 Live from San Francisco, California, extracting the signal from the noise, it's theCUBE, covering DockerCon 2015, brought to you by SiliconANGLE Media, with special thanks to Docker. Now your hosts, John Furrier and Jeff Frick. Okay, welcome back everyone. We are here live for day two coverage of DockerCon. This is theCUBE SiliconANGLE's flagship program. We go out to the events and they extract the signal from the noise. I'm John Furrier, the founder of SiliconANGLE. I'm showing my co-host Jeff Frick, GM of theCUBE. Our next guest is Scott Johnson, SVP of Product, here at Docker, Industry Veteran. Welcome to theCUBE again, good to see you. Thanks John, good to be back. John, Jeff, good to see both of you. Absolutely. Thanks for coming out to DockerCon. We really appreciate you guys having us. We're excited to have you here. Great team, up and down, great management team, and still small company. Still small? Still small. 160 people and a turtle. Hard to believe. Where is the turtle? People want to know. He should be here, you're right. He should. He's a big part of the show, but he's chilling out without us in the office. He's actually glad that we're out doing the show, so he can just relax a bit. So what an exciting time for you guys, right? One, we've got to know you guys over the years. It's a great company, but it's really important technology at this point in the history of the innovation cycle that is the tech business right now. It is one of those once in a generation architectural changes in cloud computing. It is, I mean, honestly, we all are fortunate to be living at this time where we can see it, right? I mean, these changes don't come through every year. They don't even come through every two or three years. They come through like once a decade. Like I said, once a decade, maybe once every 12, 13 years, right? When the whole stack just is completely upended, the runtime, the management stack, and everything around it, and this is that time. This is that time. And we just feel incredibly fortunate that we're a company in the middle of that and able to work with somebody. You sold your one of your companies you were involved into Netscape, which is back in the internet days, back in the day, and that was a structural change with the internet. Obviously not as big as client server and PC, but still enabled now this change, which is cloud computing. So I got to ask you, just someone who's been in the industry now leading product at Docker. You got to look at the roadmap. You're talking to customers. What is the fundamental architectural change in cloud computing that is going to make this disruption, this innovation cycle, ongoing for the next 10 plus years? No, it's a great question. And we're going to keep, I bet you we're going to keep coming back to this question, right? As the years go on, because it's changing as Docker changes, how we think about compute, right? But to try to- And software development. That was the whole process, right? End to end. So I'll try to be brief, but this is a wonderful topic. But if we look at what the last decade say, has been done at the infrastructure layer, thanks to Amazon and VMware, right? That has really taught us the value of agility at the infrastructure layer. And what Docker has done has kind of really drafted off of that and taken those lessons to the application layer. And so Docker is now all about application agility on top of infrastructure agility. So now you have cloud computing that on a dime can spin up instances. And Docker at the application layer, which on a millisecond of a dime or some fraction of a dime a cent can spin up tens of containers on top of every VM. And you just have a very, very agile top to bottom stack now. That is, as we're discussing, kind of revolutionizing the way we think about computing. And Wikibon, David Floyd, one of our top analysts coined the term SLI software-led infrastructure. Jerry Chen earlier on theCUBE this morning called it DDI developer-driven infrastructure. Whatever you want to call it, it's software at the heart of the value proposition. That's exactly right. So that is apps, right? It's software and it's apps down versus infrastructure up, which historically is kind of constrained applications, right? It's been like, oh, what am I deploying to? And what's over there? What, you know, what's storage system I using? What network am I using? And that's had to inform how we think about the apps and how we architect them. And it's going the other way. That's a reversal of where it was before. Whatever my network was, dictated what I could do with my apps. That's exactly right. While my storage was, oh, that's the capability of that node. That node is attached to my sand. So that node can only do DBA, blah, blah, blah. It's being completely upended. So what at Amazon and VMware, you brought those up, because it's a great lead in my next question. They taught us a few things. I've called this the seven year itch of Amazon since they've been out. People have been watching them. They're just blowing it out the doors. It's not a race to zero. Some inspectively, the numbers are clear. Seven to $10 billion in top line revenue. Now they make the numbers work to break. Even that's the magic of Andy Jassy and Jeff Bezos. Yeah, so they reinvesting. So that flywheel, those numbers cannot be refuted. Oracle just showed yesterday, 300 million last quarter. This is on. Game on. Game on. Game over. So if you aren't into an integrated stack, hardware to software. Well, if you're not looking at these technologies and trying to take advantage of them in any way possible, right? If you're not trying to learn as an enterprise how to take advantage of agile infrastructure, how to take advantage of agile applications, like you're going to be left behind, right? And the beauty of- Or out of business. Left behind leads to out of business very shortly, right? And so, I think you don't have to take the whole enchilada at once. And that's what we're seeing in our own ecosystem is that you don't have to go right to microservices overnight. But you got to start with something simple. And we have wonderful use cases where a developer will bring Docker onto their laptop and get so much utility just out of redoing their environment on their laptop that that's what lights them up. And that's what brings them in and starts sharing with their friends and their developers. So start small, but the impacts, even on those small changes can be tremendous. So I got to ask the roadmap questions. I love product conversations, as you know. So the speed of changes for you guys is pretty incredible. And people want to go faster. It's like, hey, can you pedal any faster? Come on. So as a product executive, it used to be slow. I'm slower, hey, roadmap, next quarter we'll have these goals. How are you managing the product team, Docker, when you have a lot of pressure to solidify, decompose, share, make frictionless consumption of Docker. And then two, peer with the community. No, that's right. It's a challenge. So how do you prioritize, what's the focus? It is, it is. And what we're trying to do is find natural cadences for these different communities, right? For these different segments. Because you're right. While there's a lot of pressure for innovation from the vendors or from the community, there's also, as you might expect, folks who are running mission critical, business critical applications, we're like, hey, hey, hey, I'm not going to deploy to prod every day, every fresh change off the tip of the tree, right? And so what you saw, for example, yesterday, that we now have an experimental branch, right? And that experimental branch will be updated in real time. So if you want the latest, greatest, bleeding edge, you go right to experimental, you can go. And that's a whole bunch of the community members. It's some enterprises as well. And then there's a team that's just going to be shipping all the time. To the branch. To the branch. To the experimental branch. Yeah, to the experimental branch, right? Every day, nightly builds, right? We also, though, will have a stable branch that will have software that you can run your business on. And then, as we announced this afternoon or this morning, we'll have this commercial solutions package that will be supported for 12 months. So the APIs in that will be supported for 12 months. So right there, you have one, two, three. When you're 12 years in your Red Hat. Right? Well, 12 years in your Red Hat, but 12 months, 12 months. Well, they are the 10-year, 12-year SLA on there on rail, right? So, right? But for that level of the stack, it's too exact, exactly, right? So for their business, their point in the stack, like that's entirely appropriate, right? It's like, hey, my applications aren't going to change, so let's run out. The customer's one. But microservices is changing as we saw in the last 24 hours, 48 hours, microservices has changed. So quickly, no one's asking us for 10-year support agreements. 12 months to them, so let's try. They just try to get the production. Okay, 12 months. Yeah, we can work within an envelope. And then, at the end of 12 months, hey, next one, right? So just to try to answer your question, those are three different cadences. And just keep shifting that value out. Customers, as well as internally for different teams to work against. This business model conversation is all about marginal economics of software. So have you just randomly popped my head, the question that asked that you as a senior product person in software changing this much? Where is it not a race to zero? Because that's the skepticism of our idea. It's a race to zero, it's commodity market. Did you ask Jerry that question? Yeah, I did. I should have been paying attention to what he said. Look, I have a big believer in value shifts. I mean, the app's going to have that value. Management stack, whatever. That's exactly what it is, right? And it's exactly right. And you will go back kind of the first principles of this change that we're in the midst of, which is software's eating the world, right? And that goes back to applications being written by developers, right? And so while developers still don't have necessarily huge budgets, they are creating the new architectures that the operations teams need to support. And so in terms of creating a business model, a viable business model that supports economics is you want to make it super easy for the developers to adopt and the CI team to adopt and really have that architecture baked in the very beginning of the life cycle. And then it moves on into the operations team, into IT. And it's like, all right, we've got to support now this Docker-based, cloud-native microservice, whatever you want to call it, distributed application. And Docker, I need the tools for that. And so you saw our price point, $150 a month. $150 a month subscription for 10 Docker nodes. I mean, that's more than, I think I joked, most of us will pay, that's less than most of us will pay for drinks in the two days here, right? So that helps the team get started. A lot of beer at 18T Park. Yeah, a bad beer, a bad beer at 18T, right? Let's go over there after this. And so you'll see a team land with that, very economically, right? But by the time you get to operations, they need support, not just for 10 nodes, but for 10,000 nodes. And they need complete tooling to integrate with their puppet, with their LDAP, with their security system. And all of that, they are very willing to pay for. And so that's really how you create a sustainable business, a viable business out of this. So I got to ask you, as you guys are really doing well, a lot of integration without touch points in the ecosystem and IBM's here, a lot of vendors are here, the big whales, pun intended are here. Yes, absolutely. But the big themes coming out of DockerCon are open container standards. Yes. The cross host networking. Yes. Which is really cool. And the Docker plugins. I would say kind of mean my big meet and the bone, some politics and some kumbaya, free love with the community. But the standard speaks, this speaks to what, where the status is. I want to get your thoughts on those three things. Yeah. The coalition around openness reminds me of Linux days, a forcing function. Absolutely. Coming together. No, you nailed it, right? I mean, look, the market is moving too quickly and has the potential to be too big, too disruptive for all of us to fragment, right? And waste time on really, really arguments that don't bring any value to the customer, don't really bring value to the innovation at the top, the higher levels of the stack. So this was a result of everyone just kind of standing and saying, look, what's important? What's important? Is that we have a big, growing, exciting market. That's the most important thing. Let innovation happen up here and not down here. The fruit in the tree is in the valley of greatness. Oh my God, absolutely. And while we're squabbling over, you know. Don't do it. Don't do it. Let's not have premature fragmentation. And so let's just squash that. So that's what the OCP OCF is all about. The networking piece, I think. By the way, it happened in a very short timeframe. Which is also an indicator. It did. Of the cohesiveness of the community. Alignment, everyone recognizing the right thing to do and everyone coming together. And again, I'll give a shout out to CoreOS and Alex Povey for just really aligning with us very, very quickly and understanding all of us together. Like understanding, like this is the important thing for the users, for the other vendors in the ecosystem as well as for obviously the participants. So. Yeah, shout out to him. Good job. How's it off to him? Yeah, absolutely. Look, number two, networking. Here's the headline. We're doing for networking yesterday. What Docker did for compute two years ago. Full stop. I mean, it's going to be as disruptive as what, with the way we've. You see it being very simple and elegant for someone to configure networks. Well, and the portability again. Right? Because think about doing networks in the battle days. Two or three years ago, right? Like, you built it once. Oh, it moved to a different set of infrastructure, different stacks. Again, going back to the previous argument of like infrastructure up or up down, right? It used to be, oh, we just moved to a different network stack and oh, what's who's our provider and what's their standards? That's bottoms up. That's infrastructure up type informing of the application architecture. That's gone. That's gone with yesterday's announcement. Like it's now app down network definition and mind blowing, right? The portability of networks. What is the big enabler for that? Because this is a huge concept. It's a huge concept. So look at the plug-ins and the things. Especially, sorry to interrupt John, but especially if like microservices is about network containers, right? So it used to be networking tens of VMs. Okay, great. Now we're networking hundreds of containers, thousands of containers, tens of thousands of containers. And virtualization. Wow. Unlimited potential scalability. How important is the network in that model? So important. So to disrupt it in the way we disrupted it, at DockerCon, like I think we'll look back and see this as a simple moment. But this makes this profound because it makes the network more, it makes the apps more compatible with the rigidness of networks, but yet there's some flexibility there. But if you try to bring rigidness up to the apps. Yeah. We've seen policy. And that's better, right? So the policy-based is very limited. That's right. I'd much rather push policy down to an unlimited to a... What does my app have to do? What are security requirements? Where is it allowed to run versus not allowed to run? Great, and that all goes with, not just the containers anymore, it goes with the containers and the network fabric. Right? It is my life. Okay, orchestration. Yes. Orchestration's a big deal. Huge. Who manages all of this? Is Docker going to manage it? You can have your own management service. And there's people out there saying, hey, I'll run some Kubernetes over here. I've got corollas and mesos. And well, this is going back to your first question. Like this is where so much innovation is going on right now at this layer in the stack, right? It's the top layer of the stack right now. And honestly, our play is similar to what you've seen at every layer all along, which is batteries included, but swappable. Meaning like we will provide an out-of-box experience that will get that to help us. The batteries included me. I know the metaphor for the folks watching. Yeah, no, sure. Thank you. Thank you, absolutely. So it means we'll include an implementation that works well for the developer on their desktop out of the box. So they don't have to go and piece stuff together. My battery's going to speak. That's right. So what's an example of that? So Swarm, right? Swarm has built in a little cluster manager that works great out of the box for a developer who wants to manage a couple of nodes and containers across those nodes outstanding. But they want to then move that application into the data center. Maybe Ops has decided on Mesosphere or maybe they decided on Kubernetes or maybe yesterday's announcement. Maybe they decided on Amazon ECS for their back-end cluster manager. Great. The developer can still use the Swarm front-end API. Again, portability, right? To launch their multi-container distributed app. You guys want consumption and frictionless acceptance. And of this app architecture. Of this app architecture. And we don't want to have it slow down as a result of infrastructure battles. Great. So we'll provide an adapter, we'll provide a plugin for those different infrastructures. So it's a complicated, it's a long-winded answer, a complicated answer to your question. But the point is like, Docker doesn't want to control orchestration. Docker wants to provide a great portable, portable application definition that can work with all these tools. And by the way, that decision of who owns it or who participates the most economically is still downstream. It's still up for grabs. It is, it is. It's not necessarily set in stone at this point. And there's no foreclosure right now in the architecture that you see. There's not. It's not a jump ball. There's not, there is. And you see, depending on use case, a whole bunch of different vectors. Which again, big pie, right? Big pie for all to share. You see cloud-based ACS, Amazon EC2 ECS, clearly a cloud-based solution. Mesospheres, trying to straddle both. Kubernetes, optimizing for Google's workloads. So there's still a lot of different use cases, a lot of different customer input that's informing this innovation. So I think, I think there's- It's still a jump ball right now. There's a lot of jump ball, a lot of innings left in the game. All right, so I want to ask you, Jeff's kind of got to get a word in. I got to ask you a question. Sorry, I keep rolling. I got it. He's rolling. Well, no, we have the big, big brain here, experience and- How about that? Well, you've got a lot of experience and- Yeah, the curl shirt. Jerry's behind me, right? You get the cool, Jerry Chen. He's just a figurehead. He's an investor. He's the bank. But Jerry is super smart. We love Jerry. Now, if I was going to ask. No, I was going to ask. With all the data, Amazon, Oracle and others, it's clear there's a money-making opportunity. So I want you to just share the audience, your perspective and vision. There's not so much a doctor-centric question, but more of your thinking of yourself as an experienced consultant to all the viewers. Where's the opportunity to make money? If the assumption, if you believe that there's plenty of beach head for everyone to play. Yeah, yeah, there is. And why fight over, you know, the apple trees or fruit trees, whatever you want to call it. Right. How do you explain that to someone who's young and in the industry? Yeah, fair, totally fair. About this concept that if we all win. Yeah. Out in the valley, it's fruitful grounds. Absolutely. Share your thoughts and vision on that. Absolutely. And look, there's been a lot of platforms and a lot of very smart platform architects that have come before us all. And I think you can learn from history a lot, right? And honestly, the X86 or the Wintel platform is one of the really good examples here where I'll simplify to two different areas to play, right? One is extending the platform, right? So I just mentioned batteries include a bit swappable. The butt swappable is a way to extend the platform. And we announced a bunch of extension points here today. We've announced a number over the last 12 months. All of those are places to play. If you're an infrastructure provider, if you have a special tool that does an implementation that is differentiating and valuable for your customers, you can absolutely make a good business out of that. So extending the platform is one. Number two is building vertical solutions on the platform. Full stop. And we saw this again in the Wintel days in spades, right? It's like Docker is a phenomenal unifying platform across the internet. I mean, our salmon's big vision yesterday, right? Make the internet programmable. We are the unifying platform across the internet. That is a horizontal platform. There could be verticals for healthcare, financial services, banking, manufacturing. Yada, yada, yada, yada, yada, yada, right? All those. How is this different from the old platform wars where it was when one guy wins and everyone loses. The Microsofts, the early days of Oracle. We own it, we own the platform. Monopolistic, some say, and this is Microsoft was ruled against it, but that's the old model. But open source is the new dimension. That's right. Thoughts on that. You see it very viable that there could be a unifying open platform. Absolutely. Absolutely. There's more wealth for the participants and the platform owner. That's exactly right. The OCP from yesterday is an example of that. So it's not the old platform days, to your point, where it's like, it's a silo lock it down, right? And you got to get through us. You got to get through our tax toll booth to make it work. But the OCP is a great example. Like, nope. We're not going to try to create taxing toll booths that are going to slow down the industry and exact exorbitant tolls from participants. It's quite the opposite. If you make it open, if you make it standardized, then it's actually a bigger pie for more people to kind of play in and eat from. So the old expression, I only count how much money I make, not the other guy's money, is your philosophy. Yeah, exactly. You guys will make a good business. Phenomenal business. Enabling people to harvest. And participants will also make a phenomenal business on this. And I think what's different about this, open source changed the game. Absolutely. If there was no open source, it would be the lock-in. Yeah, like land-grade. It was the only model. That's right. And now open source. The land owner, the building owner, and the... It clearly blows it open, right? Yeah. And no one has to be seen as like, am I inside the silo or outside the silo, right? Because there's no silo. Thanks to open source. All right. And so for you guys, value for future monetization is... What? Unlimited. Ha ha ha ha ha. Management software, on top of the stacks. Yeah, I mean, what you saw this morning is really along the ways of the management software, right? So the runtime is open source, continues to be open source, will be open source. The users are asking for tooling to help manage these distributed applications. And we talked about two tools this morning. We talked about the Docker trusted registry, which helps manage the images that are being stored and shared amongst team members across the life cycle. And then we did a sneak peek of Project Orca, which is management tooling for actually deploying and managing the distributed apps. Once they've been deployed to a data center, deployed to a cloud, deployed across clouds. And so we think the platform openness allows lots of developers to build great applications. And it gives us an opportunity to just, you're going to fraction, just provide management tools just to a fraction of that market. And we're going to have a phenomenal business. All right, Scott, we're getting the time limit here. Final point, share with the audience who are here. Some of the conversations, the vibe, top stories from your perspective here in San Francisco, here at DockerCon. Yeah, you know, it's one of those conferences and we've all been around a lot of conferences. It's one of those conferences where you walk in and you feel the excitement, right? You feel the excitement in the hallway track, hallway track of talking to vendors, talking to each other. You go into the track sessions and you're hearing. You're hearing about name brand Fortune 500s in production with the product and they are on fire. And you can tell that these are the guys that are early adopters inside of GE and inside of cap one and programs like that and enterprises like that. And they- They're the Navy SEALs, they're the elite forces. And they have a vibe about them that is super positive and can do. And you see the participants jumping up and asking questions, how do you do this? How do you do that? Not just technology, process, culture, people, like the whole stack. And so you just look at that energy and I got to say, like having been to a lot of tech conferences, you don't see that level at all layers in the stack at all conferences. And final, final question is inside Docker, what's the culture like? What's the- It feels a lot like this. It's like we can do it. I mean, Project Orca was not a topstown decision, right? That was an engineer who saw a need, went off hacked on his weekends. I should have crawled this out in the keynote. Evan has a full-time day job. His full-time day job is working on Docker machine, which as you know, is our host provisioner solution. He goes home at night, goes home on the weekend, hacks away, and he came up with Orca on his own because of that passion, because of the vision of like, I can see our users needing a tool like this to deploy distributed apps. We're going to be having theCUBE. That's a tech athlete right there. That's- Working overtime. That's working overtime. That's a vision. Okay, he's got shots of SVP of product here inside theCUBE, sharing all the data we can at DockerCon, this is theCUBE. We'll be right back with more after this short break.