But as we started building the product, this is many years ago at this point, as we started building the product, we started hearing back from larger organizations that we're talking about a couple things. Number one is they were interested in a platform that would allow them to keep things simple but have a higher ceiling. Right. It is very common that a company that adopts a platform adopts a ceiling and once they hit that ceiling, they have to then replatform onto something that allows more control. Right. And so companies where we, we had a lot of feedback for companies saying like, hey, we want something that is simple, but it's really important that we are also adopting something that we're not going to grow out of a year from now. And so through all that feedback that we got, that's when Cycle started to. Or that's when we kind of went back to the drawing board for Cycle and decided that we wanted to change the uh, structure of how we were building the platform so that again companies could own their infrastructure. Um, that was one of the big common asks was if we want to, if we want to be able to have a high ceiling in terms of adopting this platform, we need to know that we can choose our providers, we can choose our infrastructure types, we can customize that infrastructure accordingly. And so that's kind of how that all came together was being able to have a high ceiling but still not have to sacrifice simplicity, uh, uh, uh, in that effort.

Speaker B: It's so interesting that, because I think that's uh, a trade off that we just see playing out over and over and again in the industry in different places in different ways. So, you know, a lot of my professional programming was in Java and there was a period of time when Ruby on Rails was coming up and Ruby on Rails was deeply fashionable for a period of time. And one of the reasons it was deeply fashionable was because it was incredibly easy to get uh, a crud web app up and running really, really fast. It was fantastic for that. But then at some point you find something that you couldn't get Rails to do and then you were uh, sort of banging against the edges of what was physically possible, I think with uh, the framework. And it's kind of, you alluded to Heroku and it's kind of the same thing we saw playing out with Heroku and the sort of whole platform engineering space more broadly. There's this kind of push and pull between having something that is architecturally flexible enough whilst also still being easy enough to use. And uh, it's just a really hard, a really hard line to tread. I think it's a really difficult thing to get right.

Speaker C: Oh, ah, absolutely. And it's one of the kind of sayings or sayings, but one of the things that we Talk about internally within the team and within our community and our ecosystem is just in terms of that ceiling. You could build Cycle on top of Cycle to show you how powerful the platform is. And that was one of the big things. A lot of people don't know this, but Cycle originally started out as like a Squarespace competitor. Like it was all about building websites before we were even talking about containers. It was just, you know, about building websites. But I think even in the early, early days of Cycle, I knew that I wanted to build a platform that could run anything. Like, yeah, it was nice working with e commerce companies building websites, but I wanted to be able to allow companies to build their own platforms on top of Cycle. And so that's why for a while our. One of our taglines that we were testing was uh, the platform for building platforms. Right? Implying that like Cycle is not really built for companies that are wanting to just throw one or two containers up, you know, and you know, host those anywhere. I mean there's a number of places to do that. But Cycle is built more for companies that are saying, hey, I have two dozen different microservices that need to talk to each other. I have different hardware and infrastructure requirements for different parts of my platform. And so we really wanted to make sure that the platform was positioned for those sorts of organizations where there is some level of complexity inherent in the application or in the project that these, these teams might be building. And so it's one of those things like we've had companies move from Heroku to Cycle, we've had companies move from Netlify to Cycle, we've had move from Vercel over to Cycle, but we haven't seen it go the opposite way. And I think that is exactly what we all have experienced back in the early days, or not early days, but over the number of years with Heroku, you always heard about companies kind of reaching that ceiling and needing to replatform at that point, right?

Speaker B: Yeah. Now the kind of big news story here and the reason that I wanted to get you on is the announcement of a European control plane. Um, this is super exciting to me because something that I am having a lot of conversations, kind of off the record conversations with people, either as a journalist or through my consulting work, is a, uh, sense that Europe, uh, is overly dependent on U.S. technology companies for its software IT infrastructure. Um, I just really want to get into this conversation because I think it's really interesting and one of the things that's really hard about getting into this conversation is it's really hard to find people who are willing and or able to talk about what they're thinking and doing on the public record, which is why this is so exciting for me. So you have announced a European control plane. I want to get into what that actually means. But before we do, could you maybe start by telling us a bit about what, why you're doing that? Uh, what prompted that move?

Speaker C: Yeah, so. Well, first I wanted to kind of like, dive into our use of the terminology around control plane, because for a long time, we didn't really use that terminology with cycle, but over the last few months, it kind of started making a lot more sense. So originally we called cycle a platform, right. But I think everyone started using the term platform. So platform kind of lost a lot of its meaning because there's a platform for everything. And then at the same time, we started, you know, we started like, thinking about, like, what other words are there that describe what we do? And control plane, you know, kind of bubbled up to be the best word to describe that, just in the context of we don't own our customer infrastructure, but it connects back to the control plane. So, like, if you're thinking about it kind of as a decentralized orchestrator, that control plan, it kind of makes sense there, but to then, you know, dive into your question. So, yeah, so for. Since Cycle started bringing in customers back in 2019, our control plan has been just a North America control plan. So we've had infrastructure all around the US And Canada. Even though our control plan was spread across the United States and Canada, that didn't prevent our customers from being able to deploy wherever they wanted to. Right. We've had a number of companies over in Europe for a number of years now. But so again, the control plane is just like where telemetry commands, orchestration, those sorts of things happen. Right? All applications, all, uh, customer data sits on their infrastructure. So that control plane did not actually prevent customers from deploying over in Europe. But with all the geopolitical things that have really kind of started to happen over the last, I guess probably a year and a half, although some would argue they probably started way before then. But I think in the last year and a half, especially in the last five or six months with kind of the whole fiasco around Greenland, then it became a lot more real for a lot of people. And so that's where we started having companies that were already customers of Cycle, saying, hey, I know that I own my infrastructure, um, but there's a business risk to even my telemetry data Going back to the United States right now. And that is something that like, I can, I can deal with it, but I'd like not to. And so, you know, the first couple of times this came up, it was kind of like, uh, you know, like that makes sense but you know, like it's, it wasn't super critical. But when enough of our European companies started reaching out to us saying like, hey, how can we segment this so that we are entirely um, outside of the United States, that's when the decision was made on our part to, you know, to go and spin up another copy of Cycle, um, specifically to, you know, to run only on European infrastructure. And so this way there's, there's full segment or it's fully segmented between North America and Europe. And so while there's a number of benefits of doing that for like data sovereignty, and I'm sure we'll dive into those in a minute, there's also other perform or there's also other benefits around performance m and things like that. So yeah, it was something that we had a number of companies asking for and we're thrilled that it's now live.

Speaker B: Yeah, no, it's incredibly exciting. It's interesting to me. So do you see this as a, uh, because obviously the content you mentioned, the geopolitics, right. So the context of this is the geopolitics, particularly in the U.S. so do you see this as a kind of a short term reaction to that or do you think this is the beginning of what may end up being a sort of deeper, uh, long term change in how organizations in Europe think about their infrastructure?

Speaker C: I think that we're going to see it in waves. Right. So I think we just saw a short term very intense wave and then it kind of died down as the conversation around Greenland kind of paused. But I'm sure as time goes on there's going to be a number of other events that are going to just kind of keep bringing the attention back to it, which is going to drive long term change. So it's people reacting to news, you know, kind of taking a breather, something else happens, people take a breather. But over a long enough time, more and more, you know, companies, whether it's, you know, on the US side or on the European side or you know, wherever they, they may be geographically, uh, these companies are going, I think, I think data sovereignty is a good, uh, a good thing for everyone to have. I mean that's like, I know right now we're talking about the EU control plane, but if we Go back to the previous, uh, conversation about what we built in 2019. It was kind of the same, you know, idea, right. With allowing customers to own their infrastructure. Like we've always thought that people owning everything should be like, is kind of one of our core tenants. And so being able to allow people to have even more control over where their even telemetry data and things like that are going kind of resonates from that perspective too. So yeah, I think it's kind of short term waves that are going to create long term movement.

Speaker B: Right? Yeah. Um, so Cycle is still a US founded company. So how do you build trust with customers who are explicitly trying to reduce their exposure to US Influence?

Speaker C: Yeah, so Cycle is still a US company. We're headquartered in Nevada, but with all the things that were happening kind of geopolitically, like the writing was on the wall. And so late last year, my co founder and I, we moved to Iceland. So a lot of that was for personal reasons. My wife and I had honeymooned here and you know, really enjoyed the country. So back when we had started dating many, many years ago, she told me that someday she would love to live internationally. So that was one of the kind of ideas there. But it did align with a lot of what we were seeing from, from the business of uh, so many of the new companies that were reaching out to us were companies from Europe that were trying to get off of US owned hyperscalers. And so it kind of aligned there. So I think that just being here has kind of helped show European companies that we're serious about what we're doing. And then at the same time, I know I've kind of already kind uh, of poked uh, at it before, but the big component of Cycle is that our customers own all their infrastructure, all their networks, etc. So even so they still get to fully own everything. And then with what's going back to the control plane, it's mainly telemetry data. So the overall risk that I guess I'm talking about from the like versus like the US Cloud act is already relatively low because we don't have access to any of that customer data. But I think being launching the EU control plane physically now, you know, having residency in Iceland, I think are moves that show a number of European companies that it's more than just, you know, it's more than just talk.

Speaker B: Right? Yes. Yeah. So you talked earlier about the Control plane being fully segmented. So at a technical level, what does that actually mean? What are you doing differently from a, uh, uh, typical cloud Provider.

Speaker C: Yeah. So I don't know if it's different from a typical cloud provider, but. So I guess, number one, uh, we haven't talked about it, ah, yet. But with Cycle, whenever we have a control plane for a region, the control plan is spread across multiple providers. So it's not like we're just sitting on top of AWS or GCP or things like that. Um, we are sitting across many different providers. Now, the EU control plan is only sitting on top of European infrastructure providers, so that way they're not subject to the US Cloud act. Now, again, given that Cycle is technically still a US company, I guess Cycle would be subject to it. But again, since we don't own any of that data, we don't have access to any of that data, it'd be kind of moot for the US Government to come after it. But to answer your question, there's no overlap between them. There's no data layers, there's no network communication between the control planes. They are fully separate and standalone. So I know we're talking a lot about data sovereignty, but the other kind of benefit here is, and we've already had companies that are reaching out to talk about this is a company that, you know, is, is, is big enough that they're using Cycle to, you know, build a complex offering. And they might be so sensitive to downtime that even though they get to own the infrastructure, et cetera, et cetera, they can also now deploy across both control planes. So that way they're even hedging against if Cycle were to have an outage. Now, Cycle has not had an outage since, I think like March 2021. And you know, we, we put a lot of, A lot of effort and a thought behind how do we make Cycles resilient as possible? But when you have two fully independent control planes, if you were to deploy across both of those concurrently, I mean, if both of those go down at the same time, we as a society have a bigger issue, is kind of my point there, right?

Speaker B: Uh, yes. Yeah. I'm wondering as well, is there a. So there's been this sort of rumbling cloud repatriation story for about three or four years, and it's one of those things. I'm, um, not honestly sure how big it really is, but, but there are definitely cases of companies. I mean, Hay is a very famous example, but there are others. Dropbox, uh, was another. There are examples of companies that are bringing workloads back onto their own infrastructure, onto infrastructure that companies control from public cloud. So does Cycle kind of fit into that trend in some way as well, would you say?

Speaker C: Oh, absolutely. I would almost say that that's kind of a bigger trend than what we're seeing with some of the data sovereignty stuff. But again, we're trying to make everyone happy. So, you know, the data sovereignty with the EU control plane, you know, that makes people happy. But at the same time, I wouldn't say that it's one or the other because we've had some companies that are as part of. So let me take a step back and provide a little bit more context. So some of the companies that have been reaching out the Cycle due to data sovereignty issues have told us, like, hey, we can't just physically be on infrastructure here in Europe. We need to be on infrastructure owned by European providers, right? So, for example, we had a company that has been using AWS for 16 years and they came to us and said, hey, like, we're entirely on AWS infrastructure that is physically in Europe, but that's no longer good enough for us. We, we need to be out of AWS entirely. And we've had other companies say the same thing about Azure and gcp. And so we're seeing this, this pretty big move away from the hyperscalers. But these companies, while they're making these moves, are also starting to reevaluate. Like, well, have to move off of this infrastructure anyway. Like, why not, you know, evaluate other options. Um, and so since November of last year, so it's, it's May right now. And yeah, so since November of last year, I would say, I don't have an exact number, but I would say 70% plus of the companies of Adopt and Cycle have had a bare metal component. Now, whether that is bare metal that is sitting, you know, in a facility that they own or that is bare metal through like a NEO cloud provider or a COLO provider or something like that. But we've had a huge number of companies that are using this as a way of adopting, using Cycle as a way of adopting bare metal and seeing some pretty significant cost savings in the process. Um, so, uh, recently we had a company. I'm not sure if I'm allowed to mention their, if I'm able to mention their name, but they are in the advertising space and they were previously and they were using AWS before. I believe their compute spend at AWS was somewhere around $45,000 a month. And by moving over to bare metal, their spend now is, uh, hovering around 12 grand a month. So they were able to see some pretty significant savings while also still taking on double the workloads that they previously were.

Speaker B: Right. What are the trade offs that you're making? So if you decide to move away from aws, uh, or GCP or Azure or something, what would you say are the uh, trade offs that you're making there? What do you lose if you abandon the cloud effectively?

Speaker C: Obviously I'm biased so, you know, but I don't know if there's much that you would lose. And the reason I say that is we have companies that did you know, that were previously on AWS and some of them are still on aws, but they're using Cycletel Orchestrate on top of aws, so they're still able to use other uh, managed services through aws. Like we have a number of people using Aurora DB and RDS and things like that. And so they're able to still use those or you know, companies that are still using like, like Google Cloud, SQL, etc. So cycle doesn't limit people from using those different technologies and services. Our goal is just to take Raw Compute and orchestrate on top of that. So our goal is that by um, adopting Cycle, you shouldn't ever have to be in a position where you feel like you're giving up something.

Speaker B: Right? Yeah, I mean I would say there's running, it depends what kind of company it is. Running your own data centers, if you haven't done that for a while, isn't necessarily that straightforward. There's ongoing capacity planning. If you've not had to do that before, if you've not done it for some time, that can actually be, I would say, surprisingly, throughout, um, um, those of us that have been around a long time will remember having very long procurement cycles for hardware that we had to work around and buying a server and racking it and installing all the necessary software is less trivial than you might imagine. And um, I think managing a data center for the long term is not necessarily a trivial thing.

Speaker C: Oh absolutely. Uh, yeah. So if the context is, you know, if someone is moving from AWS to exclusively bare metal, you know, what are they gaining or losing at that point? Yeah, you're absolutely correct. Like there are other stresses that come with if you are owning the bare metal yourself. But in the general context of cycle being multi cloud, multi provider, et cetera, that's where I get a little bit more flexible with my answer. But um, yes, you're absolutely correct. If you're wanting to provision a rack or bare metal and uh, you own 100% of that bare metal yourself. Yes, there's Absolutely. A number of things that now you have to manage that you don't. And so most of our customers today, not all, but most of our customers are acquiring bare metal through bare metal as a service providers. So they don't have to worry about too much of the procurement or networking, things like that. They're just getting access to the, the direct machines.

Speaker B: Right? Yes. Yeah. Tell me about your. So before we, when we were speaking off mic, before we started recording, you mentioned that you were uh, partnering with Cherry Servers. Can you talk about that? How is that ecosystem kind of important for delivering your vision of what you're trying to do?

Speaker C: Yeah, so as part of launching the EU Control plan and with the context of the US Cloud act, it was important for us to build that EU control plan on infrastructure providers that were owned by European companies. And so Cherry Service was one of the first companies that we started talking to. Their, their offering was incredible, especially their bare metal. I mean it's wildly good for the price and so yes, so Cherry is one of the companies that is, that we're using for hosting the, the EU Control plan. And at the same time Cherry's also, we have an uh, integration with them going live sometime within the next 45 days. Um, and I'm really excited to be able to get that in front of our users because I think there's a lot of value and having again, just m. More. Well, number one is more providers in the cycle ecosystem, but then number two is having European or European owned providers. There's a whole data sovereignty concern that helps solve too.

Speaker B: Right? Yes. Okay, fantastic. What's next for Cycle? What are you kind of thinking about next? Uh, what are you kind of excited about for your future?

Speaker C: So next immediate in terms of like roadmap, number one is just being able to add more infrastructure providers. You know, Cherry's is the start of that. We have a couple other companies that we're talking to to add them and then beyond that we have, let's just say like a new monitoring, you know, system kind of going into cycle. So that way companies have better observability over their, their servers, et cetera. And then on the horizon, one of the efforts that we're working on is being able to release Cycle for free for people with home labs. So anyone who has like a server that they want to, you know, that they're not, you know, for professional reasons, you know, they're not, it's not part of a business, but they, they want to be able to, you know, self host, you know, GitLab or I don't know what, you know, whatever they want to host, giving those people access to Cycle for free. Um, and it's kind of the, the whole idea of like, hey, if, if, if Cycle is being built for all these organizations that are building really complex things, what will people use Cycle for in even in just a home lab situation where they're just kind of throwing it on a server for testing and things like that. So that's something that we're hoping that through those home lab efforts we'll get Cycle in front of more developers, even if it's not really for. I look at it as an ecosystem growing effort.

Speaker B: So I want to step up a level because we're getting towards the end of our time and just to help us wrap it up a bit. So one thing that's kind of stood out to me a bit from this conversation is we've obviously focused quite a lot on the control plane itself, which is one of those things I'm not sure that I've really thought about that much in terms of the potential trust issues or risk that exist there. I'm wondering if you think that that's a layer that has been underappreciated in risk terms maybe up till now potentially.

Speaker C: I think the important thing here is just kind of our definition of a control plane version versus others. And what I mean by that is in some companies like netlify, ecs, you are giving all of your trust to that company to maintain the control plane, the worker nodes, like all of the kind of things that make up that platform. Right. And then on the other side you have companies that are deploying kubernetes from scratch and they're owning everything. There's not many companies in the ecosystem that are similar to Cycle where it's, we are maintaining the control plane for our customers, meaning that our customers only have to worry about the worker nodes and like their storage. And so in uh, our case, I think it's probably a little bit unique. At the same time we also appreciate that companies want to know like what are the risks of using a, you know, control plane that might be distributed or you know, geographically in an area that has different laws. So there's concerns and you know, I mean, I guess yes, there are concerns and risks, but by us being able to deploy control planes in different geographic regions, my hope is that long term that that mitigates uh, many of those issues that uh, could be foreseen there. And again, obviously, you know, the, you know, wrapping this up with a bow is going Back to the point that our customers own all of their data, all of their infrastructure, et cetera. So the biggest risk of a control plane going down, uh, in cycles specific circumstance is like your infrastructure still up and running, your containers are still up and running, your VMs are still up and running like nothing's going to stop because again traffic is routed into your service there, it's never routed through, through the control plane. Uh, it just means that like absolute worst case scenario is that you, we'll use cycles API to perform actions against your infrastructure. So in terms of that being the absolute worst case, I would say that's a pretty good worst case situation to have because at the same time your servers will still continue to maintain state. So if our control plane were, again I'm talking about worst case. But if our control plane were to go down, your servers will continue to maintain a manifest. So if one of your containers crashes, but that manifest says, hey, like this container should be running, your server will still restart that container even though it can't talk back to cycle. So if we think of the control plane like uh, as a state engine, but where it's like based off an eventual consistency model, further de risks a lot of the kind of situation that you might have there.

Speaker B: Right, Nice. And um, then maybe to finish it up, how do you see this sort of broader picture in terms of where our industry might be going? Do you think we're heading towards a kind of more fragmented Internet, more fragmented infrastructure that's being increasingly shaped by geopolitics? Or do you think that this kind of tension we have between mainly Europe and the US but also the western China, do you see that eventually kind of resolving.

Speaker C: My opinion is that we will continue to see, I don't, I don't know if I want to use the word fragmentation. I think that we will see more regions kind of stepping up to the plate and uh, more regions stepping up to the plate and like building their own quote unquote AWS or their own gcp. And I know that the EU is putting a bunch of funding right now towards, you know, building the European version of aws. Right. And there's a number of providers that are, that are, you know, part of that mission to, to try to, you know, to get there. So I mean I, yeah, I don't think I'd say fragmented because I, at least I'm hoping, I'm hoping that we don't have multiple Internets where you know, you need to specifically, you know, when you, you know, connect, you need to talk about which Internet you're going to connect to. Uh, I think that'd be a terrible situation. I mean, I guess in theory it's possible someday that's where we end up. I'm hopeful that that is not the case, but I do think that we will see more and more, you know, conversations on sovereignty, privacy. Like, I mean, Even in the U.S. like, I know that, you know, this conversation might initially come across as anti us, but I'm not implying that at all, because, you know, as a US Citizen, as a number of people that I know back in the U.S. there's a lot of privacy concerns even in the U.S. right now. And so you have a lot of people starting to think about encryption and, you know, how do we do some of these things differently that, you know, is protecting workloads and communication and things like that more. So my hope is that technology kind of keeps outpacing some of the bad actors. But I do think that, at least in the short term, we will see more organizations focus on how do we have this within our boundaries. And I think, regardless of any geopolitical things, I think it's probably a good effort. Uh, too many people kind of handed over the keys to AWS and said, great, solve it all. And now AWS is sitting there holding your keys, saying, now you need to pay me more if you want these keys back. And I think that there's a lot of conversations that are really starting around, and so I'm excited to see kind of what happens over the next. Over the next few years.

Speaker B: That's really exciting. Jake, thank you so much. It's been wonderful to talk to you, and thank you so much for joining me for this episode of the State of the Art for Goto Unscripted.

Speaker C: Excellent. Thanks so much, Charles.

Speaker A: Thanks for listening to this episode of the Goto podcast. Head over to Gotopia Tech to discover live lots more content from the brightest minds in software development.