 Okay, we're back here wrapping up, winding down day three of extended coverage on SiliconANGLE.com's coverage of O'Reilly Stratoconference. I'm John Roy DeVolante of wikibon.org and SiliconANGLE is excited to be here and it's a great, very tiring, but exciting show here. I mean, the range of topics is phenomenal and obviously we go deep into the weeds on the tech and we talk programming, we talk data science, but the theme that really kind of hits the mainstream that really is on the front and center stage is design, design, user experience, and we have Julie Steele who's an editor at O'Reilly Media, Julie, welcome to theCUBE. Thank you. This is a fun session, you've got some props to show us, but you're involved in a lot of the work going on in O'Reilly around design, data sensing, really kind of like the emotional feeling where design is not something about the speeds and fees, but about the user experience and data can enable emotional responses, it can change things. We know, we reported an exclusive story today that's looking at an angle about the Iranian hackers and threats on Joe Biden and taking over drones. It's scary, at the same time it could be scary, but also emotional in a positive way. So Juan, just kind of give us your take of this strata with the emphasis of design and art and et cetera here and then your vision of that. Absolutely, well, one of the elements that I've tried to bring into strata, starting with our first one, but growing each year is the data visualization showcase which you may have seen outside of the main ballroom. And that's an attempt on my own part to curate a collection of data visualizations that employ design to make data more visceral and more accessible. And I've really tried hard to go outside the box and find things we had a piece this year that was done in cross stitch. So it's not just your typical interactive video or line chart, but the other thing that we've done to try to make design more visceral is that we've done this data sensing lab. We've probably seen some of these sensor modes around the event, we've got one here, yeah. So this is something that we built with our hands, we soldered it. It's built from an Arduino Leonardo board. This is all open source electronics, but we've added some sensor modes and we've used an XB digital radio to connect all of these sensor modes wirelessly. And they're measuring things in the environment, like temperature and humidity and sound and motion with this passive infrared sensor there. And what we really wanted to do is remind the attendees who are here who are mostly data scientists who often begin to work with data when it's already in a spreadsheet that actually this data comes from somewhere, it comes from somewhere very physical and that it has to be collected and analyzed and you have to design what you're going to collect and how you're going to do that. So that was our test. And then, you know, internet of things obviously is the big thing and sensors, whether it's in a data center or in public places. You don't want this big, weird looking thing. It's got to be industrial design, right? That's right. And design can have varying degrees of goodness. So we've actually had some difficulty. We've had to go back and fiddle with a lot of these things. We found that as we place them around the conference venue, they're actually somewhat delicate and even though we put signs, you know, letting people know that they were there and not to disturb them. Some of them got kicked and we had to go around and put them back together. And obviously that affected the data we were collecting. So when we go to look at that data in our spreadsheet, we're going to have to notice, well, some of it is missing, some of it may not be entirely accurate. And so the design really impacts the data we end up with and what we can learn from that. So what's this other prop here that's very interesting? It's a button. Yeah. So we've got this giant, it almost looks like a proprietary button you may have seen advertised on television but ours is, we call it an awesome button and we put it, we have one outside each session room and we ask attendees to hit the button, which lights up when they do so, to hit the button if they thought the talk that they just heard was awesome. And so we get signals back from this. We also use an online session rating form where we ask attendees to rate sessions. So I'll be very curious to go ahead and look at the data afterwards and see whether the awesome button got more people interacting with it than the online web form. And I'm predicting that it did. And will you correlate that data as well? We will, we will. And we have things like head counts for each session so we'll have to normalize, we'll have to say, well, if the button outside room A got twice as many clicks as the button outside room B, is that because there were simply more people or because it was really a better session? So we'll do some of that normalization. The awesomeness button, that's awesome. That's great. Now, so you're designer by background, right? So how did you end up sort of here? Well, I'm an O'Reilly employee. For a long time I was a book editor and I was editing books about data science and various related things like natural language processing and machine learning. And when we started the Strata conference, it became apparent very quickly that it was going to take off and so we needed additional people to work on it. The powers that be decided, it was easier to pull me over from editing, teach me how to run conferences than to take someone versed in conferences and teach them data science. So I ended up learning a new skill set, getting very involved in Strata but it's been a lot of fun for me to go from something very visual book editing to something very hands-on like this hardware. Amazing. So now, so what's next? Do you have any ideas for the roadmap? Absolutely. So this is the second time that we've done the data sensing lab and we hope to do it again in the fall in New York at Strata New York. We'll probably refine some of the ways that we're collecting data. We'll refine things like the collection rate so how often we're asking it to report back and we'll try to mash it up with other kinds of data that we're collecting. One thing that we would like to do next time is use flow meters to detect how much coffee people are drinking. We'd like to be able to ask questions about whether the temperature in the room or the time of day determines how much coffee people are drinking. Can you correlate that with a meter in the bathroom too or is that something? That comes later, we'll see. So what are the things that you're learning? I mean obviously you've been a book editor for machine learning that speaks volumes around your geekiness and you have but a little different creative view when you talk about visualization. It has to take on an art and a science. So the personal question I want to ask you is what have you learned? I mean because you probably had some things that kind of come out or left field or some things that are pretty obvious. Can you share with the folks some of the things that you've learned around this evolving important area of design and experience and... At the end of the day it's all about storytelling. I think stories are at the basic most human level the way that we communicate with each other and also with ourselves. So ultimately all of this is about trying to tell a story. We're trying to collect information so that we know what that story is. Design also tells a story back to the attendees and the people who are interacting with these things. And something I guess that surprised me was just how much of the story that you can read. I'll use that term. It was wonderful to see things like the temperature spike as you move 2,000 people into the ballroom. Or even to see things that didn't happen. We were expecting the humidity to change a lot more than it did with that number of people moving in and out of a room. So we have some questions to follow up on. Some things we need to dig deeper on and get to the bottom of it. Why did this change in the way we expected or not change in the way we expected and what's really behind that? John I'm sure we can think of an application for theCUBE for one of these things. We need an awesome button. We need an awesome button. Yeah, awesome button. And then like when John and Dave are hitting tilt button. That's... Julie, thanks for coming on theCUBE. We really appreciate it. I want to give you one parting. Where does we wind down? Day three, the final day here at Strata has been an amazing conference. What was the coolest thing that you saw here at the show here that you didn't expect? I think my favorite thing, they don't let me out of the booth very much, to be honest. Or I don't let myself out of the booth very much, I should say. But I just came from what we call the great debate where Monica, Regati and I and a couple of other folks were debating math versus design, which one is more important. And the vote went narrowly to math, although I promise you we needed to design a better voting mechanism because we were a bit inaccurate in how we counted the results. Well math should be elegantly designed in an artistic way. Elegant code was the word we heard on theCUBE today. But it was wonderful to see how many people came out and how frankly narrow the vote was and to see, I frankly expected that it was going to be a landslide to math just because of the audience bias, but it was heartwarming to see how many people are here from either a liberal arts background or some other kind of design background. And it's nice to see the meeting of those communities. One more thing I'd like to ask before we kind of wrap it up is what are you going to be working on going forward, share with the folks some things you're interested in and you'll be exploring here at O'Reilly. Absolutely, the next thing that's on my list is StratRx, which is our data science and healthcare conference. I chair that conference along with Collin Hill of GNS Healthcare and we've just opened our call for proposals, so the next thing on my list is to sift through those proposals and it's an exciting time in healthcare to be applying some of the analytics to that and trying to change this thing that matters so much to all of us. And certainly design matters inside healthcare facilities certainly and big data is important. Okay, Julie Steele here in theCUBE, SiliconANGLE's exclusive coverage of O'Reilly Media StratRx. We'll be right back with our next guest we wind down day three after this short break. Stay with us.