 Everyone, welcome to SiliconANGLE Media's theCUBE, CUBE Conversations here in our studios in Palo Alto. I'm John Furrier, our next guest is Siddhartha Agarwal, CUBE alumni, we first met at Oracle OpenWorld, really focused on the developer open Oracle model of bringing developers together with the Oracle and Oracle Cloud, Vice President of Product Management and Strategy at Oracle. Siddhartha, well, good to see you again. Welcome to theCUBE Conversation. Thanks, John. Thanks for having me here. So one of the things I like about Oracle in covering you guys for now on our eighth year, we've been at eight Oracle OpenWorlds, is we get to see the transformation. And one of the things, and we also cover Amazon web services like a blanket as well, Microsoft is sure not so much because they're just now new to the scene, but you start to see the big players really flex their muscles in the cloud, Oracle's one of them, is that Oracle on Oracle is a superior product because it's Oracle, Oracle Cloud is a superior product because it's Oracle in the cloud for Oracle customers. But you guys get a lot of criticism in the market from competitors around Oracle, not on Oracle. So people that are new to Oracle, Cloud Native, for instance, is a market. And I just think that you guys, that's not a fair, it's always people always attacking each other, but more importantly is that you guys are doing the work with developers. And so I want to talk about that with you because you have a developer focus that's about being open. Share this, this is important. I want to take a minute to explain the Oracle Open. Yeah, no, I mean, absolutely for us, it's very important for developers to understand the Oracle Cloud Platform is open, modern and easy. What do we mean by open? It means you can bring your programming language of choice. So you don't have to use Java EE. You could use Node, Ruby, PHP, JavaScript, whichever language you want. We're giving you a choice of databases. Obviously Oracle databases, you could leverage those and get great value out of them with high availability, redundancy, reliability, et cetera in the cloud, delivered as a service. But at the same time, if you want to use open source, for example, MySQL, we deliver that as a service, we give you the choice of using Cassandra or MongoDB on our IaaS platform or on our PaaS platform. So lots of choice for the programming language for the database and choice of the shape that you want to use on IaaS. So for example, you could use VMs, you could use containers, you could use just bare metal. So giving you screaming performance on bare metal as a developer, you might want that. And then running any of the open source workloads. So if you want bitnami images with one click, you can run that on Oracle. So our focus is not just for Oracle content, but to be an amazing cloud for Oracle and non-Oracle content. This changes the ecosystem dynamic certainly for Oracle as you guys expand, but you guys have open source, you have Java, you have all this stuff. It's not like it's new to you guys, right? So a lot of people, you know, kind of it's important for them to know that. So I got to ask you a specific question. As you guys have a more of a focus on open, which is just classic ecosystem development. The world's changing, right? So right now, the hottest topic in the world is really two things, IoT, Internet of Things and AI. And AI has now surpassed artificial intelligence. AI is not surpassing IoT because most people don't even know what IoT is. So they just, but AI sounds sexy and glamorous. You know, intelligent machines, flying saucers, flying cars. There's a lot of stuff going on that's real engineering work and R&D and practical applications that point to that AI and data is certainly a hot market and certainly relevant. Your thoughts on how this developer focus connects with that world? Actually, I think that developers are gonna become even more powerful than they already are because artificial intelligence requires two things. You need to be able to have these smart algorithms that are able to do analysis, but those smart algorithms cannot do anything unless there's really valuable data behind it and lots of data, right? So I think there'll be this next generation of data providers for AI. For example, with Oracle, we're providing the data cloud. 1.2 billion consumer profiles, 400 million business profiles. So now when you want to do segmentation of a particular customer, a developer can call APIs and without having to go get all of that data, we're actually aggregating that data off of all the websites visits and all the on-premise transactions that are happening at various departments, stores, retail channels, et cetera. We're aggregating all of that into the profile of who a consumer is and now these developers can access the data through their framework, libraries, et cetera. So they can actually write really intelligent applications because they've been able to leverage this data to train them. I speak into the choir here. I love this so much because I wrote a blog post of 2008 data is the new development kit and it was kind of riffing on this notion of development kits when the old Microsoft days, MSDN and people have development frameworks were there for their stacks, but the premise was data is going to be a development ingredient, a critical one. So the question for you on this AI thing is, does dated Trump algorithms or algorithms trump data? Or some are saying dated Trump algorithms, some people saying algorithms trumps data. Which comes first? Is it the chicken and the egg, the cart before the horse? What is your thoughts? I mean algorithms have always been there, right? What's been missing? The parts that have been missing is one, being able to accumulate this much data and two, to be able to process the data and three, to have the compute power to do predictive analytics at scale, right? So algorithms have always been there. There have been smart people that have written these algorithms, but those three things are the ones that have prevented artificial intelligence from becoming real. And now the data being aggregated, the speed of compute and the amount of compute that is available to you, those are things that are going to drive artificial intelligence to become real. So power's there, you got the computing power, you got the data and the algorithms that are just gonna keep on developing. And that's where machine learning seems to be the real transformative ingredient in that. Awesome, next topic is my favorite one, which is I've been talking about on my show on Silicon Valley Friday Show, which we do every Friday morning here in the studio, is the whole chatbot craze. And you and I were talking before we came on this interview about how that really is evolving, this deeper meaning behind what chatbots actually are. I've been calling them kind of like glam and sizzle to AI to give people a mental model, but most chatbots are kind of gimmicky. Customer support here and there, some applications, but there's deeper meaning behind a chatbot application. What's your thoughts? And you had mentioned to me that you think by 2018 you're gonna see chatbots. Some are saying it's already here. Why aren't chatbots ready for prior time? No, look, we as consumers are getting tired of using mobile apps. So for example, I fly a lot and I download a Lufthansa app, a United app, a Delta app, et cetera. I'm downloading all of these apps to check in, do the same kind of stuff. But I have one messaging platform that is really popular, which is for me, Facebook Messenger, for example. So I wanna be able to go and engage with Delta by friending the Delta bot and be able to say, hey, give me my boarding pass or do something, right? So I'm tired and fatigued with all the apps that I have. And so the next generation of engagement is going to be around, how do I engage using my messaging platform of choice? And you asked a very, very good question, which is why haven't these bots been more successful? It's because it's not easy to build a bot. What are the things that you have to think about? You have to think about a platform that scales across Facebook Messenger, WeChat, WhatsApp, et cetera, right? So you have to write that for various different platforms. You have to have natural language processing built in. You have to have an intent engine because people don't necessarily say the same thing every time when they want to get something done, right? Like I could say, tell me my balance, right? Now is it for checking or savings? Or I could say something different next time. And so you have to have... There's no reasoning, there's no linguistical kind of reasoning engine. Exactly, but you've got to have that, right? An intent engine. And then you have to learn as you engage with a particular user, you have to start learning how do they engage with you? Those are the kinds of things that we've built into our mobile chatbot platform. So developers can focus on writing very rich bots that engage the developers to cross the web cell. I mean, the customers into cross the web cell, but not have to necessarily worry about the underlying infrastructure that is required to make this happen. Yeah, the big theme with every interview I do in the cloud is around operating systems of the future, which is that global internet, basically. You tell them about basic IO. Think of it like a PC. You plug a keyboard in or Bluetooth. You got input, output coming in. This is kind of the interesting thing about chatbots. You want to unify your mobile experience, but in a way that unification isn't just mobile. That's other platforms too. Isn't the unification the core theme driving some of these challenges? No, it's about which messaging platform is important to you and how do you bring in all this data and the analytics capability that we were talking about before around the monocomput that is available to apply that intelligence to the bot so that when someone is engaging with you, they're not feeling like, oh, it's a machine or it's someone that has a script that they're reading. It's actually someone that can understand your intent. That's what the unification is. And also reduce the steps it takes to do things to your example. I mean, I was saying early on with mobile. I mean, why download an app when you only need it once or twice or it's seasonal. The NFL is those seasons over. The sport season's over. Now I got an app on my phone and got to wait another year. Right. So if I type in boarding pass, it should automatically figure out, okay, which is the next flight I'm going on. Am I within the 24 hour window? And oh, that's the boarding pass he probably wants. Well, Google's doing that right now. Actually, the boarding pass right there. Google's doing that right now. My email's synced up. Okay, the next one, let's go down deeper in the stack. Let's kind of drill down. Cause now you go to the next question. What is the key driver behind that? Is it the microservices? Is it the containers? Is it the DevOps? So assuming that that chatbot stuff needs to be coordinated and unified, what's under the hood that makes this all happen? Well, I think, you know, developers want to be able to build that applications quickly and test them. Right. And the key trend that we've seen is many more containers being used compared to VMs. So for example, if you think about the size and scale of development environments, they're three to 10 X that of production environments. So containers are going to become much more popular than VMs, but those create challenges for developers because now they have to think about how do I orchestrate the container? For example, Kubernetes is an orchestration engine that requires HCD as a mechanism to manage what's the primary versus the secondary, et cetera. All of that developers shouldn't have to worry about. Or whether Kubernetes is going to be around or not. And there's a new version of this and a new version of that. There's a land grab in the entrepreneurial world around who's going to own the management. It's a nightmare. That's a wonderful point. Absolutely. The technology is immature. So if you bet on a technology, is that technology still going to be around? That's the other thing. And the last thing is, for example, I have my container spread out on a few nodes. The load balance on the front end is saying, hey, I'm getting workload and the servers are utilizing this way and I'm going to send this request here. But that server that it sent the request to didn't have a container on it. And a lot of people are building an orchestration mechanism that is moving the request to some other server, but that's going against the load balances requirements. So scalability becomes an issue. And developers don't want to think about all of this. They just want to be able to use containers and benefit from it. And that's why the next trend is to be able to have containers as a service where developers can just bring the container, not have to worry about orchestration, not have to worry about scheduling, not have to worry about scalability, and it's all taken care of. And that's what we're delivering with our application container cloud capabilities at both the PAS level and the IaaS level. So the developers are becoming more relevant. You guys are obviously investing in that area and amplifying it up, which is really positive. Mark heard, again, chat with me about the use cases. And obviously the applications are coming. You're downloading a ton of apps and everyone's getting overloaded with apps and more apps are coming. So it's not like they're going away anytime soon. It's kind of like the mail. It keeps coming and coming, but more apps are coming. So, but the use cases are shifting from dev test. This is what Mark heard, told me. It's all dev test. Certainly that's the low hanging fruit. But now as workloads are moving to the cloud, there are mission critical applications that are moving to the cloud. And then some workloads have better cloud and on-prem kind of characteristics. By 2020, how does that shifting your thoughts on this and then the role of the developer in all of this? Now, I think there are a few things that are going to happen. One is that line of businesses want to fail. What I mean by that is they want to try 15 different things and they know that only two or three of those will succeed. In the past, they would think and they would think and they would think and they would think before they made a move into a particular area. Now they're saying, I just want to try and I'm perfectly fine with failing. Or what does that require? It requires one, being able to test your things really quickly and making your customers part of your development team. By putting out an application that they're comfortable with not being complete, not being completely feature rich, et cetera, but getting the feedback on what works, what doesn't work. So the number of releases that are going to happen are going to go up significantly and they won't follow the regular release paradigm because these line of business folks are themselves going to become developers. So one is the number of developers is going to increase significantly with the citizen developers. They'll be able to release the products very easily, meaning the minute they're done with making changes on a browser, it's live for customers to use. And then the other thing is developers have to get very comfortable with working in this constraint where they're not building everything perfectly and making sure that they can support the business. So there are other agarwal here inside the CUBE conversation. Final point I'd like you to make is share with the folks what's some of the action and where can they get resources around this developer? For developers watching saying, hey, you know what, I'm really not familiar with Oracle or I'm familiar with Oracle. Obviously Oracle has a lot of influence in the marketplace, huge install base, customer base. So developers want relevance but they also want to have a partner to monetize their work. And this is important. You guys have an opportunity here in my opinion. So what would entice a developer? Where can they go? Where should they go to play around and learn more about some of the things you're doing about being open? Sure. So first of all, developer.oracle.com. Easy to remember. It is going to be a one source for all the information and it's not just going to have Oracle content on it. It's going to have content on DevOps, microservices, containers, API, chatbots, machine learning, AI. And the content might come from Oracle, it might not come from Oracle. So that's one resource. Second is we've created a series of 20 city events called Oracle Code. These are free developer events. So free for developers, focused on those six or seven areas that I just listed like microservices, chatbots, machine learning, et cetera. Coming to 20 different cities, we put out a call for papers in December. We've gotten 1200 submissions for people who want to present at these events. So it's not going to be just Oracle people talking about Oracle content, but it's about what is important to developers. You're not jamming Oracle down their throat. You're embracing a conversation. Yes, and we're helping their education. The other thing that we're doing is we're sponsoring a lot of non-Oracle events. We know we have to go where developers are. So for example, we just completed a developer week up in San Francisco, and they had a ton of developers, and we showed them how they can have API-first development, how they can build microservices behind those APIs, and then how they could leverage chatbots to deliver a rich experience. Tremendous reception, they loved it. The other thing is we've created a advocacy program called Oracle Champions or Oracle Gurus where non-Oracle folks can be talking about Oracle content or topics, and we'll promote them in the environment. And last is the Oracle Code Startup Accelerator. So we want to support startups. We want to give them environments, and we want to give them mentoring and give them a stage to talk about. So the Oracle Startup Accelerator is enabling that. So if the developers want to try this, go to cloud.oracle.com-try it, and you'll get a free account with a certain amount of resources for you to try. So Arthur, thanks so much. Congratulations, and great to see you guys out there pounding the pavement, really engaging the developers where they are and being open. That's awesome. Excellent, glad to be here. Cube Conversation, I'm John Furrier. Thanks for watching.