 I'm very excited to welcome our next team talk from the amazing Bloomberg visual data team. I'm so excited to have them here. First, we have Lisa Strasfeld, who is a data visualization designer and entrepreneur. She joined Bloomberg in 2012 as their head of data visualization and leads the Bloomberg visual data team. And before that, she was a partner at Pentagram, which has done some incredible work all across mediums, from large-scale installations to prototypes, to user interfaces. Really, their work is instrumental to our field. And we also have Chris Cannon, who is a senior designer on the visual data team. And he helps create the visual standards in UI design and product ideas for the many, many interactive data visualizations that we see. So please give them a warm welcome. Hi, it's great to be here. Thank you, Irene. Thank you, Voku. It's especially nice to be presenting today with Chris. And, as Irene said, we're here to show you some work from Bloomberg visual data and just describe what we're doing, what we're exploring, what we're thinking about. Actually, I'll back up for a moment. I was at the MIT Media Lab yesterday for Bloomberg, and I was a sponsor of the lab. And it's the first time I actually had really been there in 20 years since I was a student there, which is where I first discovered this idea of data visualization. And it seemed pretty new at that time. And I left there, blithely, thinking, oh, okay, this is great. I love data visualization. Now I'll go do it and make work, which was possible for a few years. And then it was basically too early. So now, actually, as a result of this community, this conference, it's all possible. And I saw that five years ago or so, 15 years later, after the Media Lab. And that's when I thought about leaving Pentagram just to jump back into data visualization full time. And then left Pentagram, I guess, in 2011 to the startup in data visualization. And then Bloomberg called. And I had worked on some displays at the space in New York of Bloomberg and knew that not only was this the time to do data visualization, but I really felt like Bloomberg was the place. Speaking of qualitative or color mapping, color blend. Anyway, the Virgin Palace. Anyway, so joined two years ago and started building a team. I'll describe how we structured that and how Chris came in. And we'll show you a bunch of projects. But first, many of you know, think of Bloomberg in association with finance and the financial industry. And it really is, of course, it's revenues based on all this data and information that's provided to a few hundred thousand terminal customers. But it really is a news and information company and a media company. And for people like us, it's a big candy store of data. And it's been amazing just to work to be for the first time for me in a newsroom working with journalists. So it made sense a couple years ago when some folks at Bloomberg were thinking, including the CEO, they were thinking about Bloomberg's position, data visualization, they kind of finally realized that it should be a core competency and identify this need. And again, made a phone call and we started building a team. And the team has been focused on work for the consumer. There's a whole team, a graphics team that works on data visualizations for the terminal. So when I joined two years ago, I asked myself this question anew, because it had been a while since I was really practicing data visualization. I was doing a lot of other interactive work at Pentagram and observed this dichotomy between two different types of data visualization kind of in the practice in the field. And obviously I was well aware of all the amazing work at the New York Times, the Post, the Guardian. And observed that there were really data visualizations with two distinct purposes, two distinct audiences that were created differently. One was to provide explanation, the other was to guide exploration. And then these two things were really, again, sort of split in terms of resources, type of data, type of experience. The explanatory data visualizations were typically news driven, smaller sets of data engaging kind of driven by news stories. The exploratory data visualizations typically software products, larger sets of data, live and powerful, but kind of overwhelming. And this split kind of, and this understanding of these two sides has kind of driven a lot of strategy of what we do, including, here's two examples. These are examples from what we had seen when we joined Bloomberg originally, on the left side from Business Week, on the right, an example of the data from the terminal. And the first position of an idea of the team was, let's see what we can do in between explanation and exploration. Because it seems like this is what, this would be an ideal experience. You want to sort of connect with something, you want to explore. And then you get lost, you want to come back again. So the team was initially structured as a product team, and again, we're getting, we'll show you a bunch of work. And Chris was the first to join the product team as the lead designer focusing on UI. This is also what I was most familiar with doing. So it was basically a design development product team focusing on interactive data products. Then we had the opportunity to bring on a small team of graphic journalists from other parts of the organization. These guys, to me, are kind of foreign and new and exciting because they're journalists, they're trained as journalists. But they also kind of move over into interactive, and of course visual design. And then I have a co-lead who handles a lot of the production operations and works hand-in-hand in strategy. So this is the current team. And we have a catalog page. Our goal and one of our challenges is how to sort of get the work to the reader because, you know, destination sites are kind of of the past. So we, anyway, we'll get into this a little bit, but we do have a catalog kind of, if only for us and for the people who follow us, which tracks the visual data products as we call them and the infographics. The infographics are produced much more frequently. So the question is that we'll just sort of frame this and get into some of the projects. The question that we're particularly interested in and really kind of privileged and excited that we get to work on, which is what is the ideal experience of news in 2014 and how can it be supported by data? And for me what's important about saying 2014 is that all of these solutions are about scale, which, you know, there are a lot of business people who care about scale, but we care about scale, I think, and I think probably like the people in this room for almost like an imperative, like an ethical and also an aesthetic imperative. It's interesting. It becomes interesting and kind of weird if it scales. And how can that be supported by data? And again, the reader experience, we sort of find this moment where they move back and forth. This is, thank you Chris for this little visualization, which I love. It sort of scales back and forth between explanation and exploration. So what we make. We make three types of things. The first one was the one I was actually least interested in making because it's, this seems like the sort of lowest form of data visualization. You have the story and you have an embedded chart. So instead of doing that for our constituents in the newsroom, we created a tool with our R&D team. And it's something that we're actually, I think we've just decided to make open. So you'll have to sort of stay tuned. But what's exciting about it is there are now 70 people in the newsroom using this tool and it's completely open. So you can see what people are working on and these charts get embedded in the stories. I know a lot of our peer organizations have these tools. For us, it was important to have them sort of output with certain themes that relate to their various properties across the organization, different websites have different themes. Business week, Bloomberg.com, Bloomberg View. The other cool thing that we are hoping to release to the public is we can also generate charts directly from Bloomberg data sources that are publicly available. So stay tuned for that. And we started off very simply with bar charts and line graphs and we're just enhancing. But it's been exciting to move into the tool area. And again, it's interesting to see these things populating Bloomberg news stories and opinion pieces. Infographics, again, we have a small team that's been producing these news-driven graphics. About investigative pieces that are done by Bloomberg journalists that accompany those stories. But also the graphics team is really responsive to current events. But it's mapped together pretty quickly. And we sort of look, it's been interesting for the product team to observe and collaborate with the graphics team to see what opportunities we can build to scale. And these are just some examples of static examples. The team has started collaborating on some interactive examples. So again, this is where the sort of teams collide and collaborate. And this transition to Chris will sort of describe the other thing that we produce, which he has a key role in, and I'll stand over here. So this is just an example of the interactive infographics that we started to be able to build now that we have more R&D available to us. So unlike infographics, we started off creating data products. And what we mean by that is a data visualization that updates automatically and continually. And so it's evergreen. It's not just a snapshot in time. And these are meant to connect to not just one news story, but several, hopefully a lot of news stories over a long course of time. So where these data products lie is sort of where visual product interaction design and the editorial connection all meet. And they're meant to be a visualization of a large dataset for specific subjects. Could be housing. It could be unemployment, politics, the stock market, S&P 500, et cetera. So we think of these products as being cubes of data and that we're just taking slices from in order to help tell the story. And they automatically update and they automatically update in different intervals. So some annually, quarterly, monthly, but a lot of them are daily. And they're designed to be a destination for domain experts as well as a general consumer audience. So one of the main things that we want these to do is be able to link to a number of news stories and vice versa. Which is really hard. It's one of the biggest challenges we have. So one way we do this is that each of our products generates a unique URL that help capture these curated views. So every time you kind of dive into a product, you're sorting and filtering, it's changing the URL. And that URL is something that could be shared. And also we want it, not just the consumer audience to share it, but we want Bloomberg journalists and editors to be able to kind of deep dive, create their own view, and then be able to share that along with a head and a deck that they're writing along with it. And because transparency is like a core tenant of Bloomberg itself, like the company, we try to offer as much transparency as possible. And in our methodology, we outline that as well as update schedule. So we developed somewhat of a simple criteria for pursuing and developing these products. We want them to be updateable, give people a reason to return. We want them to be scalable to allow additions in the future to be additive. And we want them to remain relevant to support ongoing stories. So again, not just be a snapshot in time. Of course, there's lots of challenges. We were a new team starting from scratch. So there was a lot of problems that came along the way and we had to solve them as we went, including massive security that's involved in just working at Bloomberg and just trying to get over to that firewall. I myself come from a graphic design background. And I know that just because you have a lot of type phases, a lot of colors doesn't mean you should use them all. And obviously, it's the same with data. And Bloomberg has lots of data, but that could also be a problem as well is trying to get down to what's essential. And since a lot of what we show and visualize is very complicated, we often have to spend quite a bit of time learning about ourselves, talking to the subject experts that are already at Bloomberg before we could explain it to a general audience. And because Bloomberg terminal customers pay a lot of money for this data, we're continually struggling to be able to see what we can take off the terminal and put on the internet for free. And actually, so far, not only have we not found that much resistance, but some of the people that were initially responsible for creating the terminal have supported us in this measure. And again, it's that continual challenge of being able to connect our products, the views of our products to various Bloomberg news stories, and then have those news stories be able to connect to our products. Okay, so now we'll show you some examples. But again, in this context of understanding the relationship between data and news, for us at Bloomberg and just in general for anyone covering news, so much of news is actually data driven. And some data is released, it's evidence of the housing market moving up or down, or the economy doing this or that, and it's the story. So the first project that we built was more of an opportunity and an exercise. Its intention was to, it was done right before the last presidential election, and we had hoped to sort of connect it intimately to news stories, which turned, I can describe what worked and what didn't work as well. And we put together this huge kind of Swiss Army knife of data, which we realized was probably scoped too large. That's one of the biggest challenges is how do you wrangle, how much wrangle, how much data should you wrangle. The cool thing about this, a lot of the data that's in the terminal is aggregated public data. I mean, there's a lot of proprietary data. And of course, it's from all different sources, like the Bureau of Labor Statistics releases all the unemployment numbers and workforce numbers. We wanted to include politics data. And again, we went to all these sources. It's all there, but there's an opportunity to design something that's more integrated. And we also used real clear politics polling data. So there was a view, there's something on the terminal called the Bloomberg Economic Evaluation of the States put together by an analyst. This was also a cool index that we were able to use. I learned two years ago at that time, why North Dakota was the most economically healthy state in the unions. Everyone know why now. It's all clear. Fracking oil. Anyway, so this was there for us to use. North Dakota is still the most economically healthy state. And it was, again, an amazing opportunity to kind of start with these sources, aggregate these data sets, which again, we realized are probably too many. But we just, we were kind of hungry for them. And we worked with Chris. This was when Chris first came in and started working on the UI, which was really about allowing people to choose their own data sets, select states, change views, suggest time periods. Chris will explain this because you'll see this in other projects. We always take a look at the data to make sure it's interesting. This is unemployment data, which you could see, you know, we could see easily some blips from Katrina, a couple states. And we started, we've actually, the team has been switching from Excel to R for those data studies. And I will show you a demo of state by state. I should say that the, oops, I wasn't supposed to do that. We had an early collaborator, prototyper, developer, who's here named Christian Swinard over there, who built this for us. And we still collaborate with him on a bunch of projects. This first view, which was actually the most engaging view, and I checked this daily on my phone, was looking at swing state unemployment and polling. And the story of that election was about swing state unemployment. I was really worried about the election. And until we launched this thing in September, there's something interesting going on there, September 2012. And then I thought, okay, you know, it's looking pretty good here. These states are all polling for Obama. And then October started to get a little more concerned about some of the swing states. Again, this data was pulling, was updating daily. And it was a pretty engaging, addictive view. And so again, we could easily compare those things. I'll show you what happened after the election, which we all know. We have a lot of data sets in here. If you look at, this is margin of victory. And you can get a time series just to see, in this case, how swingy these states are. For presidential margin of victory, you can compare margin of victory in unemployment. Again, this is what the lead, the DC bureau chief really wanted to see. Let's look at unemployment numbers and voting patterns. So this view really showed that. And then of course, we couldn't help ourselves and we added, and of course, we have maps in every view. But we added a lot of other data. Of course, we have it for all states. And I'll just show you a couple data sets. And we have data about jobs, income, healthcare, housing. You could look at, you know, uninsured rate for every state. You can sort. You can filter. You can see time series. You can show percent change. And this is where we believed that we had this sort of fantasy that everyone would come to this site, that authors would come in. And they'd be writing a specific story about that would require working with a few sets of this data. But it turned out to be quite overwhelming. And we learned a lot of lessons from this. We're still always kind of going back into these products and they're live. But in this case, apart from that early view, it really definitely bent more to the exploratory side of the spectrum. And to kind of scale back from that, Major League Baseball team values is something that almost started off as an infographic, but then became somewhat in that blurry area between infographic and one of our data products. So basically, and I'll get to the billionaire's product later, but a lot of billionaires own sports teams and Bloomberg journalists kind of did some investigating and realized that sports teams are actually very undervalued, at least by all previous estimates. So there's already some of this data existing in the terminal, but again only for Bloomberg terminal customers. One of the Bloomberg billionaires journalists actually spent nine months investigating all this data and how much teams are actually worth. And because he spent so much time, we received a very clean and thought out set of data that enabled us to turn this little product around in about two and a half weeks. So we were given data sets about team values and concession sales and receipts at the gate and sponsorship rights and all that. If this was to be just a static version, it would be more like a stack bar chart like this. But we were able to create a polar diagram. So what we have here is polar diagram showing in order clockwise teams with the most to least value with the New York Yankees being the most valuable. And we can see a breakdown of how that value is derived. And if you need to get some more idea of what these things mean, you can just click on the valuation components and learn about that. Rolling over any one of these other teams, you start to be able to compare concession or sponsorship numbers from one to another. And if we click on, say, Boston Red Sox, we can shift the attention, the focus on to that team and begin to compare. And now we could sort also on league and division and most interestingly wins. And we can see that just because the team's value bill does not correlate anything to their wins. That distinction actually will go to St. Louis Cardinals, at least for the last season. So data is context. Pay of the Pump was a ranking of how much gas costs in different companies. And it relates to how much people's average income and how much they spend and consume from country to country. I think there was about 55 countries that were included in this. Again, this data existed already in the terminal. And there was rankings that were outputted in these simple generic PDFs. And on the consumer website exist in the form of a slideshow. And there was some interesting text, but it was paired with some generic stock photos of each country. And it was basically a clickbait to get you to go through all 55. We were sort of tasked to create a more compelling data visualization. So this just has five data sets, average price of gas, daily per capita income, daily gas consumption, and a percent of the day's wages needed to buy a unit of gas and the percent of annual income spent on total gas purchases. So we wanted to be able to sort and adjust the time period, change and to satisfy a more global audience who wanted to be able to change the currency and the units of gas from gallons to liters. Some early experiments in R. While they were interesting, they started leading us down kind of a murky path. We were concentrating on the pain of the pump and we wanted to be able to express to other people how another country's pain is in buying gas. So for instance, a gallon of gas in India would cost the equivalent of $155 here. So while we complain about the price of gas here, we actually have it very good. So this was interesting, but it was actually very hard to explain and we wanted to avoid having to do a lot of upfront explanation. We played around with bar charts. We played around with bubble charts as well, but it was just too complicated and again, it's hard to explain. So we went back into the data in the Excel sheet and we represented each country as just a thin line color coded by region and then we got the idea to change the whole design of this. So here we see three columns. The price of gas, unaffordability, which is the percent of the average day's wages needed to buy a unit of gas and the income spent on gas. By default, United States is highlighted, but if we roll over other countries, we can start to see comparisons. So some countries, like Norway, are very wealthy. Gas is expensive there, but their average income is as high, if not higher than the United States. Other countries, such as Venezuela, have a low income, but they also pay next to nothing for gas, the equivalent of 4 cents a gallon. Clicking on any one of these countries, I can start to see make comparisons and change the focus. So say I'm a viewer from India and I'm interested in how India stacks up. I can see that and I can change US dollars to Indian rupees and I assume they're on the metric system so I could change it to leaders and now this information means more to me if I was an Indian viewer. Also I can color code by region and start to see if there's any correlations or clusters among all of these. Industry leaderboards. So this was a franchise for the company. It coincided with a special issue of Business Week and a conference held in Chicago, but it was basically taking a look at 600 market leaders within 55 global industries. So to be able to compare companies to each other as well as the industries that they're in. There's a whole industries team at Bloomberg. They already had a lot of this data, but we required a lot more wrangling of it. So we have, well the key thing about this is that there's forward-looking data, so estimated sales growth and so on, performance, size, and profitability metrics as well. We wanted to be able to filter, sort, search for a particular either country or industry or all the companies from a particular country and be able to highlight them and retain that highlighting from view to view. And there's a view of 55 industries as well as 600 companies in their grid and the tree map format. So we got all this data trying to figure out what to do with it, came up with some just initial ideas of how we would represent 600 companies, how different ways we could color them to indicate different metrics that we have, be able to highlight them or roll over for contextual information, how to place them within their industries, and get detailed views of each industry. We initially played around with a prototype like this to try to explain how each of these squares is a company and they are placed in columns, which is their respective industries, and how these industries could be arranged along different metrics. But again, it required too much upfront explanation. We wanted the user to be able to just dive right in. And we were aware of the, very aware of, you know, prior art. We always do a kind of audit, including in this case the smart money market map, and we're intrigued by the idea of doing a kind of sortable tree map by industry, which we hadn't seen before. It may exist. Someone let me know if it does. So what we see here is 55 industries and all the market leaders in them. And the companies are sized by their market share within that industry. Now I can ask myself, one key thing to note is you can either reveal the metrics and start playing around with the different sorting or you can just simply look through all the questions here. So what are the most profitable industries? Well now it resorts to show me that the industries are ranked by operating margin and as well as all the companies within them. One view that I like a lot is, what are the lowest values companies within the fastest growing industries? Now say, I'm a viewer in China and I'm interested in all the Chinese companies. I can look that up and see them highlighted here. Now if I want to break out the companies by themselves, divorce them from their industries, I could do that and they're still highlighted, which is great. Perhaps I want to see the estimated sales growth. Looking good. Yep. Overall, a lot of them are toward the top, so that's good. And there's a tree map view too. It's good for size metrics. If I want to find out more about one particular company, I can do that. It takes me to the industry view and I can start to play around and look in here, do some sorting, get definitions if I don't know what PE ratio is, which I didn't before. I've been working on this. And any one of these companies, you can open it up, get a little bit of background information, and then there's links to Bloomberg News, recent Bloomberg News stories on that company. Okay. So after doing the leaderboard, we're still, that's a live product that we maintain and we still are challenged to connect that to news stories. So that's been on the back burner. But in the meantime, we decided to experiment with something else. Again, the team is really focused on output because it's our self-esteem, it's how we measure what we do. But we're also just trying to experiment again in this realm of news and data. So we decided to focus on something much more on the explanatory side. And we had a very nice relationship with Bloomberg View. Anyway, the idea was to sort of invert the relationship between chart and text. And very simply, and to create something immersive, to get inside the data somehow. And we've been working with Bloomberg View and helped with the redesign of their site. Bloomberg View is the opinion site, opinion consumer media site at Bloomberg. And the format is quite different. Obviously, these pieces each have, they have a voice of the contributors of the editors. And that was actually kind of exciting, because we were interested in exploring these are some mock-ups that we did with another collaborator, Takaki Okada, who also works with Christian Swinehart. And we took a piece that was written for another purpose, explaining the fall of gold. And experimented with this idea of getting inside the data, looking specifically at time series. What's important about these mock-ups are the vertical lines. There's time on the bottom. Typically, when these writers and analysts do an overview of something like this, they're talking about time. They're kind of zooming in and they're pulling out. So the design feature of this, one key design feature is the thickness of the line. So that when you're kind of getting in closer, the line gets thicker. When you zoom out, you have a thinner line, a lot more vertical lines in the chart. And again, the more, the voicier, the better. So we like this idea of a small bit of text and something conversational. So it was actually perfect to work with Bloomberg View. Then, as a lot of things do on the team, this changed hands to Jeremy Diamond, who started putting his own design interactive talents into this, Chris working on the UI, a collaboration, and then it became something else. So we've done four of these and I'll show you one of them. The most recent one was how Americans die. It's, we sort of appreciate the editorial, not just the writing, but the editorial, what the editors do in sort of titling these things is so key to how they do online. Jeremy's been building these things. We've been iterating them and we're just about to sort of move it into an authoring tool. And again, I love how specific this thing scales only to time series data. So you could not, you know, there are other charts that the writer had included, but it really told the story through time series. And we'd like to sort of make these kind of weird idiosyncratic templates for other things. Bill Gates actually tweeted the how Americans die piece. And that's the most recent one. I'll give you a quick tour of, I did this again. Okay. So we start with a cover. And again, Jeremy should give a talk just on how he built this, which this audience would probably really appreciate in D3 and a lot of other stuff. And I'm going to take you, this one goes through segmenting the population. Matthew Klein is the writer of these pieces. He's kind of brilliant young economist who takes data. He can sort of correlate anything on different axes in a good way. This one really isolates the effect of AIDS on a certain segment of the population. And anyway, you should check that one out. But I'm going to take you very quickly to the one about housing. And again, I encourage you to check these out online. There's been, my interest, whoops, is on the kind of the experience, the UI side, the sort of idea of like how we scale this into a template, the engagement, the voice. Again, it helps that what Matt Klein does in the editorial is he segments things in really interesting ways. So he takes the housing market, which typically we see as one line, which is the case Schiller index of 20 cities. So he segments that into what he calls bubbly cities and sane cities. And suddenly it's a whole new way of understanding this. And this piece is particularly amazing because it really talks about how the housing crisis and how it affected different cities was really about the loan practices in those cities, basically how freely money was available. And then of course it maps to foreclosures and I'll sort of skip through this. But it's an amazing tour, completely eyeopening. And again, I love like we also sort of find and choreograph these moments of zooming in and pulling out. I'll kind of very quickly go to the end of this thing. This one view bothers me because it was not a time series, but let it go. But it's a really engaging view. And we decided to do it. My favorite part is kind of this last part where it goes from this wide view and basically zooms in to each city. This is also talking about private equity, buying up mortgages in all of these cities. Anyway, I highly recommend checking these out. Again, the great work of Matt Klein and Jeremy. I'll let you tab back. Okay. Running low on time. 45 minutes goes by really quick. I'm going to skip this one. It's not really data this. So data is guilty pleasure. Not everything that we do is super serious. So that's where Bloomberg Billionaires Index comes in. I'll try to explain this really quick. There's a team of 14 journalists at Bloomberg. All I do is track the world's wealthiest people and uncover hidden billionaires. It is truly guilty pleasure. A lot of this existed on, well, it existed in a very abridged format on our website already. And the data already existed in the terminal, but you had to be a terminal subscriber to see it. I'm going to go through this really quick. Maybe just go right to the demo. Oh, one thing I'll say is each one of these billionaires is given a unique ID and there's some interesting anecdotes along with each of them. For instance, who left school at age 13 to start his own business? And then this one called his youthful affiliation with Nazism, his biggest mistake. That's Ingvar Komprad, who is the head of IKEA. So think of that next time you shop for a cabinet or something. But then again, who has flirted with Nazism when we were kids? Used R to kind of create some scatter plots to see if there were any meaningful correlations. Used some processing to see if a geographic view was worthwhile. Illustrations, because all these different billionaires, they had different angles, different lighting. Most of them aren't that attractive. Looked at a lot of different illustrators, found a great one named Lena Chen. She was able to take all these photos and create a nice likeness of them. You might notice that some of these people in the product have these generic silhouettes. It's because some of them literally do not have photos of them. Or in the case of John Mars of Mars Candy at the top, that's his college yearbook picture from 1962. That's the most recent picture we have of him. And that's Ann and Nicole Smith's daughter-in-law down below. Long story. So, oops. That was a really quick demo. Able to view all the billionaires, they're ranked by net worth. The numbers underneath are their dollar change from previous day. Could also change it to sort the order by the dollar change of previous day. You can see that Prince Alawid Al-Sad, he has made $1.6 billion since yesterday. And Jeff Bezos hasn't done so bad himself. But don't feel so bad because you scroll down. Sergey over here has lost $629 million. This is our Explorer view. You can highlight, say, if you want to highlight all the Russian billionaires, you could do that. Maybe all the Russian billionaires that are involved in the energy sector. And we could see that they're not doing so well. And actually, a lot of Bloomberg reporting has shown that Russian, well, not just the stock market, but the richest people in Russia, their wealth has been dropping ever since the crisis in Ukraine. That's kind of interesting. We have a rank view. Let's see if this actually loads up. We can see, this is a one-day view, but if we want to see over a quarter, over a three-month period, it loads up. Here we go. Here's a ranking of all these people. Not that much movement in the top 10 or so, but then when you get down here, you start to see some interesting, oh, and then since I already have Russian billionaires and energy highlighted, they remain highlighted in this view. So we could see what a bumpy ride Lee need had here. If we want to be able to say, see what industries all these people are in, I'll take away all these filters here. And we could start to see who has their money in cash, and that's not many of them. In fact, most of them have all their wealth tied up in stocks. And clicking on any one of these heads gives you a detailed profile of that person, and you can see some of their possessions, such as Hans's rural estate in England. Good to know. And again, you can plop them on a map as well, and I'll skip that. Okay, yes, and we're done. But what's next? We, again, we're still continuing to investigate in this area in between news and data. I swore I would never design anything for a phone, but have been convinced by the team. And actually, and kind of excited, we're starting to work to design on phones. And looking much more into, you know, how people get to the content through social media, always iterating on the current projects, current products, producing infographics, working on things for sports, the next election, and climate change. So please, oh, and of course, all of this is possible because of our amazing team. And thank you, and please visit us on visual data and Twitter. Thanks.