 Live from the Mandalay Convention Center in Las Vegas, Nevada, it's theCUBE at IBM Insight 2014. Here are your hosts, John Furrier and Dave Vellante. Okay, welcome back everyone for day two of live coverage of IBM Insight. This is theCUBE, our flagship program. We go out to the events, extract the civil noise, I'm John Furrier. With my co-host, Dave Vellante, and our next guest is Inji Chosa, vice president of analytics, big data, everything to do with what's going on here at Insight. Looking for some great insight from Inji today, because she's a CUBE alumni, she's been on many times, she's always awesome, great to have you, welcome back. Thank you, John, I'm excited to be here. So you've been doing a lot of presentations, your TED at IBM event in San Francisco is a fantastic keynote, really bringing your A game on the content side here, you're on the keynote yesterday with some really epic announcements, really putting some meat on the bone, and we've had a variety of conversations over the years around what's going on in big data, you had some great examples, really it's accelerating right now, a lot more speed going on in the market. So first, let's bang out some of the announcements, what were the key highlights yesterday's keynote? So we had several actually, so I'm going to kind of riff on them very quickly, first is Watson analytics, if anyone caught the demo, it's amazing, we're bringing predictive capabilities, a very interactive visualization tool, we've got cognitive capabilities so you can apply natural language, and one of the secret sauces inside is what we call data works. How do you actually refine information, meaning how do you take something raw and begin to play with it and massage it a bit and shape it, cleanse it, transform it, mix it, match it, and add security elements to it in order to be able to prepare for the targeted applications, that's actually inherently embedded in the UI experience of Watson analytics. So data works as our new capabilities around refining and shaping information, it's not only available for embeddability in apps, we actually put it on Bluemix as discrete services so app developers can build around that. In addition, we're also going to be launching some new capabilities around probabilistic matching. So we took some of our best algorithms around entity analytics and probabilistic matching and paralyzed that on Hadoop, right, so you could do mass scale analytics of customer insights, matching customer data with social data and other things, that's also going to be available on data works. So data works is our cloud based data refinement capability. The next item we announced was Cloudant local. So many people know Cloudant, our mobile database as a service. Now you can have that and bring it on-prem to kind of the privacy of your data center. So you could have it on the go, you could have it wherever you want. This is something that actually several of our clients were asking for, including one of our top online gaming users for Cloudant said, you know, wouldn't it be great for me to look at all the user data and run analytics on it? So we then created our next announcement, which is called DashDB. It's our really dynamic data warehousing capabilities. It has our in-memory columnar capabilities inside. It is also cloud based. It is inherent in the Cloudant experience. So if you're in the Cloudant UI and you wanted to create an analytic warehouse, you can do that with a couple clicks of the button. In addition, if you're actually a BlueMix user or a data architect or a DBA and you also want to create your own data warehouse, you can go straight onto DashDB or onto BlueMix and create an instance. On top of that, we also announced, and this was through our learnings in terms of dealing with Watson cognitive solutions and applications that when you have a large corpus of data and it's really unstructured content, so it could be documents, files, how do you begin to massage that and curate that information for knowledge workers, data scientists, and so forth. And so Watson Curator was another piece of the announcement. That was a lot. So I have a lot of questions. So you guys got a lot going on under the hood. So let's just connect the dots. So Dave and I were reviewing the Twitter earnings that went out yesterday and today. A lot of debate around Twitter and stocks down, unfortunately, but really what they talked about is in the moment, immersive. A lot of the terms that they're talking about is what you guys are saying here at Insight about engagement, systems of engagement and also engagement. So we heard things like cohort analysis, what Twitter's throwing around as Ray Wang says, these new verbs are out there. So connect the dots between the business outcomes around engagement because what Twitter's doing for their own business is what your customers are doing on the business side. They want to connect to the engagement. They want to connect with things, humans and devices and machines. But there's a lot of stuff going on under the hood. So connect the dots between the value of the technology and what the business owners want with this engagement trend. Absolutely. So when you think about on the business side, cloud data engagement, all of that is just, it creates an opportunity and confluence of capabilities and what's forcing is people to kind of rethink and redefine the way they want to engage. Whether it's in the call center, whether it's in operations or logistics. So think about real-time demand forecasting, right? So based on weather aspects, based on social preferences, based on community conversations, based on quality of service, that could actually impact and you want to be able to do it in the moment. You're absolutely right. We call that real-time context and that's a piece that I've talked about before on theCUBE and being able to marry, I would say, and integrate new types of data. Twitter being a key one, new types of data, integrating that with enterprise data, right? Historically, in the enterprise, you have information like product, asset, customer, transaction and marry that in a way that completely redefines the way you do sense and respond applications for operations, the way you rethink the quality of your customer response, which is not just a reactive man, my cell phone is down, the network's not great, to a much more proactive and predictive understanding of the quality of an engagement is going to be that, those are key elements that we're trying to design and integrate into kind of where we're headed next. So I have to ask you some questions about Watson. People want to know like, how do I buy Watson? So you mentioned data works and you mentioned Watson Curator, you've got Watson Solutions, Watson Analytics. So these are products that I can buy, correct? They are actually, and the majority of those are actually SaaS products. So you can deliver those in the cloud. That's right. So I can buy Watson as a solution, if I'm in healthcare, let's say for example, or I can buy Watson Analytics, which is a development platform. That's right, that's right. And then these two new products, data works and Watson Curator are a SaaS, presumably a subscription-based model. Correct? That's right, that's right. So you can access all of that on the IBM Cloud Marketplace, as well as in the Bluemix environment. So if you're an application developer and you're in Bluemix and you wanted to use data works and subscribe to the service, there's actually a freemium. So it's a freemium model, both for Watson Analytics, data works, for DashDB and CloudIn. And then there's a charging mechanism for per user, for the enterprise or the team. So I can buy an ELA if I want. Yeah, or like a group, kind of, if I use it. But you don't have to, for every individual, you could start with a freemium model. We really want developers and enterprise architects and IT engineers to kind of get started very easily. So if a customer says, I don't want any upfront costs, I don't want any ELA's, I don't want any, you know, penalties, if I shut down, you say no problem. No problems, get started today. And actually Watson Analytics will go live in November. Right now we're in beta. So we have, and it's a closed beta, we're actually onboarding clients regularly. But the live service will start in November. And then for data works, DashDB and CloudIn, you can actually go online today. It's available, the freemiums are available today. And get started. Improbabilistic matching, sounds really fancy. You've heard about this, right? Big match on the do. Tell us what we get with that, right? Big match, yep, no problem. Yeah, so one of the things we've done is, and it's been exciting in the lab. So we're delivering the capabilities around what clients are saying, look, I want the next generation of master data management capabilities. And in this next generation, I want to be able to integrate customer information with social and public information that may be outside the enterprise. One of the things that we did was, we took our unique probabilistic matching capabilities that we had actually developed in the labs and also acquired through our initiate acquisition, I would say five years ago. And then we took that unique set of algorithms, and rather than running it on a relational system, we actually run it directly on Hadoop. That new on-prem offering, it's an on-premise offering, is going to be available in fourth quarter as a standalone. Now, what we're also doing is we took that and are creating a cloud-based SaaS service under DataWorks as a matching service for big data as well. So we have two. That's also running on Hadoop? And that's also running on Hadoop. And it's some kind of key value store? And it has, yeah, there is a way to understand what sort of the key values are around master data sets. The other thing we've also enabled is our enterprise Hadoop as a service in the cloud. That's available also now in fourth quarter. And then we enabled our streaming analytics geospatial service. And we also just announced our partnership with Esri, which is great. And that's with your Hadoop distro? Yeah, yeah. So Jeff Kelly last week at Big Data NYC was saying, eh, you know, with deference to IBM, you know, the big three Hadoop distros, not true? No, not true. It depends on what you want to do. I mean, you don't really care. Well, no, what I would say, what I care most about is clients actually getting the value and the outcome at the end of the day and the faster you can actually apply the insights in analytics, you get to the outcome faster, who cares about just putting data into it. If you got a solution that's all IBM, you can get there faster, is that right? That's right, that's right. So let's talk developers, because obviously you have a developer aspect to this and with data works and also where again, this is the mega trend. Developers want to integrate natively into their mobile applications, the insights. So you have the freemium model, which is really compatible with how customers are buying and rolling out the cloud like technologies. A lot of stuff under the hood that's shipping and available, you guys bring that goodness. Developers, what do you guys, how do you view them? How do you make it easy? Again, there's a focus on speed now this year. So people want to get rolling fast. How do they stand up some of these capabilities and how do they integrate them into the application? Oh, it's amazing. Within five minutes on Bluemix, you can actually get quickly started. You actually pick a runtime environment. You pick a set of services that you want to run and then you can quickly get started. So Bluemix is really agile. One of the key elements that we wanted to enable for clients was, and especially developers, is look, you don't want to have to wait for a provision system or server. You don't want to have to wait for sometimes the financial aspects to get started. So Bluemix is our cloud-based open, kind of built-on cloud foundry, quite honestly, open-based architecture that allows developers to completely easily compose apps. It's our platform as a service. There's a number of data management services, application development services, there's a list of run times you can pick from, and there's even what we have a section called boilerplate, which are templates for you to build other types of applications, like the Internet of Things type application. So they're quick start modules. It's quite easy. You can go straight from developing to actually running in production. So Ray and I were, Dave, we're talking about how developers have a lot of going on in-house in the enterprises. So that's awesome, so that's good stuff there. I want to spend more time and drill down on that, but I want to get more to the customer side, the decision makers, the folks you're talking to that you're engaging with in your interactions. What are they caring about this year? What's your thoughts? Can you summarize the totality of all your customer meetings into a couple minutes? Because people want to know what to do, what sequence, how do I get rolling? How do I move it into production? What are some of the low-hanging fruit opportunities? Well, that's a great question. So in addition to what we have on Bluemix, which is really targeted to developers and IT folks, we actually have a set of SaaS-based offerings that are more targeted to line of business. So everything from pricing optimization to capabilities around digital analytics and customer service, those are capabilities part of IBM's overall experience. One is smarter commerce portfolios. Those are many of our capabilities on SaaS. In addition, Watson Analytics is a great user tool based on the individual user experience where business leaders and individuals are saying, you know what, I want to be able to play with the data. I want to be able to access it. I want to be able to understand it. I want to be able to massage it differently without necessarily having to call predictive scientists that's going to write up these models and IT engineer that provisions the data from the data sources. I want to have a set of libraries that show me what my data assets are. I can shop for them very quickly. I can look at kind of different types of visual techniques. By the way, I can interact in natural language. So those are big pieces, especially from an individual role level. From the aggregate enterprise, I would say the top three investment areas is one around improving and enhancing the customer experience. The second is around more specifically optimizing operations and assets. And then the last is how do you manage risk security and governance just in general because of the threat of security intelligence or threats around data and data protection or even fraudulent activities around claims? I mean, it's such a broad space, but the risk aspects of all those elements. So explain context computing. We've got the top three, that's consistent with Steve Mills. So you guys are all on the same page here. It's good to see you. He called it CAMS, which was an acronym for Cloud Analytics and Social Security. So talk about the customer environments there with context computing. What does that mean? Cognitive and Cognix? Oh, so context is, so we talk about real-time context computing. So there are three core elements of those. One is at a technology platform level, context is understanding relationships across different entity classes. This is when you hear Jeff Jonas talk about sense-making and G2. That's a core part of an area of investment within IBM and a core part of the area that I'm heavily focused on. Another piece of the core technology platform is our stream computing capabilities. The ability to understand data as it's coming in. And one piece at a time, not just in terms of large batch. It's operating in real-time is a very new mindset, but also requires completely new computing kind of capabilities to do that. The third area is what I consider much more around decisioning and actioning. So you take things like complex event processing, you think about workflows, you think about higher orders of rules, but you want to marry that with some of the new capabilities around machine learning. So you want to marry all of that into a platform that allows clients in terms of the value what you talked about at the beginning around real-time insight, in the moment decisioning, in the moment action. And when we talk about real-time context, it's about that. In that moment, when you have a narrow window of time and you want to be able to understand kind of the low signals and sense and respond very quickly, those are kind of the next class of applications that we're working on. Context is that, which by the way, I think it's an amazing vision. I think you're right in the money there. We're in the same religion there. So context sets the table. On top of that, it's the cognitive. That's the reasoning piece. Is that right? Well, I'm sorry. They just didn't get a question. No, I'm just trying to respect your time. I know you have a hard stop. So cognitive also includes. We're keeping it here all day. Can't go. We can't leave. We love it. Okay, so cognitive includes not only those things, but it's also the human elements, right? It's the machine learning aspects. It's the fact that systems can learn the way we learn. And, but you get to also interact much more in language, natural language. You also get to apply capabilities where you can perceive and infer and reason. And those are higher order type analytics. It's not just, you know, traditional forms of statistics or predictive analytics. Dave, go ahead. Get a question in. I'm happy if you got time and can we go? Okay, last question. All right, so I love when you guys talk about business outcomes. It's the smart thing to do, the right thing to do. Having said that, we're hearing from a lot of customers that they're sort of looking at reduction on investment and they're maybe taking a dollar that we'd say would spend on a traditional data infrastructure and they're spending 30 cents on the new one because their storage is cheaper, their infrastructure is cheaper. Having said that, it seems to be elastic. They're spending more on less for each individual component, but they're spending a lot more on stuff. Do you see that sort of trade off going on? I do. I see some. Doing more with the new? I see doing more with the new. And part of it is about speed, right? Every enterprise, and when I talk to clients, one of the key comments is that look, your competition isn't your competition in terms of who you think they are. Your competition is the last best experience that your consumer actually had. That's the bar. It doesn't matter what industry they experience it in. It doesn't even matter if they experience it in their home, right? That's the new bar. So if that's the bar, then you have to quickly figure out how do you redesign and reset that bar to say, I've got to meet that minimum expectation and the way I engage in my call center, the way I engage in my mobile applications, the way I engage in terms of my supply chain. And so for clients, speed right now trumps so many other things, and they don't want to be hindered by maybe the traditional ways in which they access information or the traditional investments that they've made. So I definitely see a shift in terms of a lot of money moving into systems of engagement type applications. Yeah. And that leads to faster business outcomes. Absolutely, absolutely. Awesome. Inhe, great to have you. Just bumper sticker the show here for the folks out there watching. What's the main theme here? If you could summarize it kind of on the bumper sticker of the car, leaving Vegas. Oh, it's all data for all people. You can put data in more places, hybrid, self-service, insight, and just outcome. We're really are focused on empowering everyone in every role to become not only data experts, but users and kind of unleash the value of data here. Okay, this is theCUBE. We are here live in Las Vegas for IBM Insight. We are special presentation inside the new digital experience, second screen, virtual conference going on here in the social media lounge. It's called Insight Go. We're proud to be part of that program. Great social media, a lot of influences here. And of course we love broadcasting that data from the noise, the signal in here. Thank you so much. Great to see you. I know you're super busy. Really appreciate your time. This is theCUBE. We'll be right back here live in Las Vegas at IBM Insight. I'm John Furrier with Dave Vellante. We'll be right back.