 Live from Washington D.C., it's theCUBE. Covering Inforum DC 2018, brought to you by Infor. We are back here at Inforum 2018 in Washington D.C., John Walls of Dave Vellante. We are in the nation's capital and joined right now by Massimo Capocha, who's SVP of InfoOS and Rick Ryder, product director for Coleman at Inforum. Gentlemen, thanks for joining us. Good to see you both. Thank you. Let's talk first off, good job, by the way. Rick, on the keynote stage this morning, you had some time to shine out there. Your thoughts about the show in general so far, we've been a couple of days in now. How's it going for you? Yeah, very, very well. The customers have received the Inforum OS and the technology innovation and what we do with AI. Very, very well. Lot of people in the hub, lots of sessions, so lots of interest on the technology innovation for InforOS and for Infor as well. Rick, for you. Yeah, it's been great. It's been interesting. What we're finding out is getting a lot of this out in front of customers and partners is bringing up some interesting opportunities for us moving forward. So it's not every day we get the opportunity to get in front of these many people within our network, so it's been great. So what are you hearing from folks? When you start talking about AI, especially those who maybe don't know, haven't embraced it yet, what are the hesitations, reservations? I mean, what are you hearing from them as far as what's gonna trigger them to make a decision? Yeah, to be honest, I think they've been hesitant in the past just because it hasn't really been clear. When we've talked about AI in the technology community, it's been hard to define. Some people might, in fact, define it incorrectly because we're making assumptions about what technology can and can't do. I think what we're uncovering, I feel we've got a pretty unique approach to what we're doing here with InforOS and Coleman connected to it. We're working directly with customers to identify use cases on how we can apply AI rather than just starting at the top and saying, hey, we should be doing all these great things and let's see how we can make it work for our customers. It's kind of, we're flipping the script and starting backwards and saying, hey, what are the issues, what are the opportunities that customers have? How can we build the technology using AI to make it meaningful? So we have business impact day one. And by doing that, I think it's a lot more understandable, it's a lot more relatable, it's a lot more trustable from our customers. We, certainly in theCUBE here, watched and observed the ascendancy of the hype around so-called big data, and which is sort of moderated now. But data is plentiful, insights aren't. And so we've sort of come to the conclusion that the innovation recipe, if you will, for the next decade or so is data, applying machine intelligence to that data and having cloud to be able to scale it, having cloud economics be able to track innovation. You guys seem to have all three of those pieces, right? But AI without the data is just, I don't know what it is, right? Data without the ability to extract insights. What good is it? And you got to have cloud to scale it. Your thoughts on, from a platform perspective, what that means. Yeah, absolutely. So I was seeing the interview that you were doing with Charles is we build out this platform from the beginning. And one of the big element is that we have made possible to synchronize in real time all this data that the applications generate into a single place called the Data Lake. So when you have the data in Data Lake, then you can do many, many things. And not only analytics and reporting, which is the classical use case, but now it allows you to do AI. And the difference is that we don't have one domain of the data. So some of the vendors have only CRM data or HCM data or financial data. With info, we have all different domains of data. So we can go from HCM, from financials, to asset management, to IoT readings of IoT devices, to ERP, and CRM also. So when you combine, when you can cross and combine the relationship with this data, then your AI is much smarter, intelligent. When you have only the AI focused on a domain, it's less intelligent. So that's actually the power that we do. And our Coleman will take advantage of that rich Data Lake. And we talked a lot to Soma earlier about the stack. And at the bottom layer is the OS. So everybody's familiar with what an operating system does in computer science. How is your OS similar and different? What's the function that it does, if we could double click on that? Yeah, so it's in for operating service and we call it as service, because it's actually not in the database and operating system level. So we believe we are more in the application technology. We are the layer that takes the bare technology and makes it usable for a business, for an enterprise, and we build applications on top of it. So what we believe at INFO, when you have an architecture with us, it is composite of a suite of application or even the new Microsoft architecture that developers built, you still have to deliver a uniform user experience, a uniform business process, uniform security and data management and even AI. So if you look at services like Facebook or Netflix, they have maybe entire microservice architecture, a thousand of that, but the experience is one. That's we want to bring it to the enterprise. The INFO OS brings that unified experience, both from the end user and business process to the enterprises. And we do it for all the cloud suites. The INFO OS is all the cloud suites, not just one, but all of them, the same services. So I love the Netflix example because if you think about digital, digital transformation, digital business, my experience with Netflix is just with Netflix. I don't have a, there's no marketing department, sales department, service department. I just have a problem. I go to Netflix, I go to my app. I interact with the app. Yeah, absolutely. So that's, I consider that, let's call it a product. So Rick, how does this capability get translated into product? Yeah, you know, one thing that you brought up a lot earlier is with all this interconnectivity and how we have to package things. So we've got all these different services that OS offers. So we've got the data lake. We've got the API gateway. We've got the integration platform and ION. All those pieces is what bring this together to where we can actually deliver something to our customers. In my case, it's an AI model or it's RPA because of all these things are packaged together. So they don't actually see what's happening because it's already packaged for them. Okay, so what I was saying to Charles, you probably, you might have seen it, is when we first discovered in for it, it was like, huh, what do you guys do? And it wasn't clear exactly what you guys were doing. But he said, and I believe him, it was always our vision to have a platform. Now that the, it's not opaque anymore. The platform is pretty clear. Now you've added the burst analytics. You've added Coleman AI on top of that. So, you know, Andy Jassy at AWS always talks about the flywheel effect. So I suspect that you're entering this flywheel phase. What is that phase? What does it kind of mean for you guys, for customers in terms of innovation? Yeah, it's a very good question. Actually, I worked for years with SOMA. We started with this platform, this journey, and with Charles. And we start really with, okay, what's the first issue? You know, we want to solve the integration problem. So we want to give it integration platform. Then we build that. Then we start to say, okay, we want to unify the experience. We build an unified portal with a single sign-on. Then we say, okay, we want to unify the data. We build a data lake. So we continue to build out the platform. We are now at a level, we have a platform. And it's a unique platform because you can say it fits in one magic quadrant because, yeah, it does the iPass and the pass. So all these magic quadrants, but it doesn't fit in one. It fits in all of them, right? So, and the analyst looks at that and say, okay, we have a platform. It doesn't fit in one. It fits in all of them, right? The magic quadrant is now becoming outdated because the cloud is, like you said, it can need 15 stovepipes. Exactly. The stovepipe thinking is the magic quadrant. Exactly. So with all due respect to my friends at Gartner. But the flywheel is, yeah, the platform is going to become more and more important, relevant. The customers that are in the cloud or not, are not in the cloud, they will use the platform to get to the cloud. It's going to be in a new enabler for those customers that are still on premises to go to the cloud. And the InfoOS is enabled for a hybrid process. So some application can be on premises or in the cloud. With the OS, they can take the journey and get to the cloud at their own place. Let me show you, I understand that. So architecturally, you don't care. We don't care what the applications are. Okay, but you've certainly done a lot of work to optimize AWS. We're an AWS customer and we know it's not trivial. You have to, you know, you've got to work it. It's simple, developers love it, but to really take advantage of it, integrate it with your processes, it'll take some work. But architecturally, you don't care. But that's not an offering statement, is it? I mean, today, can I run that multi-cloud, run your software anywhere? Are people doing that? Well, today we have a mix of, we use open source library, but we do use, utilize AWS. The data lake is built on S3. On AI, we use Lex or SageMaker for the training of the models. So we do a lot of AWS because it gives you our computing power and it's out of the box solution for certain pieces. What we do, we build interfaces to our applications. So that our customers doesn't need to take care of all the plumbing, it's all interconnected and done. So that's one of the power of InfoRaisers, brings that application technology layer between the business application and the basic technologies. And the customer doesn't even want to think about the plumbing these days, right? To most customers, infrastructure is irrelevant. You know, again, apologies to my hardware friends, but they don't care about hardware, right? I mean, it's interesting, Charles said in the keynote yesterday, when we were an on-prem software company, we didn't manage servers for our customers. Customers didn't care really about the server, any more than they care about the plumbing today, right? Right, yeah, and if I want to relate that to the AI space, all the training, all the science, all the highly computational things that we have to do, customers don't really want to know what that means or how to use that. So what we're actually doing is in conjunction with some of the AI services we're working with, with AWS, is we've built a modeling platform to where they're operating in one location. They've got no concept of where this is hosted, what's going on behind the scenes. And then when we connect it, we expose an API and they can do any sort of RPA that they want to. Yeah, so I mean, you're talking about, when you talk about your customers and they don't care about what's behind the curtain, they just want to handle maybe something up front. But yet, you have to understand what they can do, right? Or you have to understand their potential. So how do you do that when you're dealing with different companies, different sizes, different priorities, different challenges, they're in different technology stages. How do you all address them individually and help them get to that better place? You want to take that on the modeler? Yeah, I think it's never a one-size-fits-all, right? So we try to give them what we've called citizen developer tool sets in the past and I've even started to try to say citizen data science tool sets. So how can we make it more consumable by all types of users? So yes, we can provide templates, we can create these things that might work somewhat out of the box, but each one of these customers, their data is just slightly different, they need to make tweaks. So we really want to be able to provide all that flexibility and it gets back to we start with their use cases and then we build from there. So we get all that feedback and make sure we're hitting those key points. So I want to pick up on something you said about citizen data scientists. I've used that term before in front of data scientists. Some of them don't like it, right? Because I feel like it denigrates what they do. And it's true, a data scientist is a math whiz, maybe a stats whiz, they're a data hacker, they can code, you know, okay. And that's not every business person, clearly. However, when you think about things like RPA, I mean, you really want to enable business users. You don't want to repeat the same problem that we had for years with things like decision support where you had two people in the company that knew how to build a cube. And you had to have line up with and ask, please, can you build my cube? I have a deadline, well, everybody else does too. It just, it wasn't effective. So things like RPA and low code citizen data scientists spread that technology throughout. Now, part of that is having a platform that is, I vision a studio, whereas a user, I can actually create some kind of process and code that in software, you know, code it, something that's repetitive that I don't have to do every day. I do it every day, I do it the same way. Somebody gave the example, it might've been SOMA, I remember somebody else, expense report approval. Yeah, yeah. I've never not approved an expense report. And I don't crack them open, look at it, maybe every now and then somebody does, somebody does, by the way. So don't get any ideas here. I always push the approval button, right? Why couldn't, why can't a robot do that? And look for anomalies and say, oh, a $300 scotch, that's a lot of the ordinary. Yeah, yeah, absolutely. So is that a capability that you're working on, that you have today, that what I'm envisioning a studio, and then I imagine there's gotta be some orchestrator? Yeah, so if you look at throughout all of us, it's completely model driven. So either you build an integration or a workflow, or an AI model, or even we have a platform as a service, Mongo's where you build with low code applications. So you can take it to end to end where you train models in AI, you expose as an API, you can build your own app on top of it with low code, and then give it to your business users. Very, very simple and in the cloud, in the browser, and you can do, every customer can do it. So that's very important for us. We work from the beginning with this model to give the tools to everybody, not to only an elite of people that can do and then there is the rest of the people that cannot do it. Every new computer science engineer that comes out gets AI out of the box. When I did computer science, yeah, I got some AI, but it was not really like today. So everybody can program AI now, and we want to give these tools to every developer, and not just one to an elite. Yeah, and the workflow prediction model that you've been talking about, if you want to come join us down there, we've actually got a model that we're working on for that exact use case right now. Oh, cool. Yes. So yeah, giving the ability for those business users, as you say, to, it's almost like lowering the barrier to entry to a lot of this AI technology. It's not devaluing or anything data science because we've got those advanced tool sets to where if you want to do something in our studio, bring it over into the Coleman AI platform, you certainly can, we're not devaluing that. But, you know what, if we want to start and take little bites off and you want to give this in the hands of the business users, we've got a great solution for that. So this is all the cool stuff. This is the stuff that business users care about. I mean, do they, by itself, my question is, do people care about what's under the covers? I mean, are they asking you, well, what's in the database and how does this work? How does that work? Or do they just really want to focus on that functionality that they're getting in the business impact? Yeah, with the advent of the cloud, people just, those questions like, which are operating system database, which technology use, it just went away, right? So people just want to know the functionality and the value. You know, maybe there are companies that have more, you know, and IT architects and they want to know more. You know, that's what they want to go down into the details, then you go into the architecture of the OS, of the application, or we integrate with AWS. So we do that as well. We, you know, we talk to customers about that. But most of them, they just want to know, okay, how can I use this platform to make my business better, right? So it runs the cloud suite, but now I can connect to other cloud services. I can connect to the other application. I can build my own app and bring it in. So they want that business value immediately. And that's why we build this in for OS so that they can run the cloud suite and add business value. Yeah, you guys at last year's analyst meeting gave a little glimpse of some of the architecture and it was very useful. Actually, analysts love that kind of stuff. I didn't get the invite this year. Maybe it's some of the smarmy questions I asked. But I found that actually quite impressive in terms of the tech behind it and the R&D that you guys are doing there. But ultimately it comes down to what products you can build and what business impact it has, right? Yeah, absolutely. And I think where we're heading with this, we really don't have many limitations for what we're seeing right now. We're built in a way to where you can apply to every single industry, every single cloud suite. We have the unique possibility to where we can go through all these different industries and create these sort of values. So we've got a very unique future ahead of us. So how much better? Or can you give us an idea of the roadmap a little bit about where you think Coleman can go? Yeah, so we're starting to play in the image recognition space a little bit. Maybe looking at how we can utilize things like drone technology and do inspection reports, those sort of things. It's maybe, in at least my opinion, I think others kind of express the same. It's maybe the least developed area and we want to make sure we have something that works for customers the way that they're going to see value immediately. But also we're starting to look at edge AI. So how can, not necessarily just an IoT, but how can we build something in the cloud? How can we create a model? Then deploy that for our on-prem customers who aren't quite ready so that they can get that AI experience as well and that predictive insight. It's Dave Vellante at SiliconAngle.com. Is that right, your email? For the invitation. David Dot Vellante. David Dot Vellante, let's make sure. So we'll exchange information later. We'll invite you. I'm sure this is not your territory. All right, Joe. It's on me. Thanks for joining us. Thank you. It's been a pleasure. Thank you for the time, we appreciate that. Back with more here from Washington, DC. Right after this, you're watching theCUBE.