 So, I was supposed to be talking here with Kate and Chris. Chris has been indeterminately delayed because of weather, and Kate has the plague from the last two conferences she was at. So, there we go. And then there was one. A little background on Kate, Chris and myself to help set up the talk because the combination would have made more sense with the title. I've been focusing on the edge of IoT, specifically the gateway part of the edge of IoT. Kate has been leading Zephyr among the many other things she does at the Linux Foundation. So she's really working on the tiny little bits at the very far edge of the IoT. And Chris is with Cloud Native and he is the cloud side of the story. And so really, this idea of all things great and small is to put in place this story of where we're heading with the Internet of Things. And I'm sticking up a bunch of marketing slides if you search this amazingly good set. And I'm sorry, these are small and mostly unreadable. But the net net here is these are the pictures that are starting to be drawn around IoT and deployment now. And I've tracked these for seven or eight years. And I have a whole collection of these and the style of the pictures of how this all works has really changed over that time. The concepts and the understanding of it have really changed. And now we're seeing this rich picture that kind of looks like this where things are going all the way up and talking to the cloud. But they're going a little beyond that and hopefully in this talk we'll get to that. I'm kind of preaching to Dwyer here so I won't spend a lot of time on this. But there's really a perfect storm of things going on. Sensors getting really cheap, compute getting really cheap, storage getting really cheap. All of this great connectivity happening. Lots of different protocols out there now. And I think really importantly and to the talk yesterday, manufacturing. And the supply chain is really magic. The supply chain is magic because you can design a board now for a project. Push a button and three days later have the board come. And at the same time, three days later have the plastic case come from somebody else who did the plastic case for you. And so the part of building the IoT components is getting really, really good and really, really easy to do. We're also seeing, and again, as yesterday's talk was unplanned from the keynote yesterday, but hardware function moving into software. We're seeing lots of that. And I'm smiling masmos down here. Arduino helped a lot of people understand that pattern, especially really early on in this, that you could do a lot in software to make things far more flexible than they seemed. And so we're starting to see a lot of functionality move into software. So this supply chain of components plus software is really making the deployment of IoT real now. And I'm seeing lots of it in the field. I'm seeing lots of it from a wide range of companies. This is where Chris Anacheck jumps in. Hi, I'm Chris. The cloud is eating the world. I don't even have to jump into it too far. I think this one is really an awesome talk about Kubernetes and the impact of cloud and these cloud models that are emerging. And really, I shouldn't say emerging. They've emerged, they are the norm now. The Kubernetes supported by top ten cloud providers in the world. And it's one of the fastest growing projects. Why is that? Because it solves a really important problem. It makes it easy to build and deploy this stuff. Because we have all of these little things and because we have this cloud, the amount of data is rising. Anthony talked a lot about this yesterday, which is leading to opportunity and challenges. And we're starting to see this, again, like Anthony talked about yesterday, this web scale coming to the masses. Anthony talked about some services. There's a variety of services now that if you've done your little board design and you've had your plastic made, you can pretty quickly get the rest of that story together. With a few notable exceptions, it's all not that hard. At the same time, we have this trend going on that I think we're going to hear from the other keynote speakers today, and you're all here at Linux Foundation event, around close to open ecosystems. And I was really careful with the point here. And this is something Chris talks really eloquently about, that open ecosystems, not just open source software, but open ecosystems and the recursive nature of those open ecosystems to power things. Since I'm the edge guy, now I'm going to talk about the edge and really jump into the part that I think is interesting. The edge is real. Four years ago when we started this event, we were talking about the edge. And we had some examples, but we didn't have a lot of them. Now every single product company I talked to that's building a product has a very competent edge story to go along with that product. Sometimes it's in the device, sometimes it's around the device with gateways. But the edge story is getting really powerful. And the scale of it, I think this one's interesting because it was 2014. And it was kind of a look. If you look at the numbers from 2017, that's effectively 100%. The products have an edge strategy. They've encompassed edge. And the scale of dollars in edge computing is really starting to grow to the point of the cloud side. And why is this? Because when you can build a lot of little things and you can make them cheap, you really can't have them all talking to the cloud because the numbers get really big. This is my favorite stat from last year. The number of connected things exceeded the number of connected phones. This is actually a really big milestone. So now we have 2X the number of connected things on the planet if you had the phones and the devices. And this goes to the story of really the scale of it growing up. In this space, we've seen, and I was just talking to some folks here who've been doing this for a long time in the connected world. And I thought I didn't even think about this slide when we were talking earlier. But the M2M world has been doing this for a long time. The M2M world has spent a lot of time because they have big money items, building and connecting things, collecting data, processing that data, and transforming their businesses. But what's really interesting, and the edge data and the number of connected things shows it, is the size of the market that's growing to go along with that. The M2M folks set the stage. But when you have a $5 million machine, you can spend a lot of money to connect the $5 million machine to manage it. But when you have a $2 or $3 sensor, you really can't spend the same money or take the same actions to make that connectivity. But you want the same effect. And this slide is really starting to talk to where people are starting to see the dollars in it. Two years ago, three years ago, these slides just had big random numbers on them. They said that number is going to be a trillion billion million dollars. Now these numbers are starting to get ranges, and they're starting to tighten in. Because people are starting to figure out exactly where this all fits together. There's a phrase I love tracking these digital transformation slides. These have been around for about 20 years. They track every single digital trend possible. There was digital transformation for client server, and then digital transformation for web, and then digital transformation for mobile. And now it's about IoT. But the important part here is when these new technologies come in and to the opening, when they get inexpensive and they get widely deployed, they appear in all kinds of places. And they affect the way businesses work. This is a company of a Linux Foundation member, one of their customers. And this is a really interesting company. This is the trend that I have been tracking the most, and the piece of IoT and Edge that I find so fascinating. ATF, Australian Temporary Fencing Company, they literally sell fencing to job sites. They generally don't look as pretty as that nice green field, but they sell job site protection. This is in one year. Their business is now going to have as much revenue from this job site surveillance technology that they developed last year. So in 2018, they'll make as much money from the job site security technology as they will from all of their physical world fencing business. They completely transform what their business is. They already had a relationship with the customer. They already had a connection to the field. And now they're just delivering a new product. It's got an Edge computer in it. It does facial recognition. It does image recognition. It's all kinds of really great stuff. And because they're put in really terrible places all over Australia that are way at the edge of the connectivity network, it uses SIGFOX, Lora technology, to send little messages to say that there's a problem. So they have actually pushed all of the smarts to the Edge in a mostly disconnected world. Another example is Stern. If you use a bathroom tap here and you put your hands underneath it, it comes on. What's one of the big challenges of being in a place like this with bathrooms? There's a lot of them. And some of them are going to be used today. And some of them aren't going to be used today. And some of them are going to be used a lot, like when this session's over. And other ones hidden around the building, around the corner, are going to be used a lot less. But if you're a regular hotel, restaurant, airport, you have to check those bathrooms over and over again. Every 45 minutes, you send your custodial staff to them. Stern connected up these faucets. And now a big part of their business and the most growing part of their business is analytics reports for the janitorial staff to tell them which bathrooms they have to attend to and when, and which ones they can choose to ignore. It also predicts things like paper towel usage and soap usage so that you're not wasting time opening up the dispensers because they've built these really big models. These examples really fit into this. You couldn't do this before. Caterpillar could do it. But you couldn't do this before. But now Stern, in what is a very inexpensive tap that there's hundreds of in a building, a facility like this, can now do that, put informatics around it, have an edge computer that does all the calculations and then pull that data up to the building. Another example of this transformation that I love, the technology has now gotten to the point where it works. At the beginning, I showed Chris's slides and talked a little bit about the cloud picture. What we're seeing, and Patricia just jumped over, which is great too. Patricia and I have talked a bunch about this. We're seeing this model. We know how to build stuff on the cloud side. And we're now getting to the point where we need to build stuff on the edge. We have real edge computers. Dell's got a really beautiful $300 industrial edge computer. There's smaller ones if you want to use routers. There's bigger ones. So a wide range of them. Opto22 is upstairs with industrial edge computers. We know how to put computers at the edge. And now we're starting to build the models around what to do with those computers and how to make those little sensors and those little computers work. And some of the pioneers like ATF or Stern are starting to set the models. But as they're developing and as you talk to them, they're asking for it to get a lot easier. They're asking for the same things their developers see on the website to happen on the edge side, to happen in the world of things, especially in these enterprise and industrial use cases. And so what we're seeing is the same model for the cloud being flipped 100% over onto the edge side. The scale of it is slightly different. But the model is still the same. I have servers and containers and distribution. And I have load balancing. And I need all of the networking things at the edge that I do on the cloud side and then in between them. And so we're seeing this mirroring of models. At the same time we're seeing that, we're seeing this big challenge around security. Because this number is increasing. And as this number is increasing, I put the Mariah hack up here. As the number increases, companies like Stern and ATF actually start to really worry about, I have faucet taps that are in bathrooms. And now I have to worry about, is my faucet tap going to be hacked? And if it's going to be hacked, is it going to cause a vector into the hotel? Or is it just going to flood the bathroom or something else? It seems like a trivial problem. But if you're a faucet company, you need all of this security. And so in my job of being out there in the field talking to lots of these companies deploying products, they've asked over again for help putting this out there. And so inside of the Linux Foundation, we've started to work really hard to curate an architecture, an emerging open source architecture that puts together a set of pieces to make this a lot easier. It isn't all of the pieces. And there's a lot more to go on here. I left some blue boxes with dot, dot, dots because we have some announcements coming up in the very near future on some additional projects that fit in here. But on the left side of the screen, we have Zephyr for building tiny devices. And we have Yachto for building bigger devices. Plur is Tracy. Tracy, big plug for Yachto. EdgeX Foundry for building Edge computers. I have a lot of members here in the audience of the project. Linux Foundation networking for connecting all of that edge magic to your cloud magic in a safe and secure way. And then on the cloud side, besides Kubernetes and all the goodness out of CNCF, we have Acumos, an AI project, and R, and Waze to start processing that data on the cloud side. And we're going to see that migrate over to the edge side too. So Zephyr, EdgeX, Yachto. Linux Foundation networking, Kubernetes, and Cloud Native. I just popped through them all because you can go and see them all on the LF website. But I think this idea of curating a set of technologies on this side, like CNCF has done really well on the cloud side, is really important to the emergence of IoT as a real thing. Because if we don't do this, we can't deploy it. And the customers can't deploy it. And you talk to people. Four years ago, I remember Landis stood up on stage and said, there's only 300,000 embedded software developers in the world how we're going to make this work. We're going to make this work by scaling, by giving people patterns. There weren't a lot of web developers at the beginning either, and we scaled up that model pretty quickly. It's all starting to come together and work really well. But the thing that I think we need to be really careful of, and I ask from everybody in the room who's working on projects, is you're help pushing these projects along to bring them to the point where they can deliver this value. We know how the cloud works. It works really well. We have a really good model there. And the IoT, it's still really emergent. And we're still trying really hard to get that model to get to the point of having an architecture that's robust enough to look like the diagrams that everybody's drawing today. So I'm going to stop there. I'd love to talk to anybody about this and what their interests are afterwards. I'll be hanging around.