Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Episode #64: Kim Rees
135 Plays8 years ago
On this week’s episode of The PolicyViz Podcast, I chat with Kim Rees, who is Co-founder and Head of Visualization at the data visualization firm Periscopic. You may know Kim from her great work at Periscopic, her talk at the...
The post Episode #64: Kim Rees appeared first on PolicyViz.
Recommended
Transcript
Introduction to Kim Reese and Periscopic Studios
00:00:11
Speaker
Welcome back to the Policy Viz Podcast. I'm your host, John Schwabisch. On this week's episode, I'm very pleased to be joined by co-founder and head of visualization at Periscopic Studios in Portland, Kim Reese, the one and only. Kim, welcome to the show. Thank you, John. It's a pleasure to be on the show. I love your show, by the way. Thank you very much. How are you doing? I'm doing well. It's a sunny day here in Portland. The fall is treating you well so far.
00:00:38
Speaker
So far, so good. All right, good.
Future of Data Visualization and New Technologies
00:00:42
Speaker
So I thought we would chat a little bit kind of open-ended on this episode. I thought we'd chat sort of about the future of data visualization and what you and maybe some folks over there at Periscopic are thinking about when it comes to new technologies, new platforms, whether it be mobile or tactile things or sound. So what are you thinking about these days in terms of the future of data
00:01:08
Speaker
Yes, that's a fun area to start thinking about. It's something that we've been thinking about for a while. In our house, we got an Alexa. We call it an Alexa when it's actually an echo. No, it's Alexa. I wouldn't even say it's an Alexa. It's just Alexa.
00:01:30
Speaker
It's just Alexa. Siri is now Alexa to us. We call everything Alexa. We just yell at people on the street, Alexa. Alexa. There's a third grader at my kid's bus stop whose name is Alexa and I asked her for the weather and the square root of 234 and she doesn't know.
00:01:53
Speaker
So beware out there if your name's Alexa, I may ask you questions that you don't know the answer to. But yeah, I mean, it's fascinating to look at these technologies that are now assisting us in various ways.
Is Visualization Needed with AI Advances?
00:02:11
Speaker
And as data visualizer, I think about it a lot and where is the stuff going and why do we
00:02:21
Speaker
Why do we even need visualization these days? What is the point of it? Are we getting beyond visualization? Have we moved past the need for it? And I think that things like not only voice assistance, but things like AI and machine learning are now sort of jumping into a realm where people might skip over
00:02:44
Speaker
database and go right into like a heavier type of analysis to find those insights that they need. So something we've been thinking a lot about, but I think that there's a space for maybe a new way to start thinking about visualization. And it's something that I've been calling just in my own terminology, I've been calling it just in time visualization or just in time data.
00:03:12
Speaker
which is meant to sort of encompass all of those various data sources. And it's something that I think is sort of a natural extension of that sort of that mobile first mantra that
00:03:29
Speaker
you know, people have been championing for a long time, you know, mobile first, mobile first, but now it's really like, you know, it's not mobile first, I don't often go to my phone to find things out, I go to Alexa, I'm standing there, I'm washing dishes, or I'm getting my kids ready for school, or what have you, I don't have the time to go to my phone, right?
Data Accessibility and Immediate Information Delivery
00:03:48
Speaker
I'll observe that sounds, but you know, so I think that now it's, you know, the data needs to come to us wherever we are. So it's
00:03:57
Speaker
things like Alexa, things like in car services, heads up display, you know, things like, you know, I was even talking to someone recently about credit cards and how, you know, why should I have to go to my phone to check my balance or, you know, balances of my various cards or things if I find, you know, like right now I'm on the search for a rain jacket because this is that time of year in Portland. So, you know, when I'm looking at different raincoats,
00:04:27
Speaker
you know, should I buy the $120 code or can I go for the higher 250? Why can't I just hold my credit card up to the tag on a code and see a visualization of like, okay, this is gonna like put you close to your limit or this is gonna, hey, Kim, you have an electricity bill coming up at the end of the month. You know, don't forget about that. You know, I should I shouldn't have to go and check
00:04:51
Speaker
you know, my credit card balance and my banking balance and all these things, it should, you know, we need to start thinking about
00:04:59
Speaker
delivering those services at the time when people are in need of that data. So it sounds like not so much that data visualization is not needed but that the current ways it's being delivered is maybe not needed. So Alexa that would have a visual display might be the next thing.
00:05:22
Speaker
If you want to see what the weather is, you can actually see the visuals in addition to her telling it to you via sound. Right, exactly. I guess what I'm trying to think more about is not so much trying to force things into a visualization as to say, okay, how can this data be serving people? Is it voice assistant? Is it visualization?
00:05:51
Speaker
Is it something they don't even need to see, right? Like, so for instance, this is one that I've brought up before, but I think it's a good example.
Technology vs Human Involvement in Driving Tasks
00:06:01
Speaker
So, amber alerts, right? So, when a child's abducted and they send an amber alert, the alert gets pushed via text message to your phones. I guess the idea is,
00:06:13
Speaker
If I'm driving down the road, I get a text message, which I should not read, but I read it anyway, of course. So it's an Amber Alert, comes up with a make and model of the car and license plate. So now I'm driving around on freeway looking for a blue Kia with a license plate of, you know, 457 FFF, you know, whatever. I'm trying not to get into a car accident while I'm frantically looking around for this poor, abducted child.
00:06:43
Speaker
which is like, like a nice try, I think like an interesting attempt. But such a massive failure in practice. So to me, it's like, well, that that problem is like, of course, that's, you know, as a mother, that's like, Oh, my God, that's such a great thing to solve, we need to solve this.
00:07:07
Speaker
So it caused me to think about it for a time. And I think that we need to start thinking about how do we make technology do that work for us, right? Humans, one, are just fallible, period. And then, two, trying to put people into a fast-moving car and make them do some
00:07:28
Speaker
task like that is a disaster waiting to happen. But why not use the car itself that has a backup camera that can be turned on, it can do pattern recognition, it can be looking for license plates, right? I don't need to know anything about it. I don't need to know that there's an amber alert or anything. So thinking about how can data serve us in those different ways and with so much technology out there and new ways of using it,
00:07:59
Speaker
and new systems and new, you know, new technology and cars and our phones and crazy devices and sensors, you know, let's start thinking about, you know, sometimes there doesn't even need to be a human in the loop of making these things happen. So, you know, it might be database, it might be voice assistant, it might be a data service that's, you know, works with sensors on my person that help
00:08:27
Speaker
deliver messages to me in various ways.
Smart Technology in Everyday Challenges
00:08:32
Speaker
There was an incident that I was thinking about recently where I almost ran out of gas on the freeway. An incident. An incident. An incident. When the orange light comes on. That's when it's time to get gas. That is when it's time to get gas. For those of us who have grown up in cold weather climates, you fill up the tank when you get to half a tank. That's when you fill up. That's the time to fill up. Half is low.
00:08:58
Speaker
When you grow up in a snow climate, okay, go ahead, the light goes on. Yeah, so that's good. I should listen to that one. For my incident though, so I wait until my gas runs out because then you have fresh gas.
00:09:22
Speaker
The whole tank has fresh gas. Exactly. I now understand the logic. I've never understood it before. Now I get it. The whole tank is fresh and it's clean. Okay, okay. Gosh, I can't believe I'm revealing this to the world. But the problem was my light came on and I was within a half a mile of my
00:09:52
Speaker
favorite gas station, right? So I was just gonna hop on the freeway and hop off and get my gas. Well, I didn't realize it was a six car pile up on the freeway. So yeah, so but it made me think in those moments of terror of running out of gas, right behind the six car pile up. Um, um,
00:10:18
Speaker
You know, that why couldn't my car tell me, you know, it knew, it knew where I was. It knew it was low on gas. It knows that I always try to go to that gas station. It can make some inferences about like,
00:10:34
Speaker
Huh, you know, it looks like you're kind of headed toward the freeway. And I kind of know more about what's happening on the freeway than you do. So, you know, just like a heads up would have been nice. Like, hey, hey, Kim, like... See, it would have been nice. I think like a hundred miles earlier had the car been like, look, the light's got to go on, but you don't need to wait until the light goes on.
00:11:01
Speaker
No, I see what you're saying. I see what you're saying. But still, you know, I mean, I don't know about your car, but on my car, there's a little gauge and there's an F at the top and there's an E at the bottom. Okay, you're taking the long read. No, I'm taking the long read. The education route. I see what you're saying. I see what you're saying.
00:11:22
Speaker
Well, let's just take the sort of work that you guys do a periscopic. The project you did on the polar bears, the project you did on gun deaths, the project you did on salmon. I mean, those are sort of, maybe that's sort of like thinking a little more like mainstream data as it were. Like, should people who are doing that sort of work or just making, you know, charts for the report, should they be trying to think about the new technologies and where, you know, maybe that's not what they're going to be doing, but how they want to sort of in 10, 5, 10, 15 years, how they want to communicate their regular charts and their dashboards.
00:11:51
Speaker
I think so. I think that things like Alexa have a lot of utility that people may not be thinking about. It's tough. That is a tough translation to do from visual to voice. It's a completely different animal. But I think people should really start
00:12:13
Speaker
thinking about what am I trying to convey, you know, what kinds of things are people interested in, you know, like a lot of the election coverage, all of that stuff, you know, a lot of that could be voice, you know, and you just ask Alexa what's, you know, I don't even want to bring up some of the examples.
00:12:39
Speaker
But you know I think that you know polling obviously even that's on there. But I think that you know in any sort of realm that you're working in if it's something that people are looking at every day especially with dashboards those types of things are meant to be used every day.
00:12:57
Speaker
So, yeah, if you can say, you know, hey, what's, you know, if I'm getting ready for work and I just want to see like, what's trending today or what kind of anomalies are out there that I need to be prepared for as I go into work and things I can start thinking about.
00:13:11
Speaker
Absolutely. I think that we should be trying to help people because the users aren't going to come up with that themselves. We need to sort of help bridge that gap as we're trying to help them with these data services.
The Role of Critique in Data Visualization
00:13:24
Speaker
Right, right, right. Yeah, I can imagine like the Bloomberg terminal sort of thing where it's a little more voice activated, so that's six screens where you're sort of trying to...
00:13:35
Speaker
I also want to look forward a little bit and get your thoughts on critique in the field. There's more and more projects, the Makeover Monday project that Andy Cottreeve and Andy Kreebel have done over the past year sort of got a lot of attention.
00:13:50
Speaker
rightfully, you know positive attention I think rightfully so and I'm wondering what you think about the sort of current state of Critique and discussion in the field is it you think it's going down the right? Right path. There's gonna be sort of different things going on Are we not talking about the right things? I mean are we too stuck on like zero baselines and pie charts? Yeah, well that's a that's a lot um
00:14:18
Speaker
I like to narrow the questions into the biggest things. It makes it easy on my guests. Yeah, absolutely. I think there's always room for the zero baselines and the pie charts and all that stuff. I think data literacy is something that as visualizers, we don't really realize how bad it is sometimes.
00:14:43
Speaker
because we're immersed in it and we know it and we are constantly improving on it and we sometimes forget that the general public doesn't have that much of an interface with it. So always those reminders I think are always good. Always remind ourselves too that there are always people, newcomers to the field, there are always students who are trying to figure this stuff out and the more we give them in that realm, the better.
00:15:11
Speaker
The space of critiques, I know that's such a contentious space. I mean, it's tough. It's tough to get a critique, of course, or a makeover. I think that they are vastly necessary though. I think that they, one, doing a critique or makeover of the visualization, first and foremost, for the person critiquing, they really have to think through
00:15:41
Speaker
the pieces of it. Even if it's cursory, even if they don't take everything into consideration and obviously weren't a part of the full project and they don't know who the client was or any of the constraints, just going through that thought process is immensely valuable to the critiquer.
00:16:00
Speaker
Um, we do, uh, at periscopic, we do a show and tell every Monday where we bring in, um, you know, everybody's allowed to throw things into the hopper and we look at them. Uh, and so there are some examples of really great visualizations, some really bad ones, some in betweens, and we just talk about what, what's good and what's bad. And it's a really, really valuable learning experience, um, for everybody.
00:16:26
Speaker
You know, we have developers, designers, project managers, strategists, UX people, you know, every person comes at things with a different perspective. So the more you hear those different sides of the story, I think is, you know, all the better critiques, I think sometimes get a bad rap. You weren't there from the start. You don't know what the constraints are. You don't know the time, you know, time or budget constraints or the layout or any, you know, the clients was, you know, had demands or whatever.
00:16:55
Speaker
You know, there's all of those are, you know, those are valid reasons. But I don't think that we should say stop doing critiques or makeovers because, you know, the things we get out of them are so valuable. It was Jonathan Corum of the New York Times. He gave a talk at, oh, gosh, I want to say it was at Visualized.
00:17:19
Speaker
And he left this portion out of his talk, but he left it in his slides. And it was actually a critique of our gun visualization. And when I read through his slides, it was like, so thoughtful. And the ideas that he had were so interesting that we actually went back and tried them out. And I had our developer go back and implement some of his ideas. And they were really beautiful. I don't think they quite
00:17:47
Speaker
had this gave the same message so we didn't actually publish it but it was a really beautiful way of looking at it and you know opened our eyes up to you know just a different way of thinking about things so I think it's really valuable to have those out there especially for you know for students who are people who are coming to the field fresh they can you know we got we have to give people
00:18:13
Speaker
some credit and let them make their own decisions like, do I like the original or do I like the makeover or would I combine different things? It gives people another way of looking at things and to show.
00:18:25
Speaker
All of these things, there are so many choices that go into visualization. Really, you could have a million different solutions to the same problem. What I find interesting about the whole thing about critique or makeovers is they often tend to be on the final piece.
00:18:43
Speaker
And I think it'd be really interesting for people to be, I mean, so you have this process within Periscopic where you are critiquing each other, but I'm sure there are places where there's one or two people who would benefit from critique on a thing that they are working on from the, you know, have that conversation sort of wider. And, you know, I guess, you know, you just don't see that as much, I guess, because people, I mean, aside from any like data security constraints or, you know, they just don't want to put something out that's not done.
00:19:12
Speaker
Right. Like I think, you know, like your help me this blog was fantastic for that. I love that idea. So maybe it's something like that where it's more of a yeah, or maybe there's like a way to
00:19:28
Speaker
make a private version of that where you're not exposing it to the world, but you have like a private way to share this with other experts and say like, hey, is this going down the right path? I think that we've worked that way with a few clients where, so for instance, there was a client we had that was trying to build out their database capabilities in-house.
00:19:52
Speaker
And so they would come to us and just sort of vet some of their ideas early on. So we worked as a consultant in that respect to give them, you know, guidance. So I think that's valuable. I think that, you know, I think people should definitely reach out to others and people are really willing to give feedback.
00:20:15
Speaker
But the field also has, like, definite snark. I mean, the field, like, let's not fool ourselves. And I'm as guilty as this is anybody, but, like, let's not fool ourselves that we tend to be, like...
00:20:27
Speaker
what were you thinking lol emoji emoji emoji and then some poor person is crying in their office yeah yes that's true I mean I'm definitely guilty of that I mean it's
00:20:47
Speaker
There are times when you're like, really? Well, yeah, I mean, so there are times when it's blatant, like either blatantly trying to be misleading, or just blatantly, you know, there's something blatantly wrong, either purposefully or not. And so maybe some snark is okay. But I think the fact that that snark sort of
00:21:08
Speaker
seeps into a lot of the conversation, I would guess, turns a lot of people off from being like, oh, I have this thing. I'm not really sure on the right path. I'm going to put it on Twitter and see what people suggest. I would not recommend putting it on Twitter. Don't use that as you go to guidance, unless you have really thick skin. But I think that if you reach out to people professionally,
00:21:38
Speaker
and just ask them. People are much less. I mean, it's certainly true. It's much easier to be snarky on Twitter. Yeah. Well, it's certainly true in my case. I mean, one of the first projects that I did, I emailed to you. I mean, when I first started, I was like, ooh, I'm going to add. And I said, LOL. Are you kidding me? A big donut chart. That's crazy. With an emoji in the middle? No.
00:22:07
Speaker
Yeah, no. Okay, so that is probably true that the snark sort of is toned way down when it's one on one as opposed to the world at large. Right. Exactly. I think and people people are happy to help out generally unless they're very busy, in which case you shouldn't be you shouldn't take that people shouldn't take that personally, you know, and just, you know, try someone else. You know, I think reaching out to other
00:22:34
Speaker
people getting started out to is often a good way to go about it. I know that when we were starting out, we looked to our peers a lot to see, just sort of get a gut check. Yeah. Yeah. Okay. I want to talk about one more sort of future looking thing. We've talked about the future of technology, I guess, and sort of critiques and makeovers.
00:22:55
Speaker
You guys at Periscopic do a lot of work on sort of socially conscious issues, the gun vis death that everybody knows, the polar bears and the salmon thing and the terrorism thing. And I'm curious where you are standing right now, especially in a time period we have a fairly contentious presidential election. We have situations around the world like in Aleppo and Syria that are serious and often sort of horrific and depressing.
00:23:23
Speaker
serious where you sort of stand right now on the relationship between visualizing data and the emotion that comes through and talking to people and maybe showing images and just sort of curious where your thoughts are evolving or you have new thoughts on that sort of relationship between the two.
Finding Humanity and Stories in Data
00:23:40
Speaker
Right. I think so finding the humanist, the humanity
00:23:48
Speaker
The empathy for the data is something that is undercurrent with us. It's always part of our goal is to sort of reach some level of empathy. I think one of my mantras is to respect the atom of data. So whatever your individual piece of data, look at one row of your data. Is it a person? Is it an animal? Is it whatever it is?
00:24:17
Speaker
respect that thing, respect that entity and sometimes look back before the data. I think a lot of people just start with a spreadsheet or their database or whatnot and go from there and just they start with numbers and it's really devoid of any sort of humanness.
00:24:39
Speaker
So my mantra is always, you know, go back to that item of data, go back to before that. Where did that data come from? Who are we talking about? What's behind this? And really respect that. You know, I think a lot of people sort of see, you know, they're, they're communicating numbers and therefore they should be as, you know, impartial as possible.
00:25:04
Speaker
But to me, that's sort of missing the whole point of why you're visualizing anything in the first place, you know, it's sort of like the goal is to understand what it is you're communicating. And if you are putting it in a chart, then that essence should be in the chart. You know, if you're doing it with a story, you're going to use a, you know, you might use a photograph that a photograph is full of emotion, right? You're not just gonna
00:25:32
Speaker
take a photograph of a number. We have to look at other ways that people are communicating things. So like I think that journalism is a great example where people are afraid to put emotion into their visualizations, their charts, because they are meant to be seen as this sort of impartial, objective,
00:25:59
Speaker
reporting, when in fact, every article you read, it's full of stories, personal stories, personal accounts, people, images, photographs, you know, and somehow
00:26:12
Speaker
Why aren't they held to that same sort of rigor of like, well, let's take out every sort of hint of emotion or empathy. But so practically, you would say you have an article on your, you know, whatever you're in the newspaper, you write an article on Syria, write an article on Aleppo, it has the interviews, has the pictures.
00:26:30
Speaker
You're arguing on the data vis side, there's a graph that to practically sort of input the emotion or the narrative into the graph would be what? Would it be a more emotive title or annotation or is it imagery on the graph? Like practically when people are thinking about how do I take my line chart of the unemployment rate with this report? Like how do I make that, like how do I get people to connect with that more, right? Maybe the unemployment rate's a bad example, but you know what it's like.
00:26:59
Speaker
Yeah, so maybe it's not a chart about unemployment. Maybe it's something deeper. And it's something that's not I think a lot of times we get stuck with what's in front of us in the data. And we sometimes have to look at like, well, what isn't being said in this, you know, it's not just a chart of numbers of people who are unemployed, it's like, what, what does that actually mean? Does that mean more people are homeless? Does that mean more people can't afford to pay for
00:27:26
Speaker
you know, food for their kids, you know, like if there's more behind it than just a number of people unemployed, which is sort of like, if there's if there's no emotion in it, then there's no emotion in it to be had. But if there's no emotion in something that you clearly know, it's emotional. And then I think you need to dig a little deeper to find what you know, why we should care about the unemployment rate, right, you know, right.
00:27:50
Speaker
I mean, I may think being unemployed would be freaking awesome because I could just sit around and do nothing, you know, that'd be fantastic. But, you know, I don't think that's quite what they're trying to get at. Yeah, probably not. But, you know, I think you're thinking more of a vacation. Yeah. Which has its own sort of emotion with it.
00:28:15
Speaker
But yeah, I know I see what you're saying that people they're looking at the topic of what they're writing about or what they're showing, but not scratching below that. That's right. Here's a really good example. So for we recently did a project about salmon, right? And they this client came to us with tons of charts that they had already done, right? And we went through them and there was one that was just like, what?
00:28:42
Speaker
What is this? Why is this a chart? What does it mean? It just like vexed me to no end. It was about when... Oh god. So it was like, okay, these are scientists, so bear with me. Prepare yourselves, everyone. It was charting what they... I forget what they call it. It was like it was sort of a one type of fish that was
00:29:20
Speaker
of an image of a middle school cafeteria right now, by the way. He wears a leather jacket. Right, right, right. Got that cool hairdo. Cool hairdo. He's got his jeans are coughed and pinned together. Exactly. I'm going to date myself a little bit. Yeah. All right, the cool fish. Got him. The cool fish. Cool fish. So they were charting this fish, and when
00:29:40
Speaker
like the popular fish, if you will.
00:29:48
Speaker
like the time of year that this fish was swimming back to its spawning ground in this river, the whole visualization was based on this one river, right? So here's the popular fish and here's when it's swimming back to its... So you can imagine sort of this normal distribution of when, you know, around, you know, July 1st or whatever, of this fish swimming back to its spawning ground, right? So this is a popular fish. So now being the popular fish,
00:30:16
Speaker
It's also the most fished fish. So the fisher people come out and fish for this fish. So the problem is, now that's fine because this popular fish is popular because there's a lot of this fish, right? And they do what's called enhancing the population, which is fish farmers add more of this fish to the stock, right? So it's a farmed fish, basically. Very popular. People like to fish for it.
00:30:46
Speaker
So basically, by charting when this fish is swimming, it's sort of a proxy for when people come and fish, right? Where they come to fish on this river. But the problem is there are tons of other salmon that also have to swim through this river. And lots of little subspecies of salmon have to swim through this river during this fishing season.
00:31:12
Speaker
So the problem is that you have some populations of these fish that have only 500 fish left or 293 fish left, you know, like really, really tiny populations, right? So when they came to us, they had showed us a chart of just overlapping when these different fish were migrating, right? So when you look at when the fish are migrating, you're like, oh, you know,
00:31:41
Speaker
Okay, this one fish is sort of like on the tail end of this other popular fish, so whatever. It's all good. But what they didn't show in the chart was the actual numbers of these fish, right? So the popular fish is like, I don't know, 14 million or something. And then some of these other populations are like 293. They're just a little tiny blip along the bottom of this chart. When you put the actual numbers together along with the timing,
00:32:11
Speaker
you can very quickly see, oh, this fish could become extinct in one fishing season, right? So, it's a matter of taking, like, what is the goal? What am I trying to say? And this may be vastly important, you know, saving a species of fish. Like, how do I say that in an image? You know, and sometimes it's not the data that you're looking at, that's just coming straight out of your database. Sometimes you have to think about, how can I get this across to somebody?
00:32:41
Speaker
How can I show this? And that image that we made was simply like a bar chart side by side. But once you see it, you're like, oh my God.
00:32:53
Speaker
It's not like you don't have to put a dead fish up there to make it emotional. It's just seeing the numbers sometimes. It's interesting. You kind of touched on two things
Iterative Approaches to Effective Visualization
00:33:03
Speaker
there. One is the idea of digging deeper and finding the story behind the story. But you also mentioned there are two charts. I often feel like people forget that it doesn't always have to be packed into one chart. Two charts often work best because you get
00:33:18
Speaker
You get the story that you just told where you get these two aspects of things. You get the level and the change or whatever it is. Right. Yeah. You have to go through the work of doing it. I think people sometimes forget how many iterations of things it takes to get the idea. Sean Carter gave a talk a long time ago about
00:33:43
Speaker
one of the New York Times visualizations. And I forget what one it was, but it was, I think it was like many roads to the White House or something. And he showed all of the iterations that they had pushed in GitHub. And it was literally like the first three iterations were vastly different. But once they got to the third one, and it was sort of like the golden nugget, then like the rest
00:34:05
Speaker
the learning of iterations, which is like tiny little play with this, play with that, try different things within that same, those constraints. And I think that people forget like, oh, once you have the idea, you're not even close to being done yet. And now you have to do the actual work of figuring out what does convey that information in the right way, in the most impactful way. Good. Well, on that note of being impactful,
00:34:32
Speaker
pulling it all together. Gamala, thank you for coming on the show. This has been, as always, it's been fun chatting. As always, John, I wish we could do it more. This is fantastic. Thank you so much for having me on. Well, thank you for chatting with me and thanks to all of you for tuning into this week's episode. So until next time, this has been the Policy of This Podcast. Thanks again for listening.