 All right, this is Jason Porter with the Red Hat Developers Program, here with James Krugland, who runs our IoT initiative here at Red Hat. Yeah, yeah, I'm chief architect for Internet of Things at Red Hat. So work with customers and partners to understand their use cases and requirements. Try to work with the partner ecosystem and also try to work with our business units to help them understand how partners and customers are using our products in the space. Oh, very good. Now, as I understand it, our foray into IoT is a little bit different than a lot of the other companies. A lot of other companies are focused on the consumer aspect, some of the really small devices. Where do we play in? It's really in the industrial and enterprise markets. That's where we have our core, a customer market. And so we really entered the market initially by customer demand. We had customers like large logistics and shipping, rail companies, energy companies and things like that that were saying, hey, we're using JBoss technology or REL or whatever the technology is over here, let's try to leverage it in this use case where we've got it out in the back of a store or we've got it on a train because we know this technology and we trust it. And that was sort of the genesis. And then we began to looking at how can we integrate our products together so it makes them easier to consume. How can we represent these use cases in the products and extend them to add things like adding MQTT support in AMQ, enable them to support things like smart meters and things like that through the broker. So it really was an evolution and then we've been extended that to working in the Eclipse IoT community so that we're working with partners like JBoss and EuroTac, GE, a lot of Zool systems, Samsung, Cloudera, CodeNV, folks like that where we can come together and have individual projects that solve problems. But we've also created testbeds where we integrate these projects together and not even Eclipse projects, but Camel, ActiveMQ, all the upstream projects, OpenShift Origin, so that we can have this community of practice, community of development and really sort of represent use cases there, get the best practices going and really develop the software so that we're integrating the best software to work with the devices in the field, act and ingest the data or act on it and ingest the data and then use it for deeper analytics, route it to enterprise applications, mobile applications, things like CRM and ERP integration, so I think it's, you've got to have both an IT mindset and an operating technology mindset to do this so it really spans beyond the skills of any one company and I think so it's sort of that, it takes a village sort of thing to really represent all the capabilities that are necessary and if you look at Bosch, for example, they've said that they're doing their development in open source because they can't do it all. They need a community of people to work together that it's just, there's too many disciplines to be able to cover it. Oh, that makes sense. So if I'm understanding right, you're not going to be deploying a device out in the field that's going to say run Jbos or rel, but you're going to have rel in the back end and then something that'll talk to the devices that may be pulling in data via ActiveMQ or something else. So you either have a device to cloud model like a Nest thermostat and in that case we would have the cloud technologies that would ingest that data, route it and use it and use it for analytics, present it through applications and then there's other use cases, industrial use cases where you have, may have hundreds of sensors on a train and then you have a gateway that has the cellular connectivity in it that's running rel and Jbos and other applications and it aggregates that data, decides what data to act on to send an alert up to the driver, what data to batch and send back over cellular, what data to wait until you're in Wi-Fi coverage to dump like maybe engine performance data, things like that. So you've really, you've got bandwidth restrictions and cost restrictions on transmitting the data but you've also got regulatory restrictions on if you're a retailer that's working across Europe Germany may have different regulations about data locality than others. So you have to build a flexible software infrastructure to handle that and that's, we help, our thing is really we have those enterprise technologies and we can extend them out to that gateway layer, that aggregation layer out in the field so that you can write IT type applications that interact with the field and the devices and you can make local decisions and make local control and so you know, like I said, back a store when you look at an airplane, an airplane if you go in the hold it's a mini data center down there if you look at the 787s there's probably 10 plus racks worth of electronics down there to run them so in a lot of these use cases you really have enterprise like capabilities that then send their tendrils out and work with the actuators and the sensors. Okay, so the developers that are interested in IoT you know a lot of them are going to be like the booth over here that you know the makers and going through and soldering everything but what we're going to be focusing on is not those devices but you'll have those out in the field and then the developers will be developing stuff for the data center for that gateway area and they're going to be developing that on rel maybe it's a native thing they develop in C, C++ and then also on the back in the data center you're going to have something like data grid or something like that on the Java side or some other framework. AMQ, Fuse, and we've been very successful with Fuse out on the gateways because it really does a good job of aggregating and routing data so Fuse is really integral in a lot of our customers you know we've got a major shipping logistics customer that uses Fuse and AMQ out in the field and at the ingestion layer so I mean it's really providing those services and capabilities as far out into the edge as you can while still maintaining some semblance of IT control over those environments where they're going to run. Okay very good so for the people that are interested in getting into IoT what would you tell them? Yeah I would say you know we've got a lot of resources ourselves you know I started out going and getting a Raspberry Pi and beginning to just understand a lot of the protocols and understanding things. I'd say go to developers.redhat.com slash IoT and go there we've got some code repositories with examples and our test bed is there and available on GitHub. There's also iot.eclipse.org which is where we do a lot of our open source work I'd say go there there's a lot of good projects like Capua, Quora, Hano and different things where you can get involved and learn with that. There's a lot of simple step by step how to get started instructions and you know that's really it is get your hands dirty, get out there and you know start small and learn and understand it and it's better to fail fast and really understand it. I'd say as you get further on the real thing is to begin looking at you know when you start to go beyond prototyping look at the scalability and manageability because those things get left until the end and it's really important to design those into anything from the beginning. I've seen multiple projects failure because they didn't do large scale testing before deployment began for the volumes that were going to come into a cloud environment for example or they you know they didn't understand that they were going to have to update certificates to manage security or roll out patches in a logical way and schedule them so that they don't cause outages. So I think start with prototyping but once you get into real design and implementation you've got to have the security and manageability features built in. Oh definitely, definitely. So those of us following the IoT space you know we're familiar with the idea that the number of devices will eat fairly quickly actually outpace the number of humans that we have on the planet. Where do you see or where should I be focusing on in the IoT space for the next I don't know three to five years? Yeah so we see I mean obviously autonomous driving is really big and I think that's going to change the world over the next 10 years. We actually had something yesterday that yesterday showed AI. Yeah I think in the shorter term window we see a lot of stuff with predictive maintenance. So instead of replacing or repairing assets based on the amount of time and number of hours. You've got monitor, yeah you have a 95% chance that this will fail within a week you need to pull it in and maintain it rather than having it fail in the field. Right. And so your maintenance costs. You know a simple example is at an airport going and cleaning the airport every 100 people that go through rather than every hour because in the middle of the night you don't necessarily need to clean it as often. So I think that's big. I think fleet telematics where you have a lot of mobile mobile trucks, boats, ships, airplanes where you're getting the data off of it that you use for the preventative maintenance and then asset tracking. Whether it's theft and loss prevention in retail or whether it's in logistics integrating that with just-in-time manufacturing there's a real push to know where all of the assets are not just for you but all the way through your supply chain all the way back to the factory. And so those are areas that are really high growth areas that we see a lot of PFC and production implementations coming. Very good. And again if you're interested the places to go are developers.redhat.com slash IOT and then IOT.eclipse.org, right? Yep. Okay, very good. Excellent. That'll do it for us today. Thank you, James. Thank you, it was good to see you. Thanks.