 Live from San Jose, California, in the heart of Silicon Valley, it's theCUBE, covering Hadoop Summit 2016, brought to you by Hortonworks. Now, here are your hosts, John Furrier and George Gilbert. Okay, welcome back everyone. We are here live in Silicon Valley actually in San Jose for Hadoop Summit 2016. This is Silicon Angle Media is theCUBE. This is our flagship program. We go out to the events and extract the signal from the noise. I'm John Furrier with my coach, George Gilbert, Silicon Angle Media, Wikibon analysts and our next guest, Merv Adrian, great analysts from Gartner, always great to have on theCUBE with great insight. Merv, welcome back. Good to see you. Always a pleasure. Always good to see you. He has to be following this ecosystem for 10 years now. We've seen each other now. This is seven years in a row now. We had theCUBE at pretty much all the events up until Strata and now we've done our own big data events in New York and Silicon Valley, but we've been here covering it in the front lines as have you. Yeah. Thoughts because, you know, growing up. Yeah, our baby's growing up. It's growing up. Yeah. Can I have the keys to the car? So it's not, teenager? So what phase would you- No, I don't think so, not quite. I don't think we've quite hit the raging hormonal storms stage yet, but new friends, new activities, play dates are going to be something very different in the near future. There's a lot of near adult activity going on and the key I think for the business user looking at the news coming out of this event is to understand that the intent is really good. The right buttons are being pressed. The right software is beginning to enter the stack, but it's quite new. Yeah. One of the big topics here has been Atlas, which is a very, very important development. Metadata-based governance hub for the whole stack, but it's important to note that Hortonworks was very proud to talk about the fact that they got this from ideas and meetings to code in seven months. And, you know, in the open source community, that's a badge of honor, but in the enterprise community, it's a hang on a second, I'm not quite ready for this moment. Yeah, yeah, yeah. Yeah, in one hand, it's great accomplishment. On the other hand, whoa, no one goes that fast. Well, teenager, slow down. That's also an interesting dimension. You know, when you and I started doing this, much of the not supported yet innovation for critical strategic opportunities was being played out inside companies here, in the Valley, LinkedIn and Facebook and Yahoo were all building these pieces and happily contributing the open source code and being part of those projects to serve their own business ends. We've been hearing now from companies who are not software engineering companies doing the same thing. One of the poster children for Atlas here was Capital One, who's done some fantastic work because they're getting competitive advantage out of it. We heard Progressive talking about some of the work they've been doing with sensor technology and cars to do segmentation and variable pricing and giving back a half billion dollars to their customers. This stuff is really cool and these are not software companies. These are mainstream businesses with big brands participating in the community at a level that is substantially different than the one we saw five years ago. And again, the ecosystem is another metaphor of growing up, play dates, car pools. But as you- Sleepovers. As some parents realize, well, I love those parents, but the kids kind of don't get along and you lose the friendship and then you grow up and you find your real friends, if you will. Kind of happening now, but to the adult conversations you're pointing to is the customers. This is the dynamic. This is where we're seeing some great proof points. I don't want to get your thoughts on this because the KPI key performance indicator is how fast can you step up the value over the cost and complexity? So I know we've identified complexity and total cost of ownership, which has been a wet blanket of IT deployment. I mean, that's the- The department of no. You pull that out when you need to just kill a project. However, the competitive advantage component is so significant in this data-driven examples and you can't look no further than LinkedIn selling to Microsoft for $26 billion. I'm going to go to Ben Gallup at Docker, I'm like, hey, if they bought LinkedIn for $26 billion, I mean, you've got to be looking at at least three to 10 for Docker. My point is data has value. Yes, and- Thoughts. Data has value is the big story of the LinkedIn acquisition. The vision of combining the two professional graphs of the two companies in order to serve us a whole set of applications that will be enhanced dramatically by the ability to use that information, it's very clear, and they made a very compelling case for it. You know, there is one dimension of it that I think has not been commented on nearly enough and that is everybody likes to talk about the $26 billion and they always refer to it as a cash deal. It's a borrowed cash deal. Oh, because they couldn't repeat that. Well, money is cheaper right now than it's ever been and if you're Amy Hood, the CFO at Microsoft and you're thinking, what do I do with this pile of money? How do I invest it and separate it appropriately? This kind of an investment is actually pretty smart. It's likely to be a creative pretty soon once they start working with it. And having contacts in the cloud like with Office 365, having a social graph that's got interest graph, the future of work, collaboration. The other big takeaway, because we could talk about the same takeaways everybody else has gotten out of this and by now they're well-worn, but the other thing for me is that it really was the most visible and sizable manifestation of Microsoft's intent to continue to focus on the enterprise. This is not a consumer thing. This is not gaming. This is enterprise business applications being served better by enhanced collaboration and an enhanced social graph. Well, let's go original analysis on this and I'll take it a step further to go there. I agree. One, it's not a one-off, trying to make a phone like they failed before and other things, Skype and Bing, whatnot. Here, you're saying enterprise, but the consumerization of IT is certainly with the cloud and certainly what Docker containers are proving is that rapid agility is a consumer-like environment, but the value in the data, the consumerization of cloud-based technologies is going to change the work streams of users. That is an enterprise land graph from Microsoft. The expectations of the new generation of workers in enterprises are completely different than the ones that I grew up with on green screens years ago, or even people in the 90s with client-server technology. We had very different expectations of how the technology would interact with us and we haven't shed all those assumptions and bringing it back to what's going on here for a minute. When we start thinking about governance of data and we start thinking about Atlas and how people are going to use tagging to manage data security, we watched an industry or a category start to grow up two or three, four years ago in this space with companies like Trifacta and Paxata, all these data wranglers who are creating artisanal data management capabilities in an era when nobody wants to do that work anymore. Why can't the system do this for me? I want to automate this. I don't want to crawl the data. I want the machine to crawl the data and tag it for me and that's where we're going to go. We're going to let the machine do it, but there's a fundamental difference between this data that we are now thinking about securing and the data that we always secured before and it's this. It's the Dick Cheney unknown unknowns problem, okay? We built governance and security frameworks and hardened them and made them work very effectively for data we knew. We knew the content, we knew the policy and we knew how we wanted to optimize it to answer the questions we already knew we had. That's the data warehouse model. Now we're looking at data that is new. We don't know it very well. Sometimes it's not ours, it's coming from outside and the whole model of how we secure it changes. In the old days, it was authentication, authorization out at the network and there was a point at which, okay, you're through, now you can log into the DBMS. And at that moment, it became the DBMS's responsibility to manage security, but this isn't about DBMS's. These are file systems. So none of those capabilities- Yes, so streaming matters, so all this stuff matters. But to take that one step further, I'll even go up to the stack a little bit and say the application world of CRM and ERP automated known processes. Yes. And had unknown technology. Now you have the complete flipping the vector around saying we have known technologies, aka Lego blocks and Agile, and unknown processes that you can't automate until you see it. Well, let's go back to our opening analogy. Our kids want to play with a bunch of kids we don't know yet. What kind of safeguards are we going to need to put in place? We're going to give them social, emotional quotient. You got to give them some brains. That's machine learning. Teach them how to make good choices because they're going to have to make them. So the final question I got to ask you is we have limited time. Let's take that step one step further. I want my kids to be smart. They go to school. They teach themselves. They become learning machines. So AI and machine learning is all the rage. We kind of talked about LinkedIn as a big bet by Microsoft. In your opinion, what are the big bets right now because there's so many, there's a multitude of use cases, IoT and beauties in the eye of the beholder, but there have to be big bets to enable that ability to take advantage of the future, not foreclose the future, to teach our children, our machines and AI. What are the big bets do you see in the landscape out there? Some of them are not new. Sometimes we have to teach our kids the same thing three or four times. If we look back, we've tried with things like Duet before to connect the office worker to the enterprise apps. In many ways, some of this is the same. Some of those connections are going to look the same. There are apps in LinkedIn that are going to go away and be replaced by stuff in the Microsoft Dynamics Suite. All this stuff is going to get tied together in a different way. We have more ability now to combine additional data feeds that both inform and instantiate the actions that we want to pull off in the marketplace. So we can drive behavior, not just inform behavior. The truth is, you just reached over and touched your phone. You probably were looking to see what time it was or who's coming on next. My daughter's calling me. Your daughter's calling you. She's in college, she's not a teen yet. If you're out in the field and your Salesforce automation laptop or mobile device screen comes up and says, take another meeting in the next half hour. There's a guy five miles from here who's actually ready to buy. We've done all the work. Here's all you need to know. You're going to go do that. That's an opportunity to harvest that you didn't perhaps develop yourself, but you're going to rely on that technology. A notification economy we're coming into. An event-driven one. A curated event-driven one that will allow us to realize opportunities that we didn't know were there before. And so to get out of that mind, you got a mindset, you can't be overdosed with notifications otherwise you won't know what to pay attention to. They have to be the meaningful ones that count. And we can assess the likelihood of that being true with the data science we're building now. And we've got more data than ever before to test the models and prove them so that we can comfortably rely on the things they suggest we do. Murph, thanks so much for spending your time on theCUBE. Appreciate the insights. Great to see you. Merv Adrian with Gardner. Breaking down the analysis here in theCUBE with Wikibon and SiliconANGLE. I see all the analysts are here. It's theCUBE. I'm John Furrier with George Gilbert. Merv Adrian, you're watching theCUBE.