 I wanted to end on a high note because we have a team that's been working on a really interesting project based on the work they've been doing with Steel2. And they're going to end us with a little demo and some lights and maybe some strobe action on something they've created. So I want to bring up to the stage Tim Hess, an engineer at Pivotal, and Jenny McLaughlin from the Senior Platform Architect at Pivotal. Come on out, Jenny and Tim. I won't do that again. Thank you for coming out, Jenny. So we have support devices that are placed strategically around the stage if you're wondering why we were hanging lamps off the corner. And we're going to let Jenny and Tim tell us a little bit about what they've been doing and why we have lamps. Well, thank you for that, Abby. I'm actually an engineer on the Steel2 project. So we've been working to build tools for .NET applications. So, Tim, what was the motivation for your team to develop the Steel2 framework in the first place? Well, the same way that Pivotal has been working with the Java community to develop Spring projects, we wanted to have tools for your .NET applications to solve problems with microservices. So here are some of the examples. In the most recent release of Steel2, we added a client API library so that you can work with Credhub to store credentials like passwords and certificates. So as a .NET developer, I can leverage the same cloud-native patterns such as the circuit breaker for preventing cascading failure, service discovery at runtime, and centralizing all my application configurations, if that's right, right? Absolutely. So that is all completely open source and we recently joined the .NET foundation. Well, that is very cool. You know what else is cool? These light bulbs, so these light bulbs are smart. They allow me to change the color remotely. Spark light bulbs are getting really popular in home automations and among Internet of Things products. So Tim, can we demonstrate how easy it is to use Steel2 and then microservices to change the color of these light bulbs? Absolutely. But rather than just arbitrarily changing the color like you did just now, how about if we work in that new Credhub library? We're going to be running this demo in Pivotal Cloud Foundry, so we've already got a Credhub server available. Oh, that's an awesome idea. We can analyze the strength of a password and convert that strength to our RGB color code, then send that color code to the light bulbs. I think it's probably also a good idea to plug in a circuit breaker here, since this is a traditional systems integration. So that way we have a fallback algorithm if for any reason we can't reach the Credhub server. That makes sense. So let's switch to the demo. All right. So we have a color gradient here. Strong passwords are green and weak passwords are red. Let's start with something a little on the weaker side here, five characters. Not looking like a very complex password. We probably shouldn't use that for anything important. What should we do? Should we get something a little bit stronger in here? Abby, what do you think? Absolutely. We are, it's everyone's attention spans are waning. We need to amp it up a bit. Amp it up. Let's get something tougher. Well, I don't think I could memorize that password, so it ought to be good for a few centuries. Wow. Yeah. Literally it's going to take centuries for us to crack that password and we got green light bulbs. That was fun. So what else could we do with these smart light bulbs and the studio framework? How about if we throw some sentiment analysis at them? So sentiment analysis tools are getting better and becoming more popular. Anything you say can have a sentiment score. For example, after a hotel stay, when asked to give feedback, if I were to say something about my experience, well, that's not what I meant. So that was a neutral statement. We got a sentiment score of 50%, which is mapped to color yellow. Sorry. Sometimes I get a little literal. What I really want to do is to give a positive feedback, right? So something like, I love the staff and food, maybe. Right? Now we get 98% sentiment score. That is a very positive feedback. And we mapped that score to color green. And now if I were to say something negative in the receptionist, we didn't mean to say that. Sorry. She was kind of mean. So now we get some really red bulbs. So if we switch back to the slides, we'll talk you through how we did this real quickly. We take whatever was entered into that form, no matter how silly or basic, we send it off to the Microsoft Cognitive Services, specifically their text analytics API, and we request a sentiment score. That sentiment score comes back as a decimal, which is effectively a percentage. We map that using our scale to a color. And the color to the light bulbs, voila. And cognitive service requires an API key. Obviously, we're not going to put that API key in our source code and upload to a public GitHub repository. So we decided to leverage, you know, config server backed by a private registry to store that API key. And you're still told to retrieve the key at runtime, all right? So how does everyone like our demos so far? All right. Let's do a retro. If you like the demos, please tweet with hashtag CF bulbs right now with all your wonderful compliments. If you don't like the demos, please don't tweet at all. Of course, I'm just kidding. We welcome your constructive feedback as well. So the way this demo works is we're going to reach out to Twitter. We're going to grab the 10 most recent tweets with the hashtag of CF bulbs. We're going to bundle them up, send them off to cognitive services, get individual sentiment scores for each tweet. We're going to average those scores out together, apply the score to the color scale, and send that to the bulbs. We also implemented the service discovery pattern between the Twitter monitor and our smart bulb UI application. Of course, there's a config server to store the location of the sentiment analysis tool. Let's go ahead and switch over to the demo. Yeah. We can hopefully get all your feedback in. We'll see what you have to say. Not bad. Thank you. Thank you. 90%. I'm happy with that. Thank you. That's awesome. So the goal of this demo is for you to see what is possible, and you can do easily as well. If you would like to see more behind the scenes, things like circuit breaker dashboard and how we actually use the spring cloud services to build out the demo, start by pivotal booth for the next couple of days, and we'll be happy to explain all the details to you. If we switch back to the slides again one more time, we have the GitHub location, and you can pull on our code. It's open source. Feel free to do whatever you'd like with it. Have some fun with it. Get a chance to see how easy it is to work with Steeltoe. Yeah. Everybody should download at their house for other smart bulbs, and that way they can detect what's actually going on. Yeah. Yeah. After you have had some fun, you can start to apply these cloud native patterns to your own .NET applications. Things like you can wrap your integration points with circuit breakers in your system or apply all the application configurations, put those things in a config server. All right. To sum up. Thank you so much for coming up. That was awesome. Come on, everybody.