 Live from the Mission Bay Conference Center in San Francisco, California, it's The Cube at Google Cloud Platform Live. Here are your hosts, John Furrier and Jeff Frick. Okay, welcome back everyone. You're watching The Cube live in San Francisco at Google's Developer Conference. It's called the Google Cloud Platform Live event. Breaking out from IO to get their own event. A lot of stuff, so many updates going on. They're going to have their own event and continue to pound the pavement. I'm John Furrier, the founder of SiliconANGEL. I'm showing my co-host, Jeff Frick. Our next guest is Stephen O'Grady, co-founder of Principal Analyst at Redmont, big fan of the firm, James Governor. I was just on Twitter calling out. Someone said that App Engine was the first past layer and then they were re-corrected here on The Cube. Stephen, welcome back. Thank you. Good to be here. So, James jumped on my little tweet, like, oh, I haven't just said App Engine was the first past layer. I'm like, no, maybe they're correct in saying the first real past layer that had traction. So, interesting how they talk about that. But shout out to James. Good catch. What's going on, Redmont? What's going on with the developers here? Give us the scorecard. The scorecard is basically that the things that we've been saying for a long time are now finally popular. In other words, when we were running around 10 years ago saying that developers were the most important constituency in technology, people said they were crazy. And you come to events like this and it's all developer-developer-developers. All people want to talk about it. And I think a lot of the announcements at the show today reflect the fact that like Amazon, like all the other public cloud players, the essentially intent is to compete for developer mindshare and to try to give developers what they want. Everyone has this bumper sticker, got to win the developers. I mean, that's kind of like a weird statement, winning the developers. Developers have their own muscle right now. So, I guess if you were going to say how do you win the developers, which means win them over in an open-source framework, which is now the standard kind of ethos, how do you do that? I mean, how do you kind of ingratiate into the developer community if you're certainly IBM's trying to hack the league and HB? Well, how do you do that? The simplest way is honestly convenience, right? So, in other words, you come to conferences like this, everybody always wants to talk about features and engineering and performance and scalability. All those things are great, all those things are wonderful. At the end of the day, the thing that more often than not wins the day from the developer perspective is convenience. In other words, we use this analogy all the time. If you think about the public cloud, in almost every case, the virtual instance that you spin up on a cloud is going to perform worse than a physical equivalent. The difference is the physical equivalent, best case you can have it in a day, probably you're looking at days to weeks, and I can spin up an instance or more importantly, thousands of instances in 90 seconds. So, from our perspective, trying to get out of their way, give them the things that they want and then... Faster time to beer. I mean, that's basically the algorithm. That's right. We know that developers like to drink beer and tea from our analytics. That's what we do at our conferences, too. We just drink beer. So, convenience means a lot of things, basically time, energy, getting down and dirty and stuff they don't want to do. That's right. And that's where the DevOps thing kicked in, where we're hearing here, cloud, like, okay, I'm going to extract away the network stuff from the app developers, make the network guys do their thing. So, how far are we in on this revolution there? I think it's still actually early days. In other words, I think a lot of the technology has come a long way. There's a guy by the name Flip Chroma, we use this quote all the time, where, you know, essentially the cloud means that anybody with a $10 bill, you know, can have a 10 node cluster with a terabyte of distributed storage for eight hours. And if you think back five or six years ago, I mean, that was literally incomprehensible. If you wanted any of that, you have to go out and wait weeks and spend lots of money. You know, we can do that now. All that being said, you know, the quality of the tools and quality of the mutation, there's a long way to go. We're still figuring out, all right, we have all these wonderful tools. How do they fit together? Where do I use this? Where do I use that? How do I sort of sift through all the available options at my disposal and make things easier? So, I'll get your perspective on this. Dave and I always talk, and Jeff and me on the Cube, is a great tsunami of organic growth, certainly in the developer side. You guys have been really the best analysts from out there covering the developer angle. They've been watching you guys for a long time and you're number one in that area. And they're certainly awesome. You just, you don't have to rack and stack and get a startup up and running and do a project. As you said, it's not ideal, but it's good enough. And you get the job done and you're fast, right? And then you got on the other side, infrastructure massive shifting going on. So, is there a glass ceiling here where we're all kind of bumping our heads saying, that's the down and dirt. You said convenience. Now we're at, okay, I got my app up. Now my boss wants to fund it beyond POC. I got a role in the production. And these guys over here got this new conversion structure box. That's changing everyone's role from TBA. Storage admin goes away because there's no more storage. So, I mean, you got this massive like shift. What is that ceiling? Are we hitting that yet? I don't know about a ceiling. I mean, I think, you know, part of what organizations are struggling with now is, you know, they recognize that, look, we can do things faster with the cloud. The question is, you know, you can't have everybody spinning up their own instances on different clouds and expect to have any sort of central visibility, accountability, you know, cost control and so on. So we're beginning to see sort of the early days of, all right, look, developer, you want to use a cloud that's fine, but you need to do it through central accounting. You need to do it through centralized tooling such that I have some idea of what's going on. I want to give you what you want. I'm going to get out of your way. I want you to be productive, but you have to give me something in return, which is, again, some accountability, some visibility into what's being spot. Well, it's interesting. Yeah, that's a great point. We just had the Forrester guys on earlier and we brought up the Amazon dominance calling the secretariat thundering away, still winning. And certainly Google bringing this play where they have the large-scale back-end saying, so if you get lucky in this, I'm feeling lucky mindset, you know, you get pops and back-end scale quickly. That's a great, love that vision. However, the CIA is now moving aggressively 100% on Amazon. They're going to announce next week that CIA is going to go all consuming. Another government agency has Coroner Forrester saying, that has actually the big issue because of one reason, accountability and essentially answering to voters. So now that mindset comes to the enterprise. Is that trend hit the enterprise? Certainly that can make sense on the government side. Spend less, do more, and show me what you've done. Well, we're always asked by press among other people, hey, is private cloud a real thing? And my answer to that always is, let's think about the characteristics of the public cloud, right? It's elastic. It's quickly provisioned. You have granular billing and so on. Why would you not want your private infrastructure to have those characteristics? So yes, on some level, we're going to see a lot of the public cloud elements infiltrate. We typically thought about the enterprise data center. The question is, how much is going to be public versus how much is going to be private? Because look, at the end of the day, the reality is that the Googles, the Amazons, the Microsofts, and so forth of the world are probably going to be better at running these data centers than most private organizations. And I don't care how big they are. So that's the question in terms of, it's not whether private infrastructure looks like cloud, it's what the percentage is in terms of adoption. All right, so we had a joke earlier on the cube called the hot tub time machine, IT, where you drop into the 80s and you have client server infrastructure, your IT staffs, look at the storage admin, who hates the network admin, and they really have no development force. It's all been outsourced and that carried into the 90s. And so that being said, great, virtualization came along, great, server consolidation, a little boost, a little steroid shot. But now there's a huge demand for enterprises to hire developers. We want to bring it in-house almost during like the heydays of the mainframe days where in-house development is cool and being invested in. So that's kind of been realized, people kind of agreeing with that general direction. So I got to ask you, how does an enterprise who really has not have a core competency, I'm not talking about the financial guys, because guys are always kicking ass, right? Your traditional enterprise, you know, stuck in the 90s, IT, or give them 2000 and one maybe, or they got a boot up fast. What do you guys advise those practitioners? What do they do? It seems difficult. Yeah, it's funny, we were on a conference every year in October, we had Adrian Cockroft, I'm going to speak, the CUBE listener is may or may not be familiar with Adrian, he's one of the original architects of Netflix. He's been on the, yeah. Yeah, he's been on it for many times. Very, very good guy, very smart guy. He's pretty good on Twitter too. He's very good on Twitter. The funny thing is, is that he apparently has organizations, he's left Netflix, he's now over at Battery. So there's a lot of organizations coming to him and saying, where do you get these people? Like Netflix is like special unicorn, and you can go off and sort of run only cloud and so on. Where are you finding the developers that can build this stuff? And his response is that, we heard them from you. Like none of your businesses would let them do the things that they want to do, and therefore they come to work for us, as we get out of their way. So outward migration from the enterprise, now how do they get them back? It's like living in North Dakota. It's like everyone leaves as soon as they go to college, and they never come back, right? So like, no offense to North Dakota, but I mean, great place to fish maybe, but how do they get them back? Is there a way you found that, because that's a real strategic imperative? Honestly, you get out of their way. Really, you get out of their way. And like I said, it's not just sort of inmates running the asylum, like let them do whatever they want. But on some level, there are certain tools, there are certain things that they want to do. They're probably going to be clouded, they're probably going to include open source, they're probably going to include a lot of things that maybe your organization is or isn't comfortable with. So you can either say, no, you can't use these things and see them all sort of run off, or you can say, all right, you know what? You're going to use these things, probably be more productive and faster and more efficient the process. So yeah, I'll just get out of your way. What's your take on legacy migration? Old apps running on old boxes that may or may not have a benefit. Obviously, if it's new development, it's a great environment, easy to spin up. But if the guy in North Dakota is stuck just running old systems, are they just eventually going to die? Do they last forever like the airplane reservation system? I mean, what happens to those? Honestly, a lot of those things are never going to move. Just flat and never going to move. And they shouldn't, right? Not really, not really. In other words, every year Oracle and suppliers do a lot of those enterprise legacy applications will raise prices and the press comes to us and says, hey, is this when everybody runs out and leaves them? And our answer is, look, let's walk through this. In other words, sure, the prices went up, but the cost of moving is a lot higher than the price hike. And for legacy applications, are there advantages, both sort of from a functionality and a jolly standpoint, as well as a cost standpoint, in some cases, to moving towards the cloud? Sure. But in other words, you want to bear the risk, right? I mean, again, it becomes a cost benefit. If I have this application and it's been running forever, the airline reservation system runs fine where it is. We know how to maintain it. We know how to manage it. The risk of moving that is very, very high. So if the risk is high, the benefit to moving it has to be that much higher. In a lot of cases, we're just not going to be there. So a lot of the, I mean, you see conferences like this, all the public cloud conferences, a lot of what they're talking about are net new workloads, as opposed to, hey, let me take that thing that you've been running for 30 years and moving into the cloud. So the hardest question I want to ask you, this is a tough one. So brace yourself, it's controversial. I see you wearing your red socks now. I'm going to say better be by the red socks. You know, I love my Giants. I've co-opted that tee because I moved out of DF-50, but I'm a Diard red socks fan. They give away their best pitcher. What's going on? Give us the analysis. Break it down for us. I never understood that deal from the get-go. Why they got Sassapadis is a great player. He's fun to watch. But yeah, you trade away your best player and eat for a guy with one-year service time. I trust Jarrington in spite of the fact that he went to Amherst. But yeah, I never understood that. I think you're going to ask him to lay out on getting Pablo. I thought that was the question. Yeah, I think they're going to renew it. I liked Pano, but I think he ends up back in San Francisco. So what's your prospects for next year for the Sassapadis? I think they're actually in good shape. The farm system is really good. The lineup will be much improved. They'll have to trade some of the outfielders to get some pitching. But yeah, I throw a couple of pitchers in there and improve lineup. I'm a Bruins fan. Chara's injured still. I think he's out. So Bruins don't look good. While the Giants are awesome here, you know, of course I'm reliving that. Reliving that on my red side. Back to my two favorite teams. How's James doing over there at Redmonk? James is doing very well. Yeah, James is doing very well. He's like, we all are. He's running around. We've got conferences and engagements and speaking engagements and so on. So yeah, we're all busy. So what's the coolest thing you've seen here at Google Cloud Platform Live? Coolest thing I've seen here. Probably the... It's not terribly surprising. I think one of the more important things I think we'll see out of this is essentially the container components. In other words, having a container engine docker for those of you who've not seen it say this before. Out of all the technologies, we've tracked at Redmonk over the years. We've been around since so too. This is easily top two or three in terms of quickest growing... Docker or... Docker. Docker. Okay, docker. Docker more specifically. So basically having a system for the maintenance and orchestration and management of docker containers and giving developers again what they want in docker is a big deal. What's number one? Historically? It's a good question. I don't know. We'd have to look. No.JS was very, very quickly growing. It's not number one, the top three. Yeah. I don't know. What would be first? Well, we're going to get you guys breathing our crowd chat application which is all known in the cloud. Breathe that ass real time database. Any other predictions for the show? Ooh, predictions for the show. People are going to like it. I think they'll like it. In other words, Google historically has done a pretty good job with giving developers what they want. They have a lot of expertise. And like I said, I think they're, you know, obviously based on the investor space announcements, I think taking it to the market seriously. So yeah, I think they'll like what they're saying. You think they're pushing fast enough and they think they're just taking their time with the army first waves coming in? Well, I mean, in other words, I think they have to look at it, say, you know, you're not going to necessarily go toe to toe with, say, an Amazon, you know, from the feature perspective. You know, because I think it was 2013, I want to say Amazon dropped 230, 240 individual features. And none of the cloud providers are going to compete with that. So, you know, if you're trying to sort of make up ground, basically you try to figure out, all right, what do we go to, what do we focus on? And particularly, you look at their container strategy, they know how to do this, have been running containers forever. You have Kubernetes, you know, managing Docker containers which developers love. That's a good place to go. Are you surprised by the speed of how much things are shifting right now from a vendor landscape? I mean, hopefully the stack seems to be on its heels. You know, this is kicking ass. Docker, you said just first on the scene, goes kicking ass. You got Kubernetes, a lot of pressure at VMware, almost chameleon and all this. We're actually, it's interesting that you asked that, because we're beginning to track a little bit of a, a backlash is probably an overstatement, but things are changing so quickly, the developers are actually having a difficult time keeping up, right? So, you have web developers, and I wrote about this a little while ago, Marco Arment, who's, I don't know, probably like eight or nine at Tumblr, created Instapaper, you know, a very, very high profile developer, wrote, not too long ago, he's like, I'm actually tired of being a web developer. There's too much to keep track of, there's too much going on. So, we're beginning to see a little bit of fatigue. Yeah, it's not convenient. I have too many tools. Look, I just want to build something. I don't want to have to figure out how all these things work together. So, am I surprised by how quickly things are changing, not necessarily surprised, and that reaction I think will be, we'll see a lot more. Certainly seeing that in the social media space as well, people want another platforming tool like a whole room here. They want more convenience. Yeah, convenience speed. All right, Stephen O'Grady from RedMonk here inside theCUBE, breaking it down, developers, developers, developers. There you go. And they want convenience. They don't want another tool and another platform to figure out. They want convenience, faster time to not value, faster time to be here, as we say. So, this theCUBE, we're going to have a beer shortly. It's dropping off towards the end of the day. Stay with us. We have more coming here. Live in San Francisco, Google's Cloud Platform Live, their developer conference in San Francisco. We'll be right back after this short break.