 And these things have gone through phases. The first examples were like this one, mechanical or electromechanicals. Layer models used analog electronics moving on to digital. And then finally, the most recent models out there today are fully connected to the internet. Connectivity lets the controller operate based on the current weather and forecasts. As an engineer, I once designed systems for manufacturing industrial control and pipelines. And these moved through the same generational advances as home thermostats, ending with what people think are modern implementations based on digital electronics. And people put in these devices as the generations changed because each generation offered real advantages in terms of performance, resulting in better safety, efficiency, and capabilities. My point is that what is viewed as modern changes every 10 or 20 years as tech and new ideas surface. Things like AI and machine learning now present a new opportunity. Examples in the sensor space include occupancy sensing that not only detects somebody entered a room but identifies who that person is and what the role is. Industrial automation that can not only tell that a pump is running, but predict impending failures before they're allowed to happen. We've got talks coming up this morning on machine learning and AI at edge, and I think you'll find them really interesting. We also have people talking about entertainment applications and next generation health care delivered out to thousands of edge locations. We have talks about ways to handle the challenges of connecting and managing huge numbers of devices and software installations at edge. Also, there are opportunities related to new types of edge devices that improve our environment. But I think the bigger payoff here relates to just doing existing stuff in a better way. Suppose we could improve the efficiency of manufacturing or transportation operations by just 2%. These kinds of advances improve quality of environment, standard of living, and quality of life. I came across this recently. If you process data intensive workloads, such as image recognition in a public cloud, the downside could go beyond just latency and added points of failure. It might bring with it wasteful energy burn. And even given that you're adding extra hardware in these long-term compute supply chains contribute to long-term e-waste, running pipelines of 4K 30 frame a second video all the way up to public cloud might be wrong on many levels. The elephant in the room here relates to the original design of Kubernetes. It was designed for large data centers, achieving efficiency and availability by running workloads on centrally pooled hosts with good network interconnects and lots of interchangeable compute hosts. Edge is actually decentralized, often with IO and location constraints causing one-to-one relationships with the workload in a particular cluster host that that has to run on to be useful. People have come up with Kubernetes projects to make Kubernetes better suited for edge. And we have a talk on kube edge today related to this. But there've always been questions about how to map Kubernetes to edge, whole clusters at edge, worker nodes only, CRDs, controlling edge workloads. Some are asking if Edge might actually be a new type of cloud calling for a completely new type of orchestrator. But it isn't just the orchestrator. If you think people talking about throwing out Kubernetes for a new orchestrator at edge, hold my beer. Heretics are now actually challenging Docker, including this guy who founded Docker. There's a relatively new form of container called Web Assembly that has some very interesting characteristics related to edge. For example, smaller size when the packaging is moved on the network, significantly lower startup latency, cross CPU compatibility, lower resource demand for the runtime. But Web Assembly also brings with it some limitations. So first, they challenge Kubernetes itself, now Docker. Next thing you know, they're going to be alleging maybe the world isn't flat. We have a talk on this Web Assembly thing later in the afternoon. And I really would like you to be there as we dig into the suitability of Web Assembly at Edge, because this could be the next big thing. So I promised you, I tell you, the real theme of today's event. And that theme is community, sharing ideas, sharing things we build together. And in that theme of community, let me start with a community anti-pattern that plays into Edge. I think we're all aware of recent world events that are causing high levels of mistrust across certain borders. Governments have been banning black box, commercial, closed source products from critical infrastructure. And this trend continues or is strengthening. There have been localized shutdowns of cloud-hosted services. The world maybe could benefit from Edge solutions built and maintained as shared efforts, in other words, built as open source. And not just any is open source, but open source with proven governance models that help people work together and trust each other across geo-boundaries. I submit to you that in the Linux Foundation and CNCF, we have a system that has been proven to work. And a key element of that is the code of conduct that helps people from different countries, different companies get together. This code of conduct applies to all events. And I'm not going to lecture you on the full extent of that code of conduct here, but I do want to point out one particular application of this code of conduct that's relevant to today's event. The CNCF and Kubernetes Slack channel have had some recent discussions regarding the mask rules for this event. And it's been a difficult situation with imperfect knowledge and actual changes in conditions. For example, if the virus in the wild out there today is different from two years ago, that might be the case. If survival rates have improved, if overall community immunity levels have changed and treatments have been discovered, maybe the rules that were appropriate two years ago are different from today. But we have to recognize that an organization like this has to put in place a rule. And we have one in place now today, which says that we're supposed to be wearing masks. I guess I get off because I'm a speaker and I'm a long distance away from you, but I'm asking you to understand that when people are involved, they don't have this big bank flip of opinion like a computer would where instantaneously everybody flips their attitude on masks and we have people with different attitudes, so let's just accommodate them here. Arguments over Docker or WebAssembly or the best Kubernetes form factor are thoroughly encouraged at today's event, but when it comes to masks, no debates, the rule says you wear them. So the way to get the most out of this event is to go beyond passive listening. Make an effort to meet people and hold discussions during the breaks. I'm asking you to look at the people around you at your table now and when we have a break, go out there and introduce yourself and hold an actual conversation because this is where the biggest value of this conference comes from. That kind of sharing. So it's important to realize the reason for these discussions is that at a conference, nobody is always right or always the smartest people in the room on all topics and you get smarter yourself by talking to people who have a little bit more experience and little niches and that isn't gonna take place listening to speakers like me or the people that follow me. That's gonna take place up in the hallway as you have these conversations. So I just wanna point out that along this theme of community, one more session I wanna plug at the end of the day before the closing reception. We have professor Sergio Mendez, I hope I'm not butchering your name, who's going to show you how to stand up a multi-node Kubernetes edge cluster on Raspberry Pis if you're new to this and want to learn how to get started. At this stage, we're going to recognize the event sponsors. We've got two diamond sponsors, SpectroCloud and VMware Tanzu and also a platinum sponsor, Red Hat. And I'm going to yield the podium here and allow the next speaker to come up. Thank you very much.