#84: In these times, everyone is familiar with Slack and Microsoft Teams. However, there are other companies that offer similar solutions. Today, we talk with PJ Hagerty from Mattermost and find out how Mattermost sets itself apart from its competitors.
If you like our podcast, please consider rating and reviewing our show! Click here, scroll to the bottom, tap to rate with five stars, and select “Write a Review.” Then be sure to let us know what you liked most about the episode!
Also, if you haven’t done so already, subscribe to the podcast. We're adding a bunch of bonus episodes to the feed and, if you’re not subscribed, there’s a good chance you’ll miss out. Subscribe now!
PJ is the Senior Developer Advocate at Mattermost. He is an organizer of DevOps Days Buffalo, CodeDaze, and ElixirDaze, and a board member of Open Sourcing Mental Illness (OSMIhelp.org). PJ is a developer, writer, speaker, musician, and celebrant of all things community. He is known to travel the world speaking about programming and the way people think and interact. He is also known for wearing hats.
Viktor Farcic is a Principal DevOps Architect at Codefresh, a member of the Google Developer Experts and Docker Captains groups, and published author.
His big passions are DevOps, Containers, Kubernetes, Microservices, Continuous Integration, Delivery and Deployment (CI/CD) and Test-Driven Development (TDD).
He often speaks at community gatherings and conferences (latest can be found here).
His random thoughts and tutorials can be found in his blog TechnologyConversations.com.
I go to a lot of these events and you see these people who are flying in jets and driving in cars and coming up and saying, well, open source is the only way to go. You shouldn't make people pay for software. How did you get here if no one was paying for software?
This is DevOps Paradox episode number 84. Mattermost Saves a 30 Year Old D&D Campaign
Welcome to DevOps Paradox. This is a podcast about random stuff in which we, Darin and Viktor, pretend we know what we're talking about. Most of the time, we mask our ignorance by putting the word DevOps everywhere we can, and mix it with random buzzwords like Kubernetes, serverless, CI/CD, team productivity, islands of happiness, and other fancy expressions that make it sound like we know what we're doing. Occasionally, we invite guests who do know something, but we do not do that often, since they might make us look incompetent. The truth is out there, and there is no way we are going to find it. PS: it's Darin reading this text and feeling embarrassed that Viktor made me do it. Here are your hosts, Darin Pope and Viktor Farcic.
We all love Slack or Microsoft Teams, or even dare I say it AOL. It doesn't exist anymore or Yahoo. Instant messaging has been around for years. What was your first instant messaging client, Viktor? Do you remember?
I have no idea to be honest. Probably Yahoo something
Probably Yahoo. Okay. And most people are, or Microsoft Messenger. Again, things that no longer even exist, but it was a way to get started. On today's show, we've got PJ with us. PJ is with Mattermost and if you've never heard of Mattermost, well, you've probably never worked in a large organization that only has on-premise solutions. PJ, thanks for joining us
Thanks for having me.
So let's talk about Mattermost for just a minute and understand what it is and what it does. And then we're going to not talk about Mattermost for a little bit and talk about open source versus commercial.
Absolutely. Mattermost I mean, you kind of hit the nail on the head there. It's very similar to Slack. It's very similar to Discord and other messaging services. The idea is to provide communication for teams, to put people together in a way that they can communicate either asynchronously because they're remote and around the world, or to have a nice tight team trackable situation, so that you can say something's happening. Here, I've commented or I've dropped an emoji. I know that that's happening. I'm acknowledging the existence of the situation is something that's very familiar to a lot of people, I think.
But the difference in comparison to Slack is Mattermost has primarily been only an on-prem play.
That's right. It grew similar to Slack in that it grew out of the video game industry, but what we found was there's an entire group of technology companies or non technology companies that require heavy security. We're talking banks, healthcare, government agencies where they can't just spin up a Slack instance in the cloud and say, all right, great, we're just going to have our communication out and open. We need something behind a firewall. So Mattermost built an open source tool that allows you to create a very Slack similar situation to communicate in your own terms. You hold the server behind your firewall. You can have it air gapped. We have a lot of air gapped customers. So that basically you have your own instance and you're running the software, without having to worry about exposure to the outside world.
So right now, You're probably the only player? Up until what a year or two ago, your competition probably was HipChat.
You could run HipChat on-prem and then once Atlassian sold that off, is that story correct? Atlassian sold off HipChat to Slack.
Right and then it just kind of disappeared. Yeah.
So basically once they exited the market, you were the only player.
Right, right, which is an interesting place to be mainly because I think a lot of folks think, oh, this is chat, it's open and if you look at a lot of the applications for chat things, it's to deal with communities or deal with external users or external customers. The understanding is that if we can put this on-prem, then we have that edge. I don't think we specifically went out there saying we're going to be the number one or only available on-prem solution. But what we were able to say is like, hey, we can set this up in a way that's convenient and it's got a little bit of a difference from what's available for cloud or SaaS solutions.
Did Mattermost start out as open source and they commercialized it? What was the history behind that?
Yeah, it's always been an open source product. It was originally built similar to Slack. It was built as a communication tool for a game. The game didn't work out. The communication tool did. It was decided from the very beginning that this needed to be open source in order to allow people to set up a server, run what they wanted on it, and then there's a extremely large plugin and an integration community to, oh, we need LDAP or we need this extra security feature or we're DevOps focused, so we need an integration with Grafana so we can see our graphs real quickly. The open source attitude was there from the very beginning.
How is it licensed?
That's a great question. I do not know the answer because I just had a conversation. I believe it was a GPL license. I'm sorry, an MIT license initially, and then moved to GPL. I could be wrong on that though.
well, let's, let's take a look at it. For those of you listening, obviously you can't look over my shoulder. It is Mattermost licensing, so it is a
so we have a proprietary license. Yeah.
Yeah. It's MIT with AGPL and a commercial. Okay. It's already confusing and APL. Okay. I'm not a lawyer. I don't play one on TV. I can't figure this one out.
I too, am not a licensing expert at all.
Yeah. So it was produced by the company, open sourced and you also brought up, it has a very large plugin ecosystem.
It does. Yeah. One of the first things that we do a lot of times when we bring someone into the company is we say, hey, build a plugin for something that's interesting right now. When I came in, they were like build a plugin for Grafana so that we can easily integrate Grafana dashboards, with a simple command. Part of the reason why they do that is because A, there's a rich ecosystem already there, but B, the variety of things, the flexibility of how Mattermost works, it becomes really obvious when you're building a plugin. The main bits of Mattermost software are built in Go with a React front end. So it becomes really malleable when you take a look and see, I can jump in, I can write a plugin in a completely different language. One of those languages doesn't really matter and get it easily integrated quickly. We're talking a matter of a couple hours after building the thing. That's kind of what they wanted to focus on. The idea that, and this is a philosophy, this is part of the reason why I joined, that keeping the core product built the way that it should be built. The community drives that building and allowing them to build anything they want onto that kind of Lego brick technique makes it easier to manage and easier to figure out what the community really wants and how to make a better product, as opposed to saying oh, we heard through a focus group that people really want these plugins, so we gave them to you and you can buy them through the marketplace. It's like, no, do you find this interesting? Cool. Build it. Proof of concept it. Our team will help you flesh it out and we'll go from there. That's kind of the attitude that Mattermost has with the open source community. We're willing to kind of embrace whatever stretch you're trying to put through Mattermost.
So I guess then most of what open source contributions outside of those made by Mattermost people are mostly focused on plugins, right?
I would say yeah. Plug-ins and integration. You can also create special slash commands and things like that and add that to the ecosystem if you want. There's going to be a cloud version that comes out that will be a little bit different and their plugins and integrations will be a little bit more controlled because if they're being built for a different security situation, we have to look at different options and things like that. But for the most part, our focus is on allowing people to bolt on what they want to bolt on and to make it as easy as possible.
So, do you have a plugin that you saw somebody made and it made you think, heck, I never thought anybody would come up with this idea. I mean, something beyond the usual things, right?
For the most part, I think people who use Mattermost are a little more seriously minded. I haven't seen anything come across yet that would come across as ridiculous or like, why would you build it, why do you need this? For the most part, the people who use Mattermost and are part of the community of builders for it, they mainly stay focused on things you'd expect them to. DevOps tools, end user tools, things specific to the sector that they're working in, whether that be financial or governmental or whatever. They keep their focus on the more serious side. We've seen a few things like an integration that brings in Reddit that just shows the top 10 Reddit posts of the day. It's like, okay, I don't really see the value in that. You're basically pointing people to stop doing work, but if that's the integration you want, maybe there's a reason for it and I just don't know what it is.
This is out of pure curiosity and you probably don't know, but I'm going to ask anyway. How much of contributions come from company sponsored work and how much is just individuals?
I'm not sure of the exact statistics, but I would say like a rough estimate, I would say maybe 20% come from sponsors or corporate interests and 80% come from the communities that are using the free version.
I ask that question quite often, because I'm curious whether enterprises are going to jump into that train and start saying, you know, this matters to us.
They do to a degree, but I think the community drives things more. If you look at it from the perspective that I've been doing DevRel for a long time and I've always had the attitude that developers drive what gets done, but someone else at some higher level decides what gets purchased or what gets adopted. When you look at a community version of Mattermost, you look at something that's purely for developers and so they're really going to look at it. They're really going to find the balance. They're really going to want to build a thing that's better for themselves as opposed to building something that's better for their bosses, if that makes sense.
One thing I did see when I was preparing for this conversation today was, hey, let me go download it. And I have to admit, one of the things that sort of took me aback was I had to give my email address or name in order to download anything, even the open source side. And it makes sense because basically, even though the open source side or the free side that I can install, it's the full version and basically all it needs is a license to unlock the enterprise features. That's sort of a weird smell to me. Compared to most other open source projects that have commercial parts to them, or that also have a commercial flavor.
Sure and I think that it's because it is in fact a product. Even if you're not paying for it, it is still a product. As opposed to say, I think of things like CrateDB, which is a database infrastructure. We don't need your name for that because you're not really getting anything from us. You're getting a set of instructions. With Mattermost, you're actually getting something that's being put up on a server that has the potential to do something. There's some accountability that needs to happen. If you take Mattermost and do something nefarious with it, we need to make sure we know who installed that Mattermost instance. So it's kind of a security measure. That's my understanding of it.
Yeah. It makes sense and the way you phrased it, even though it's not so much open source, although it is open source, it's a product that is free and I'm using the word product on purpose. It's a product that is free that you can install and you can run completely free. The only thing you ever have to give up is an email address.
Actually, I'm checking it right now. It's optional.
Oh, it is optional.
Yeah. So there is that field. Give the email address, but it's optional. Now I'm ruining the business. You can click the button without filling in anything.
I don't think it's that big. Chances are, if you're going to use it and you really want to upgrade, there are enough features in there to just click upgrade and we'll get the email address eventually, if that's what you want to do.
It was a different download experience than I was expecting. So, again, it's not huge, but, and obviously I went down the rabbit hole and Viktor pulled me out of, I actually didn't check it. Um, I thought about it, but then it was time to go eat a snack, so I didn't followup on it.
I understand that completely. I've been down that road myself.
What is the strategy of open source versus commercial? How do you make the decisions? What do we put in the product that is going to be on the completely free side and then what is the stuff that we need to be paid for?
I think it depends on the universality of the feature. So what I mean by that is if someone comes and says, listen, we need stricter guidelines on who can do what, depending on their login. Okay, well, that's something that probably everyone needs. That's something we should build as a feature into the core body of Mattermost. When someone comes and says, well, you know, we specifically need the ability to do this on mobile devices that are air gapped. Well, that's probably not something that everyone needs. That's, I'm going to guess, probably an enterprise specific need and that's something that we can build to make an option just for the enterprise, just for paying customers. Chances are, if you're running a Ruby community in Moscow that you want a Mattermost instance for, you do not need to worry about air gapped, mobile devices,
I always liked that approach to open source and enterprise, where actually you're distinguished by size and then by the business really using it, right? In a way, not necessarily explicitly, but hey you're individual, you're group of people. You earn less than a hundred million dollars a year, kind of, open source.
That's kind of the idea. It's kind of a behavior driven development model. Is it a matter of someone saw a focus group that said cloud is the way to go and we must deliver a cloud product. Probably not. Is it a matter of people saying, we need to develop this cloud product because some people need something specific that is extremely similar to Slack, but not Slack, more secure, but can be cloud accessed instead of being on-prem. We can fix that. We can make that product and we can make it both free and enterprise with slightly different determinations in the version. It's not a matter of saying we're going to build all the features and then take them away from the open source version. It's about who can use what, and what's important to what parties.
It's curious that the vast majority of people I speak with tend to go to one extreme or another. It needs to be a hundred percent open source without realizing really that every open source project needs some funding, right? One way or another. So every open source project needs to sell something in order to continue thriving as open source.
I've always thought that attitude was interesting, especially when you talk to developers. I'm in developer relations. I fly around a lot. I do a lot of different things. I go to a lot of these events and you see these people who are flying in jets and driving in cars and coming up and saying, well, open source is the only way to go. You shouldn't make people pay for software. How did you get here if no one was paying for software? This isn't a capitalist rant. I'm just saying you can't have one without the other. You can't live this lavish, I live in the Bay area and pay $3,000 a month for a 12 square foot apartment, while I drive my Tesla and also say all software should be free. You can't have both sides of that coin.
Especially maybe for some tiny projects done by one person but as soon as it grows. Then you have that sometimes it isn't obvious how that funding comes along, like in your case and sometimes it's less obvious, but Kubernetes is heavily funded. Even though it might not seem so obvious because there is no official Kubernetes enteprise. Right. But heck it gets millions. It's a positive thing. I encourage people to contribute to projects sporadically. You have two hours today. You have three hours next week, but you also need the dedicated group of people who are making really sure it's my job to make sure that this project is working.
Right and I think the great thing about open source is it allows people to walk through the door. There are a lot of very easy entry points through open source to get into the world of technology, get into the world of development. I have a four year CS degree. I know a lot of people who do. My CS degree was worth about bupkis as soon as I actually got a job. But there are lots of ways to enter. People have asked me like, I wanted to do Hacktoberfest but I don't know anything. What have you heard that you like, Oh, you know, I've heard about Python. Okay. Well, Python is an open source language. Go in the documentation. Look for some grammar that needs to be checked. Check it out. Learn how to fork a repo. Learn how to do everything on GitHub or GitLab or wherever. Fork it. Make the grammar check. Submit a PR. Congratulations. You're an open source developer. It's just that easy. Start with documentation. As you're reading the documentation, you're going to learn more about the language or the product or the software. You'll build things up and that's why I love open source. Literally anybody, caveat with a computer and a decent internet connection, but just about anyone can get involved in open source if they want to and your level of participation is completely up to you.
So at the time we are recording this, Hacktoberfest is going on. I'm curious, which side of the divide you are? You probably heard, right? Some people are disappointed that people are making false PRs.
I feel like this is where we put my Facebook status is it's complicated. I don't have a Facebook, but if I did. The issue I think is, what is a false PR? Did the change that you make add any value to the code or the documentation or whatever it is that you forked and changed? If it did, even if it's I felt like there should be an Oxford comma in this list of things that you had in the middle of a sentence, you've added value. Okay. I can argue that's a decent PR. Get your t-shirt from DigitalOcean and have a great day. If you literally just went in and entered some white space to push, that's not valid. You're just doing that. To be honest, I was surprised at the amount of PR and attention that this was getting, because this isn't a new phenomenon. This has been going on since Hacktoberfest started. People do Hacktoberfest literally to get free t-shirts. We can talk about the brotherhood of humanity and getting together to make everything great for open source and that's true and it's great. It's fun. People want a free t-shirt. When people visit your booth at a conference, they're there for the free socks you're giving away. They're mildly interested in your product in the same way that people in Hacktoberfest are mildly interested in your project. I'm not saying there aren't people who are extremely zealous about what you're doing, but when people are like there's a bunch of fake PRs. Define a fake PR. If you want to be strict about it, that's fine. But I mean, you did this thing to be social and to promote an idea. Not everyone agrees with your idea, so you're going to get different reactions and I think that's what we're seeing in Hacktoberfest this year.
But to me, that's similar like, I feel that some people have as a relation to open source in general, is that like, if you go back to the previous conversation, when you said flies in jets and says it 100% needs to be everything open source. And yet most people really don't want open source because it's free not because of for being open source. Right. And to me that's similar, like t-shirt. I want to contribute because I get the t-shirt, not because I really believe in open source.
What you're speaking to here too, is the whole idea behind Mattermost. So we understand that we have a product that is open source, that you can use completely open source if you wanted to. There's a lot of people that stay dedicated to that idea, but also you can pay for it and get a good experience and there's nothing wrong with that too. We don't feel like we've compromised our moral high ground by selling a product. I think that that idea is a little antiquated that open source for the sake of open source is great in concept, but it's not really great in execution. Part of it is because we do live in a capitalist society. You do have to make money and it would be great if you could give things away for free and spend 40 hours a week building a product or a piece of software or some code or a language that you could then give away for free and not have to worry about having to pay your bills. I like to make this joke a lot when I do talks. You know why this year isn't the year of the Linux desktop? Because no one can afford to make one. And it doesn't mean that they don't have the time. Everyone has the time. What you don't have is the ability to make something for free constantly without any income whatsoever. And it's unfair to ask open source developers to work a 40 hour a week job, and then work another 40 hour a week job building the next great piece of open source software. When people argue, well, I don't use open source because it moves at such a slow pace. Well, yeah, because this is a lot of people volunteering their time. To expect them to work at a regular job level is obscene. Now I think there are some positives coming out. There are companies like Tidelift and foundations like Ruby Together that are working to collect funds to pay open source contributors. But that kind of underscores the idea that you can't be a truly pure 100% free open source contributor and be successful. It won't work only because we live in a society where that's not going to be possible if you want to survive.
I definitely do.
Yeah. I mean, I'm fond of eating dinners and being able to clothe my children. If it was just an idealistic everything's perfect utopia, I would do nothing but code open source all the time, but I do have to put food on the table and that's the reality of the situation.
Why don't enterprises seem to believe that story?
Well, I mean, it depends on the enterprise. And I think the irony here is when you look at the huge enterprises, they believe that things can only be valued if you put a price tag on them. We look at some of the large older technology firms, places like Oracle and IBM, where they talk about the only way we can value even our open source contributions is if we say there has to be some sort of price associated with it. Like, yes, Oracle delivers MySQL and you don't pay for MySQL, but what you build pays for MySQL. Oracle makes millions of dollars a year. IBM goes out and buys Red Hat and says, how do we profit from the great open source community? The great open source tools that Red Hat has built and how do we make our tools that we've been asking enterprises to pay for it for so long, like them, but for money. And I think it's just an old business idea. It starts at the top. It's not driven by developers, which is where the open source movement comes from. It's driven by folks who are looking at focus groups and charts and saying, we only use Microsoft Teams because Microsoft always makes the best software. That's why we use Office 365 and not Google Drive because Microsoft has always been the one that they do it for business. There's a price tag associated with that. And yeah, there's a price tag associated with Google Drive, but it's a lot smaller, so it must be less valuable in the eyes of the business-y people.
Yeah. I mean, that's similar to Apple in a way. I mean, Apple creates great products. Right. But there are some not so great which are selling great simply because they are very expensive. There is that association between price and quality.
That's kind of how they magically came back into the business. I remember when I was in elementary school, it doesn't matter how old I am, but when I was in elementary school, we had Apple IIGS that were replacing the Texas Instrument computers that plugged into the back of a TV with the Atari controller switch. We were like, Whoa, we must be at a good school. Look at the computers we've got, Apple IIGS. They went into decline for a bit when Steve Jobs left. They came back with that weird bowling ball looking iMac thing and people were like, wow, that's fancy. It must be better because it's fancy. It kind of faded and came back with the MacBook Air and things like that. It became the default development tool because it's a *nix based system. It comes preloaded with half of your open source, popular, open source languages, ready to go. If not in the most recent edition, but ready to go. It was like being able to use a Linux system, but prettier and more versatile. Plug and play. Everything works. But that was 15 years ago. How much innovation have we seen out of Apple in the last 5, 10 years. They're like, Oh, it's faster and has better chip set. Okay, cool. Um, but so does the Windows Surface Pro. Why are you convincing me to stay? Like, what's keeping me with Apple other than familiarity and convenience? Where are the innovations? And I think that that's also where enterprise software gets in trouble. They're like, Hey, we built this really great thing back in 1992 and we're still using that. We just made more of them and we called it cloud. Okay, cool. What have you got for me now? Whereas the open source world for the most part is consistently pushing things forward, because they want to see new technologies. No one believed that you could build a computer that is usable with a Linux operating system and sell it for 30 bucks but look at the Raspberry Pi.
At least in my case, I'm always torn between the two in a way. At least once a week, I play with a new project. I try to explore thing s. On the other hand, like my laptop is, for years now, Mac simply because I know what it is. I know what it is. I don't know whether it's better or worse. That part of my brain has no time to spend on checking Linux laptops or Windows laptops, right? I'm not saying that they're better or worse, but I have no time for that, but on the other hand, actually have more time than I should to explore some other things.
It's like time to development. Let's say you get a new job, you get a new position and how much time do you want to spend shaving yaks, making sure you have your IDEs installed and any plugins that you needed to make that work are installed. Your browser is set up the way that you want it. Your security protocols interact. Does the new jobs still have the same is everyone using LastPass and you have to integrate that? Well, some browsers don't use LastPass very well, so now you have to change what you're doing. Is it just easier to send somebody a MacBook and say, hey, this stuff's pre-installed, this is what we do. We're ready to go. Install your time tracker, install your browser, put in your ID and have a good day. You know exactly what you're getting as soon as you open that box. I don't hate Linux. I don't want anyone who listens to this podcast to be like, that guy really hates Linux. I love Linux. I think it's great solution for a lot of things. I have a great Mint laptop that I use outside of work for various testing purposes. But at the same time, if it was my day to day, I think it would be problematic. Little things like you can't do Zoom calls. Jitsi doesn't work well on Linux because the cameras don't want hookup and you have to install drivers. Why do I want to be installing drivers? I want to be doing my job.
So Mattermost. Completely free version. Download it, use it. If you're working in a company, I still see this from time to time. If you're working in a company and there is zero, zero real-time communication tools, Slack, Microsoft Teams, whatever else is left, look at Mattermost. Because it's on-prem. The argument of, well, we can't use cloud goes away with Mattermost. Now there is an enterprise version and also you were talking about, there is a new beta coming.
Right, relatively soon, we're going to see the cloud version come out. It's going to be a very similar experience to Slack where it's like, Oh, I need to set up a chat server. I need to do it now. Let me click, click, click. There it is. And you get a URL that works. Even with the on-prem version, we already have the mobile capabilities. So I think the mobile capabilities will kind of give you an idea of what the cloud capabilities will be. It's very similar to that. Like you said, Darin, the key right now, even with the cloud coming and the excitement that we have around the cloud products, I think a lot of people enjoy the fact that you can just pop up a server, drop Mattermost onto an instance, or a VM or a Kubernetes cluster and just have it run. The fun story that I have for how I got involved with Mattermost. First time I used it is because I am, as we can tell from this conversation, a really super cool guy, and this pandemic hit and my D & D group needed some. We had some questions about how we were going to keep going. We've been playing together for almost 30 years and we weren't about to say, well, we just won't get together on Thursday nights. Obviously we needed a solution. One of the guys was like, Hey, I use this thing at work called Mattermost. I have an old beat up Dell running Ubuntu in my garage. I'll upgrade. I'll make sure it's up and running. He installed Mattermost and we had our chat server immediately. So we had our treasure tracking and where is everybody? Side conversations with the DM as we are all trying to betray each other? Because again, we've been playing together for 30 years. When I was just like, this is great. It was such a better experience than trying to spin up a Slack or a Discord, much less a Microsoft Teams and it amazed me cause I mean, there's 10 of us using the server. So he was like, yeah, I just spun it up on an old machine that's now sitting under my desk and as long as the power doesn't go out, we always have a way to talk to each other every single day. And I think that's kind of the idea of Mattermost. It's super easy to use. Super straightforward. Set it up. Make it as secure as you want.
If you're interested in learning more about Mattermost, just type in Mattermost. It'll pop up. There's a .org. Most of the stuff ends up on the .com. Yes, you can download without giving them your email address. If you don't understand that, go listen to the earlier part of this episode and be on the lookout for the cloud version that's coming out. And if people want to follow you PJ, where's the best place to find you?
On the internet, I'm most easily found probably on Twitter where I go by @aspleenic A S P L E E N I C. That is also my username on several Mattermost servers, on GitHub, on Xbox Live. Long story short when I was 19, I lost my spleen and that username is available just about everywhere oddly enough. So yeah, if you want to find out more about that story @aspleenic on Twitter. It's well-documented.
We hope this episode was helpful to you. If you want to discuss it or ask a question, please reach out to us. Our contact information and the link to the Slack workspace are at https://www.devopsparadox.com/contact. If you subscribe through Apple Podcasts, be sure to leave us a review there. That helps other people discover this podcast. Go sign up right now at https://www.devopsparadox.com/ to receive an email whenever we drop the latest episode. Thank you for listening to DevOps Paradox.