 At the end of the day, businesses are looking for an abstracted way to handle this cloud and infrastructure interaction. They want to do that with as little drama as possible. They need a reliable way to say, I need to build this cluster or configure this system. That's what these products do. Hi, this is your host, Bill Bhaltia, and welcome to you for our Let's Talk. Today we have with us once again Rob Hershville, CON co-founder of Rack and Rob. It's great to have you on the show again. Rob, it's a pleasure. This Terraform change and the TOFU project now, I think, is the source of endless possibilities for discussion and significant and important ones too. I remember our discussion when Hashi Corp made an announcement and then I was at Open Source Summit, so I talked to Sebastian when Open TOFU was announced there. Have you seen that interview with Sebastian? I did. It was an excellent job. You answered a lot of great questions and I think you raised some questions also. Right. Once you saw the interview, you had a lot of questions and we want to kind of tackle some of those. But before I jump into the main topic, what role do you see Open TOFU is going to play in the whole Terraform infrastructure code space where do you feel that it is filling some of the gap niches that was created because of business source license? I think that from a business user perspective, I think Open TOFU really opens up the market and the portability of infrastructure as code at that layer and the cloud and API interface lightweight orchestration. I really think that as a business user, as a general market, that we are well served by having guaranteed open governance for something that has been so widely embedded. I think that portability really benefits. It means that companies who are using Open TOFU as a base and are used to using Terraform had been using Terraform. If they switched Open TOFU, have a confidence in vendor portability, vendor neutrality when they embed that platform into their systems. I think if you look at the current status quo of the industry and you look at how people are using Terraform today, I think that that provides actually a lot of commercial benefit and assurance to companies who are involved in the market using those products. What does it mean for Racken? Oh, boy, for Racken, I don't know. We have abstractions that take and run Terraform on customers' behalf. We have abstractions that allow people to use Terraform interface to our product and digital rebar. Without a doubt, the changes that House Sheet Court made to the license would prohibit our ability to use Terraform without a dispensation from them or something else. Open TOFU allows us to continue to use cloud interfacing in that model. However, our designs didn't assume that Terraform was the only way people would do that interfacing. We already anticipated that customers wanted neutrality from an API perspective. And so the expectation here, I think this is what customers, what I've heard when I've talked to our customer base, they are expecting to use both. So they'll use Terraform where they're continuing to use Terraform. The business license change doesn't actually impact people doing that. If they were using alternatives to Terraform Cloud, then this allows those alternatives to continue to operate. And so they're happy about that. It removes that complication. But it doesn't actually change their choice that much. If they're using one of the alternatives, then they'll be using one of the alternatives. If they're using Terraform and House Sheet Court-based products, they'll continue to use those products. There wasn't that much need, if you will, for customers to reuse as much of the binaries, at least. And that's one of the items from that interview was a lot of talk about binary compatibility. And me listening to that, thinking about how our customers consume that technology, you're going to use a stack. The need for binary compatibility isn't necessarily important as long as you could take the Terraform plans or the state files. Even the state files, I think you could translate state files, so they don't even have to be compatible. So the need for compatibility here is, I think, much lower in practice than what the market is right now worried about. And I think if we can say, we don't actually have to worry as much about compatibility as we are, then the open tofu community can follow an innovation path based on what their needs are. Right. I mean, as in the discussion, he clearly mentioned that they will try not try to drop in replacement just the way MySQL or MariaDB became over time. But since legally they cannot look at the code and the project will evolve over time, you know. And at one time, maybe Hashi Garfield will come back to it and actually even join open tofu as their own upstream. But since you mentioned binary compatibility, I do want to quickly talk about just going deeper. They're really, as you said, businesses really won't care who are in Terraform. But as we use REL versus CentOS, you know, it's both used in the same organizations use SUSE or Linux also. We do multi-cloud. That's why it was a hybrid platform that we used to talk about. Now we talk about hybrid cloud. So can you talk about it from your perspective, because you do deal with clients directly? We do. And this is the thing, you know, HashiCorp, my expectation for them is that they are going to continue to innovate and add into the platforms that they have. They actually need to build a suite of products. You know, I've talked about this in the past. And they're going to add into Terraform and their other products in ways that aren't going to be compatible and actually might not be well served for the other users of open tofu. And the trying to keep them the same like we do with Linux distributions, I think is actually going to be a challenge, right? For Linux, we say the core is the same, but the distributions actually have different packaging. And where you download packages from varies depending on the distribution you have. And maintaining those package repos is actually the work of the distribution. And so, you know, the important thing to users here is that they have a maintained innovative piece of software. That's the value that a customer wants to do. Nobody's compiling open tofu or Terraform. They're getting it from a distro. That distro is going to have add-on pieces that create license and commercial value that they expect people to pay for. And that's how people want to consume it. The challenge with Terraform and people forget this is that Terraform itself was free. People were charging for the management layers for it. And that was the design. That's actually pretty typical for open core type software here. And for the companies that are going to build on top of open tofu, they're going to have to differentiate their management platform, right? Because open tofu is not differentiated. It can't be. They're going to have to differentiate the management platform. The question becomes, do they add things into open tofu? Just like HashiCorp is now doing with Terraform, where they're adding things into Terraform that differentiate their management platform, you know, how will the community do that with open tofu? Because that's ultimately what the end users care about. Do I have a management platform that governs my use of Terraform, incorporates it into more things, provides an API, gives me management control and maintenance and auditability, right? And governance, oh, those are the things that people are going to pay for. And that's where the competition is and should be. What kind of ecosystem you see will be created around open tofu as I was talking to him. The response came from, once again, the user. Actually, some big users, they were not comfortable. That kind of also led to open tofu. Because do you also see it's emergence as a distro? A very good example is Ubuntu. There are so many derivatives based on Ubuntu. I mean, Kubernetes is a great example. OpenStack is a great example. What kind of ecosystem do you see of vendors who will be offering distros? And now they might add all those differentiators on top of open tofu. Oh, my goodness. There are so many challenges in front of the open tofu community. I wish them a lot of luck here in a very sincere way. Because the challenges is that the providers and the way you plug things into Terraform fundamentally create distributions. And so if you were adding something into open tofu and it breaks compatibility with the providers, then you're going to have to have your own providers. You're going to have to compile your own providers. And let me, this is a little bit confusing, but I want people to understand how open communities and distributions work. If you are a customer of open tofu manager X, I'll just call, I don't want to use X anymore for this, open tofu manager, Kai. And Kai gives you a distribution of open tofu, which they would do, and then attaches you to the provider libraries that open tofu is maintaining. All that's great. But when they find a bug in open tofu or in any of the providers that you're depending on, you now have an expectation just like you had with HashiCorp that they will fix that bug and then move that through the distribution channels. But because it's an open community project, you're going to have to upstream that fix or have an alternate distribution model to build your own provider, build your own version of open tofu, and then distribute that to the customer. That is the challenge with open source is what Red Hat has done with Linux and the other distros too that makes it so valuable. And so the thing that you're getting in this case that we're still going to have to work out how this goes is how well can the community take fixes to these providers and then put them back into the system? And it could be, and this is the way the provider model was supposed to work, that the end, like the Amazon provider is maintained, it's not currently maintained exclusively by Amazon, but you would hope Amazon would then fix the bug. But, and there's a huge community benefit if manager Kai then turns around, fixes a bug, and then everybody gets the benefit. That's the beauty of open source and it's a huge deal. But it takes time. And if somebody needs that bug fixed quickly, there has to be a process to get it through the system. And that's, and then, if it's not reviewed or not accepted, there's all sorts of wrinkles in how these things work. And so at the end of the day, that customer relationship with, between Kai and the customer ends up, they need to own that whole stack of technology. It's a huge benefit that they can now own that huge stack of technology. In the past, they were beholden to HashiCorp to do that work and kudos to HashiCorp for it being high enough quality that that whole system was working without having a way for communities irrelative fixed bugs and funnel things up and then take control of their own distribution. So at the end of the day, this should create a better customer experience. And it also is gonna create more complexity within how those things get managed. Exactly like you were describing, all of these derivatives of the popular distros, right? There's one of the reasons there's a ton of near copies of distros is because it gives the vendor who's using them that distro the freedom to add, inject, change, fix, control their own destiny. And that ultimately in open source is the key power, but the distros are part of a successful open source project. The good thing is that since OpenTov is Paralynics Foundation and Linux Foundation, it's a foundation of foundation. They have Kubernetes, they do have expertise in how to manage Kubernetes is a good example. They are Kubernetes distros, but they're all, it doesn't really matter which Kubernetes you're using. It's not the same story as Suza versus Red Hat versus Canonical Linux. So do you also think that that experience will also benefit the community and also the community will also overlap because it's the same cloud native kind of community here. It's interesting with all of the providers, and I don't know any of the back room machinations that landed Terraform in the CNCF specifically. I think this is more adjacent to CNCF than in CNCF, and I haven't heard the arguments for one way or another. I think it's a convenient and smart move, but showing up, but I could see a whole parallel conference to KubeCon for Terraform alone. Because all the providers, all the consumers on it, it's its own provisioning engine. In some ways comparable to, certainly not comparable to Kubernetes in a lot of ways, but in some cases it is. It's a DSL, a domain specific language for cloud provisioning, and that's not the same thing as what Kubernetes does. So I think that there's, we're not at the end of this road from a governance perspective on these projects. The good news is, I think most customers at the end of the day can shrug their shoulders and say we'll use Terraform as long as we can. We'll, the alternate managers will have a mechanism to at least in the short term, not have their businesses disrupted, which was their goal with this. And at the end of the day, I think, for those businesses to unlock value, then they're gonna have to find ways to differentiate and add into Open Tofu in ways that work within the community framework. And they'll be good players and they'll be bad players, like there always are in communities. And I also think that HashiCorp is not in anywhere done or seeding the ground. They actually know the product in space really, really well. And I expect to see innovation coming out of Terraform that will likely impress, based on their track record, will impress people in how it works and what it can do. And that will give businesses an opportunity to evaluate, do they wanna go and license the HashiCorp versions and take advantage of what they're adding into the product. And this shouldn't be a shock to anybody. We're having an open source conversation, but at the end of the day, customers will pay money for products. Right, the opportunity, if you're building something internally that mirrors some of these products, which I see a lot of customers doing, Open Tofu might make you more feel safer building your own stuff on top of this. We're actually in a place where there's gonna be a lot of managers. And if you're a technical leader and your teams are building their own managers, at this point, I would start questioning that decision. There's just too many on market. Let's look at it purely from customers, users, businesses perspective. Just take the technology out of it. If you look at, you know, once again, you talk about Terraform or you talk about Open Tofu. What is the main business problem they're going to solve? And does that really matter whether it's Open Tofu or it's Terraform? I think that cuts to the heart of the matter here. At the end of the day, businesses are looking for an abstracted way to handle this cloud and infrastructure interaction. And they wanna do that with this little drama as possible. They need a reliable way to say, I need to build this cluster or configure this system. That's what these products do. And they do it as part of a series of operations in a CI CD pipeline or an infrastructure pipeline as part of, right, other tools. And so at the end of the day, what customers need is a reliable flow drama, predictable way to manage their infrastructure, right? And really it matters to us because we're down in the industry and the weeds, but from a customer's perspective, they wanna spend less time worrying about Tofu or Terraform and more time setting up, tearing down and using infrastructure. And the extent to which companies are helping people do that, that is actually the value proposition here. And the only thing we're trying to do is reduce the friction of a potential license confusion. It really doesn't matter if it's open or not open, if you're solving the customer's problem without adding friction. And that is at the end of the day, the way to understand this problem and not become so embroiled in the, this license, that license, this company, that company is your infrastructure, are you running your infrastructure better today than you were yesterday? And do you have a plan to do it better next week? And if that's true, a lot of what this is really doesn't matter. Rob, once again, thank you so much for taking time out today and talk about Open Tofu and Terraform. And as usual, I would love to chat with you again soon. Thank you. Thank you. Thank you, Swap.