 Good morning everyone. I'm Josh Wood from Red Hat. I think you've already heard that. I am a developer advocate for Kubernetes and OpenShift, so I concentrate a lot on automation around these platforms. Some things I've talked about this week, including operators, relate directly to that. We're not going to talk a lot about those things except at a very high level in today's talk. Instead, I come to you folks with shocking news. Software is people. It's people. And what I mean by this, I'm going to try to illustrate, and I actually think by pure accident that it made Dovetail fairly nicely with the longer form talk that you just saw, and I hope that it does so and serves as kind of an exhortation or proration on the end of that. Now, in our industry, we produce a litany of words. We produce more words in terms than the best novelists. Just here at this conference open-source summit, just my company Red Hat has addressed a long list of topics that you can find in the schedule. I'll highlight a couple of them here. We've of course talked about OpenShift, our Kubernetes distribution. We've talked, as I just mentioned, about automation on that platform in the form of operators and some of the other things we work on, both upstream and in our own projects around Kubernetes. Some of my colleagues have spoken about the security of the containers that actually run on those platforms and best practices for maintaining security as we move into these kind of architectures. But we've also talked about things I might refer to as more of the people side of this equation. Another of my colleagues has spoken this week about using open data to make civic technical systems more sustainable and help improve decision-making processes. The same fellow has actually spoken about automation in the form of processes to make community management and collaboration easier among larger and more diverse groups. Now, out of all of that litany of words that we hear even in a short talk like this, I think we can narrow down at least a few clarifying principles. First and foremost, on the technical side of things we believe at Red Hat, security through automation is our only real weapon against the vulnerabilities arms race in which we find ourselves as software practitioners. Only by automating updates can we keep up with attacks against vulnerabilities that we know well are themselves automated. The nice thing about that is these practices of automation lead to velocity in our ability to deliver software faster and with better features. Velocity is kind of this business word or a favored word of project managers, but in engineering or even at a philosophical level, I think it's one of the dimensions that can help describe something John McCarthy wrote about in the 1960s and called utility computing. Utility computing can mean a lot of things, but what it's really about is making computer power meterable, switchable, like the outlets in your house or your faucets. In other words, it's about making this technology and the increasing power of these technologies easy for people to consume. So that's sort of at a high level a technical approach to these problems. Now at Red Hat, we see those two principles in security through automation, illustrated by CoreOS technologies about automatic over-the-air updates for the platform itself, Kubernetes clusters and OpenShift clusters. At the top end of that, we see OpenShift's traditional concern for developer experience, for the automation of builds and tests and other things that make it much easier to build software and deliver it onto that platform. We are performing that automation on new public resources that have led to new kinds of collaboration, I think brought us to a sort of golden moment of collaboration. Think of announcements from Red Hat and Microsoft working together, Red Hat and IBM working together. OpenShift on Azure and suddenly every Java developer I know is running Linux and a little window on their Windows laptop. You can insert your favorite recent tale of unlikely bedfellows there. I think from this, we can say that collaboration of this sort works is happening at unprecedented scale, and I think we know that it cannot happen without open source, and that's what brings us all to this conference. And the reason for that is because collaboration happens between people, not between projects. Another way of saying this, I saw this on Twitter this week, I really liked it. I don't know Mr. Nolas, but he's a data scientist. The most underrated algorithm in data sciences go talk to people. Again, why? Because those algorithms, all systems and again software is people. That's why communities build the best software. Open, diverse communities building open source software. So I urge you to go forth and collaborate. Join us in making utility computing a dream since almost the beginnings of our industry a reality by making these things automated and easy for people to consume to help developers concentrate on features instead of platforms and infrastructure and deliver continually more interesting and more powerful software. Design for security with automation and you will find velocity in those same techniques for automation. Most importantly, design for humans and for consequences. We all compete or at least our companies do, but I urge you in the spirit of this conference that brings us together this morning to collaborate along the lines of that same competition to build better, faster, easier software in the future. Thank you.