Jaime Snyder_mixdown.mp3: Audio automatically transcribed by Sonix

Jaime Snyder_mixdown.mp3: this mp3 audio file was automatically transcribed by Sonix with the best speech-to-text algorithms. This transcript may contain errors.

Speaker1:
Welcome to Radical A.I., a podcast about technology, power, society, and what it means to be human in the age of information. We are your hosts, Dylan and Jess. We're two PhD students with different backgrounds researching A.I. and technology ethics.

Speaker2:
And welcome to the second episode of Season two. In this episode, we interview Jamie Snider about how we see, understand and track our lives through data.

Speaker1:
Jamie Snyder is an associate professor in the Information School at the University of Washington. She leads the Visualization Studies Research Studio and is also an adjunct associate professor in the UW Department of Human Centered Design and Engineering.

Speaker2:
And this is an episode that's near and dear to my heart, because in this conversation we cover some of the topics that I'm currently researching, including the visualization of mental health data. And so without further ado, we're so excited to dive into this interview with Jamie Snyder.

Speaker1:
We are on the line today with Jamie Snyder. Jamie, welcome to the show.

Speaker3:
Hello. It's really exciting to be here with you guys. I'm looking forward to the conversation.

Speaker1:
And we are really looking forward today to talk about data and personal informatics, how we make sense of data and how we see ourselves in data, which is actually our first question for you, Jamie. How do we see ourselves in data?

Speaker3:
Well, there's there's probably a lot of different ways of answering that question. But from my perspective, I think a lot about how we visually represent data through data visualizations and infographics. Really, any time that that we create a visual image as a way of of encountering, accessing, exploring data. And I think one of the the the well, three of the big questions that I often think about are what types of things are we visualizing when we say we're creating a data visualization, when we're when we're taking personal data, community data and visualizing it, what are we counting as data? And then how are we actually encoding that? How are we representing that? And you can you can think of points, dots, lines, right? But all of these visual encoding, these ways of of of graphically representing data that represents people. Right. Are ways of reducing ourselves down to these graphic components. Right. And then so how what are we visualizing about ourselves? How are we visualizing it? How are we actually translating that into a graphic form? And then what are the implications of those design choices? That's that's the bulk of my research really looks at that, that trajectory. And and so that question of how we see ourselves in data for me comes down to, well, it depends on how we represent that data that really influences what we can see about ourselves in data.

Speaker2:
And for me, who's very I'm getting a little bit more accustomed to data visualization through my own PhD program and some of our colleagues that I work with, but I'm still new to it. And so for folks who are new to data visualization, just even as a concept, can you say a bit more about what it is and maybe if there's any sort of case study or example that might help folks understand?

Speaker3:
Yeah, yeah, yeah. I think that data visualization is funny because you hear it and I think a lot of people have an immediate reaction like, oh, that's that's a data science stuff, that's statistics, that's computers. My approach, I have a background in fine art, so image making and visual representation, I come at it from a pretty different perspective than you might encounter if you went into a data science lab someplace. So for me, very, very simply, data visualization is the is any time we create a picture out of information or knowledge or what we know about the world. I've been looking at a lot of documentaries about the kind of birth of alphabets and numbers and pictograms, right? So so this this idea of of putting information into pictures is old to completely oversimplify, right? So in some sense, that's what I'm talking about when I'm talking about data visualization is how do we and when do we to what end do we create images to help us store our information, what we know about the world, access it, share it, reflect on it, analyze it. Those are all things that I think about. So. So cave paintings. Yeah. Okay, that's that's one version. But I think another another example that gets more at the heart of the questions that we, that we started out with, which is how do we see ourselves in data or all of those charts and graphs and maps that we have seen fly by in the last two years related to the pandemic and COVID and COVID transmission and COVID symptoms and COVID vaccination.

Speaker3:
Right. I think even if you somehow missed a lot of it, you probably are familiar, at least in passing with the term flatten the curve. Right. That was a direct reference to this one chart, this one graph that was made by the CDC in the beginning of the pandemic that showed what's going to happen. The trajectory of of. The impacts on humans if we don't do anything right. The steep, steep, steep increase of of transmission and of deaths, followed by a kind of a radical drop off when there was no humans left to infect. Right. It's pretty stark, right? You know, to to to think about the simplicity of that graphic and to think about the underlying story that's behind it. Right. And so when when we hear that phrase flatten the curve, if we go back all the way, all the way back to that to that one graphic, the idea was to say, hey, what do we do so that fewer people die and that the the leveling out that maybe kind of knock on wood we might be seeing now happen sooner, less abruptly. Right now, as we sit here two years in right in the midst of of yet another wave of another variant. Right. The the we often see stories in the newspaper now kind of reminding us about the people behind those numbers, the the lives behind that that flatten the curve story and how many people are actually have been impacted.

Speaker3:
So that's a great example for me. Yes. Part of it is the storytelling aspect of when we put pictures to some of these really abstract quantitative phenomena. Right. But it is also an idea of how how these representations start to. Influence what we think we know about something as big and incomprehensible as the pandemic. Right. So I think extending that, we have that flatten the curve graphic that is. So simple yet so huge. Right. In terms of the lived experience of of a global pandemic. But we also have these these daily, you know, these these graphics that are updated kind of in, quote, real time that show how many people are vaccinated in your county. How many people are getting sick in your county? How many people are getting hospitalized in your county? Right. And I know that for myself, as things have opened up at various points and I've started traveling, I start looking at where I live here in Seattle, which has had relatively high compliance and high anxiety compared to other places where I have family that I've traveled to say in Ohio. Right. And so part of my travel preparation is looking at these these maps and trying to make sense of them. And so for me, again, I go, okay, this is what it looks like here in Seattle.

Speaker3:
And my risk is X, Y, Z. I'm going to travel to Ohio. This is where that where my risk changes in these ways based on these graphs. But then when I actually get there. The kind of social and normative practices of mask wearing, social distancing, attitudes towards vaccination. Are so much more complex than those little colored squares on a map. Right. That that those maps are a way to support a narrative. But they're not telling the whole narrative. They're not telling the whole experience of what it means to. You know. Be in the social and social communities where there are these really different attitudes towards towards risk and danger and health and following the mandates and things like that. So that's my example. So when I say data visualization. Don't you? You asked me a very simple question. What is data visualization? And I went off into this rabbit hole. But to me, that's my answer. And that's going to be different from an information visualization system. Researcher Right, someone who's building tableau or building power. Bi Right. And someone like me who is a design researcher who is really using the act of visualizing data as a way to reflect on these ideas of of how we represent ourselves and how we are reduced by data, but then also how different narratives can kind of emerge from data.

Speaker1:
Yeah. So now that we've gained a better understanding of what data visualization looks like at this really large scale and for a very complex case study like COVID, I'm thinking of maybe an equally complex case study, but maybe a smaller scale on an individual level, because data visualization is, is something that's also really commonly used and seen amongst individuals for things like personal data tracking or personal informatics. And so what is personal informatics and how do we encounter that in our daily lives?

Speaker3:
Yeah, so so bringing it bringing it down. Right. I've done a lot of work related to this, this term use personal informatics. We use that word to encompass any collection of data, usually quantitative data points, but it can also be more narrative based that is intended to represent our personhood. So the electronic health record you have that's associated with your medical record, that could be considered personal informatics, we often call that personal health information. The the data in the app that you have on your phone that connects to a wearable device like an Apple Watch or a Fitbit, that is personal information, personal informatics, the amount of electricity that you use in your house, when and how you're managing that some folks also consider as personal informatics. So. It's one of those things like if I'm in a school of information science, right? And when you when I first told people that this was a field I was going to be going into, because, again, I came from the arts. So people weren't really super familiar with with information science. They would say, Oh, where are you? Where are you getting your piece? What is that? And I'd be like, Oh. Information science, information studies. And they're like, There would be a pause and they'd be like, What? Isn't that right? Like, isn't that everything? And so in some sense, when you push the edges of what personal informatics is, it is, you know, a lot of data is about ourselves, about about representing ourselves either as an individual through my my sleep data, my activity data, or as a community when we try to to look at things like trends and during the pandemic.

Speaker3:
So kind of backing up one of the the case studies that that I've worked on quite a bit looks at how people use self tracking data and that so how people use self tracking data to manage serious mental illness. Specifically looking at bipolar disorder, bipolar disorder, for those of you not super familiar with it, is a it's a chronic mental health condition that's characterized by sometimes extreme swings between manic elevated states and depressive low states. And it can be extremely challenging to to predict, to manage these swings, because they they are usually accompanied by changes in mood physiology. Sometimes attitude and medication can can help with that condition. But it generally takes a really big toll on social relationships, family relationships, career finances. Because all of these different ways that we. Make sense of the world. Right. Are deeply impacted by all of these these factors that are in constant flux or can be in constant flux. So one of the ways that people manage over the long term is, like I said, medication therapy, having a relationship with a clinician is extremely important, but also through self tracking. So understanding that. Oh. I haven't been sleeping so well. My you know, I usually when things are good, I sleep for 8 hours a day. But my sleep has been getting shorter and shorter and shorter and shorter. And now I'm only sleeping 3 hours a night. Something's going on, even if I'm not seeing the effects.

Speaker3:
Other places. That's a red flag, right? It could be. Oh, wow. You know, every week. Every day this week I've come home and there's been. 20 to 30 Amazon packages waiting for me at my door. This is an example that was given by one of my participants. Where if it's the holiday season, it's not unusual to have like a pile of packages. But every single day that amount for that person was an indicator of, of an imbalance of, of, of one of these shifts. So self tracking, when you when you think about bipolar is used to help people answer the really basic question, how am I doing today? In a context where you can't always trust yourself. To know that no one and have an answer to that question. So it's a process and. It's a process of tracking of of. Creating systematic measurements or assessments of yourself that is intended not to reduce you down into just a number, but it is intended to give you some insight into aspects of how you're doing that might be inaccessible to you at any given at any given point. So I got interested in this space because I was working with some collaborators and colleagues who were building apps to support this kind of self tracking. And I said. How are you visually representing all of that data? And they said, oh, just line graphs, charts or charts, the normal, the usual. And I said, Can I can I can I come in here and actually start talking to some of these participants? I got them to.

Speaker3:
And this is Steve Boyda and Mark MATTHEWS. Steve Boyda is at University of Colorado in Boulder now. And Mark MATTHEWS is at University College in Dublin. At the time, we were post-docs and in the interaction design lab at Cornell, and I said, Hey, you guys, can I just add a couple of questions to this big survey about people in mental health and technology use? And they said, oh, yes, sure, throw it in. So we asked people a couple of questions. We said, how do you what words do you use to describe your experience of bipolar? These are all people who had been diagnosed with bipolar. And how do you explain your your experiences to friends and family? And. The results of that. That survey were really interesting because the number of people who said, I don't I don't talk about it, I don't share my experiences because no one's going to understand was overwhelming to us. Right. The other part the other answer to that of like what are some words or images you would use to describe it were incredibly rich. They were these visual metaphors that were complex, that were mixed. So people would say, I feel like the the I feel like when I'm when I'm having a a manic or a depressive episode or a mixed episode is when you have experiences of both of these both of these states. It's like being both the ship in a storm and the storm itself. Right. So this out-of-body, decentralized experience, we also heard people talk about the disjointed nature of time that it's not, oh, I'm on a roller coaster, I'm at the top and I know that I will be going down the hill, hit a bottom, and then come back up in this linear way that when we think of a data visualization aligned chart, right, like stock market, it's it's going to go up, it's going to go down, but it's going to be linear no matter what.

Speaker3:
Right. They're like, I don't know. I don't know what's going to happen. There's no there's no predicting it. There's not an idea. And I had people kind of in one of my studies, I had them use a really simple graphic element. I said, just draw a line that represents your experience, like these these rhythms, these day to day lived experiences of like just draw it. And in a simple line, you don't need to be an artist. You don't need to be a fancy designer. Right? And the visual mechanisms and I'm going to use fancy words now, the visual mechanisms they use to encode their non-linear experiences of time. Right. We're amazingly sophisticated, right? Because they're an expert in what it feels like to have that disjointed sensation of not knowing. From one point to the other where you're going to land. Right. They were able to use drawing an image making to represent aspects of that lived experience in ways that then as researchers, we can then do the work to map to the ways the computational models of change over time that are embedded in a lot of of visualization systems.

Speaker3:
Right. So one of the. So this case study then really focused on this idea of the ways that the the the quote, normal quote, typical quote, correct ways of representing change over time. Meaning today at 10:00 in the morning, the temperature is, you know, 60 degrees Fahrenheit. In an hour, it's going to be 62 degrees Fahrenheit. In another hour, it's going to be 67 degrees Fahrenheit. And we're going to show that by these plot points. And we're going to draw a line that connects those points. And it's going to be smooth and we're going to say the day gradually got warmer. Right. That, honestly, is not useful to many of the people we've talked to who have bipolar disorder, because in between those two points. Like all sorts of things can happen. And sometimes people will describe it as like. Like, I can't even and I have these really great transcripts and drawings of videos of people drawing and they're kind of being like OC at this point in time. I felt this way, and at this point in time, I felt this other way. I can't draw what's in between. I can't actually like a line is not is not adequate. Right. So in that sense that seeing yourself in the data, right, I take that really literally like really like concretely, like seeing yourself in that line and in those points and not seeing not not feeling like it resonates, not seeing yourself. Does that make sense?

Speaker2:
Yeah, absolutely. And one thing I hear you pointing to and I feel I work with Steve Boyd as part of my research and one of the conversations that that we have, especially when we're thinking about app design is and mental health is, these are deeply individual and personal journeys that folks are on. And yet folks who are designing apps are designing for some level of universal some level of universal experience, because that's what you need to do in order to scale an app. And so my question for you is, is about that balance and not to to bring in this element, too, but especially when markets being flooded with apps for mental health and for mental health tracking, some of them that are getting the rubber stamp from, say, a CDC or a government organization that deals with health and some of them that are just the wild west of Silicon Valley apps. And so how do we make sense, I guess, of of all of those factors?

Speaker3:
Right. And I've had this question where people, especially folks in industry, are like, och, this is it's it's an interesting kind of compelling set of ideas to think about. But I got to get back to work and I got to get the app launched. So what I think is important, yes, this this work and the questions I ask and the design methods and the image making practices that I try to to use, really the goal is to try to provide a shared space in order to understand diverse lived experiences. To say we you don't need to have a PhD in technology design or human computer interaction or data science. Right. In order to. Have insights and perspectives on how you want to be represented through data. It's up to us to do a lot of the translating work and us meaning researchers, in my opinion, to do some of that translation, to be aware, to ask the questions of who are the people behind these points? Who are the people behind this, this smoothing? What are those experiences? And is it appropriate to make these design choices to reduce the data in these ways? Sometimes it is and I say this, I mean, my example like. The global effort to. Test and deploy. The vaccinations that we have seen in the last two years would not be possible without data. I'm not against data. Right. I believe in. The incredible importance of things like chemotherapy, right. We need to make generalisations at times, but we need to make sure that we are informed and we understand what choices we are making when we create those models, when we are creating strictly computational models. I think sometimes those choices are masked, are made more opaque, are not really as transparent as they could be for the users, but also for the people who are sitting in the labs.

Speaker3:
Creating those algorithms and those models, right? I think in that sense, the power of the visual is that it analyzes it, gives material form it, and I'm doing this with my hands. I always do this like it takes it from your head out right into this shared space. Right. And some of my my earlier work and my and my doctoral studies had to do with what does it mean to actually make a drawing, to put a mark on a page when you're having a face to face conversation with someone? So I think that the for me, the impact of of the visual. Many times, oh, pictures are pictures worth 1000 words or just show me a picture and I'll get pictures. The images, data visualizations are intuitive. Right. I push back on all of those ideas. Like, I mean, I'm sure we've all seen a social network diagram. Those are the ones with the nodes and the lines, and they look like giant cobwebs, like layers and layers of cobwebs. And, you know, you say, Oh, see, it's evident from this visualization. And you're like, No, it's not like I study this stuff every day and I have no idea what that means. Right. So it's not just the images are accessible. I think it's that image making plays a specific role in our communication dynamics that we can leverage and we can use. So yeah, I think that a lot of the work I do does have implications for more generalizable visualization system design. I also think it is just as important that it provides a means, a method, a way of having conversations that get at things that might go unarticulated or unsaid, using other ways of communicating.

Speaker1:
Something that we've seen on this show and also just witnessed in the tech space is that there's this tension between scalability and locality. And when we talk about these really sensitive topics that require this really local, individualistic perspective on what real human experiences are, and we attempt to scale those through technology design, it's sort of bastardize all the work that was done to attempt to encapsulate what these individuals are experiencing. And I imagine in this scenario there might be something similar happening. If we do have a lot of apps coming out about health tracking and behavioral tracking, and I'm wondering what the standards are and what generalizations have been made in health tracking apps just generally right now. How have they attempted to standardize and scale some of these more local experiences?

Speaker3:
Well, you know, it's funny, because now I haven't I haven't been to some of the conferences in a couple of years for obvious reasons. But but have in some ways, the visualization industry is a is a field site for me. So I've heard academics talk about some of these issues. I've heard industry folks talk about these these issues or designers talk about them. And I think that. There's this. There's a tension. We use this word tension. Right. That's a nice academic word to tension just means, like things don't fit quite right together. Right. And sometimes. When, when, when things don't align, they don't fit in. We can't kind of tell a story that encompasses all the parts. Right. We get frustrated and we're just like, Oh, right. So sometimes that's what happens with some of these things. We can't make everything individual for everybody, right? Yeah, that's a nice idea. But I need to get back to my job. I need to. To create something that will work for most people. Not all people. Right. That's not just unique to visualization. That is a problem with technology, with design in general. Right. Do you make a piece of clothing? That fits me perfectly. Right. And then it won't fit anybody else in the world or it won't fit me next week or me next year when my body has changed. Or do you create something that will fit most people? Okay.

Speaker3:
Right. And I think that in some ways. There. I mean, this is this is a huge question, right? I mean, it's about accessibility. It's about universal universal design. It's about power. And for me, it's about our models. It's about how we are creating a model of what the human form is, how we're creating a model of how data looks. That again, it's about being aware of the design choices and the trade offs. If I say I am not going to make one off shoes that so that just you get and you get the perfect pair of shoe for you that will fit no other person. Right. I instead I want to make shoes for everybody. I want to make sure that everybody who needs a pair of sneakers get something on their feet. Those are two really different goals and to really different intentions. And I think that that is something I talk to my students about a lot. But what are your values? What's important to you? What are you trying to. Who are you trying to support? Who, what what compromises are you making and why? Right. Again, I want to go back to those questions. What data are we are we choosing to represent? What data are we saying is the most important? Is it my minute by minute sleep data from last night for me, Jamie Snyder? Or is it to look at at a higher level the sleep data of the whole city of Seattle over the last two years to see what impacts have been felt by the community in terms of anxiety, mental health.

Speaker3:
Physical fate. So, again, you know, I'm probably maddeningly agnostic in a way. Right. But my my rallying cry is just recognizing that we are making choices and making sure that the choices we're making are aligned with our values. And so. It's for me, my background, my tools, image making, representing what we know about the world, what we experience about the world in a visual form is a way I can help that helps me understand people's values, and then I can kind of expand that, have that inform the design of technologies. And I work in really applied settings. I have some fantastic collaborators both at Seattle Children's Hospital and at University of Washington at the Brite Center, which is a behavioral research center. You know, they are they're doing amazing work related to clinical care and collaborative communication among patients and their families and providers. And I'm just to me, I get very excited when I think about the perspective I bring and this idea of using. Image making as a way to understand values and models of the world, how that can influence some of these more applied projects.

Speaker2:
One thing I think I hear you saying as well is that there's also there's a social dynamic to this, that it's not just the technical how do we do this? It's what is the context that surrounds us. And one thing that it makes me think of is data literacy and how there's this know you can show ten people the same graph and people aren't going to interpret it the same way. And it's not necessarily in our education systems, depending on where you're looking or in terms of power. It's accessible to certain groups of the population and not others. And I'm curious your perspective of what the role of these social systems are, and then how can we make data more accessible to more people?

Speaker3:
Yeah, yeah. I think that's a really great thread to pull on from all of this and something that I do I think about a lot there. There is an area of work, visual literacy and having the the the privilege of of being able to interpret images. Now, there's a version of that. That is when you see a high renaissance painting, you can tell the story of the figures being depicted and be able to deconstruct the perspective that is being used. Right. That's one idea of visual literacy. There is an idea that when you see a chart or a map or a graph in the newspaper, you can look at it and identify whether it's reliable, it's trustworthy, whether it's valid. I think that there are other types of literacies that are equally important that I refer to. Again, another another academic term is vernacular literacy. Right. Vernacular just means local. It means situated. It means what you know and how you know it. Based on your. Personal your your community experiences. So an area that I'm thinking about a lot has to do with environmental data and how my understanding of how my neighborhood is doing has a lot to do with what I encounter in my backyard and that I garden. So how I pay attention to when things are blooming, how much it's raining, what's growing, what's not growing right? That's a great example of a vernacular understanding.

Speaker3:
Then I can go online and I can I can look and I see. City of Seattle. What are people seeing like? Example, my Rosemary Bush completely died this winter. Right? And so I go, Oh, what did I do to it? How did I destroy it? What? It was the old and beautiful and I loved it. It was here when I moved into my house. What did I do? And then I go. Let me ask let me put my feelers out into my community. Let me go to the garden center. Let me look at the newspaper online and say, are other people having this experience? I start to kind of cross check and collaborate or corroborate my experiences. This is still all anecdotal. It's still vernacular. It's still based in my own experience. But by doing that, I can start to learn. Quite a few people seem to be saying that their Rosemarie didn't make it through the winter, but I also learned that Rosemary's only live a certain Rosemary Bush only live a certain length. So all of this is me gathering that information not as an information scientist, not as a designer or not, but but as somebody who is living these experiences. And then I can go about addressing that. So vernacular practices lead from those that vernacular knowledge, right? To say, what have I got? What do I need and what have I got in order to meet that need? What are my resources around? Right.

Speaker3:
So when you think of literacy in these ways, what do I have available to me to make sense of the world that I'm seeing? So data literacies that are not to me. How do I carve out time to do a six month study course on statistics? How do I learn, ah, the program, the open source programming platform, so that I can do my own statistical analysis of all the public health data I can get my hands on. To me, that's not that's not the kind of literacy that I'm interested in more. I'm interested in saying, how do we take these again, these ways that people go about evaluating and assessing how they're doing and how their community is doing and how their families are doing on any given day and incorporate that into that scaling up. You guys are talking about scale to scale that up and integrate it into these higher level models of of what's happening in the world. Great. So it's not. I do want to also put out another 25 cent word of epistemic burden. Right. That is a term that is used to talk about the. The the. The weight of So how do I say it? I think of it in terms as a researcher when I go into a community and I say, Hey, you guys join my study, show up at a certain place at a certain time, and here's a consent form.

Speaker3:
I am already putting way more burden on you than I am putting on me because I know where the place is. I know what a consent form is. I know what I'm going to ask you to do and why. And you are just showing up with this high cognitive load of like, what is she talking about? What does she want me to do? What if I do it wrong? Where am I? How long is this going to take? Right. And so that that idea of of. The epistemic burden of data practices and seeing ourselves in data. I think is a really important thing for us to be considering. We talk about personal informatics, even using that term, right. Even saying, do we see ourselves in data? You know, it's beside the point in some ways. For me, I'm really interested in. Before we even say this is a data set. This is data. I want to go back a step because as soon as we call it data, we set up that Gulf feeling. You're saying, I don't know what this data visualization thing is. I don't know what this data is. If I ask when I when I was was working with folks who have been diagnosed with bipolar disorder.

Speaker3:
I specifically and one draft of the questions what do you do to self track. And after testing that I don't self track I don't self track. I don't do it. I don't do it. If I change the phrasing of that to How do you check in with yourself every day? How do you know how you're doing? Then people go, Oh, well, I look back at the text messages I sent the night before. I look back at how much I've slept. I look at my bank balance, I look at my calendar to see how many, how many times that I've been able to make my appointments over the last month. Right. So I think that when we talk about literacies. That willingness to. Use the tools that we have of categorizing, of recognizing, of naming, of describing. These are important research tools, but not not letting that so put a blinders on us to what we're willing to see as we go out into the world and we start encountering lived experiences. It means that things get really gray and messy and but that's the space I like and hopefully can can create a bridge again, do that work of translating those lived experiences to be to help inform some of those more. I'm going to use the word reductive, but not in a negative way. Processes that are needed in order for. Other data practices to to succeed.

Speaker1:
Absolutely. And for those bridge builders who are listening to this interview, who are maybe interested in understanding better how they are represented through data or interested in improving data literacy or interested in personal informatics, any of the topics that we brought up today, where is one of the best places or several of the best places for them to go if they're interested in these topics?

Speaker3:
We're really fortunate to to be working at a time where there is some amazing work being done in data studies and the Ethics of data podcast. Like, like what you guys have been working on, you all have been working on Rupert Benjamin, Virginia Eubanks, you know, these, these really smart folks who are doing the hard work of breaking down these practices. To me, I think of everything as a practice is just a way of saying what people do with what under what circumstances write to help us start to to see where this data again, the choices that are being made on many different levels, that produces what we call data, what we call knowledge, what we think we know about what we think we can represent about ourselves. I think that that's so so there's those are great resources. I am a huge advocate of making as a way of of encountering people, places and situations. So one of the the. I think. Taking taking some time to to. Learn about how data is visually represented. Just some of those tools. One of the things I will say that I. I teach a visualization design class, and I'd love to teach that class as an entry way into programming. I know enough programming to know how much I don't know. Like I would never describe myself as a programmer. Right? But I teach a visualization design class and I have taught it for.

Speaker3:
Yeah, probably over a decade at this point. And the idea is that it gives people a framework for going from writing the code, writing these characters, and then hitting run, hitting, compile and seeing it turn into a picture and seeing that, making that encoding, launching that like themselves through, through what they're doing. And I think that I try to give students opportunities to use a couple of different systems for doing that. Like, I'm sure many of the people listening to this have used Excel. Maybe they've used something like Tableau. We, we do really hands on work with some with I mentioned ah, the programming, the open source programming platform. Ah, an adobe illustrator. This is not sponsored by either one of those, those, those tools, but the really different systems for visually encoding going from, you know, beatbox and bloops right to pictures. And I think that the more you can kind of play around with that if you're interested in that, even on those basic level, I mean, like there's a whole like hello world tutorials for any programming language in the world, right? Even at that very basic level, I think understanding that translation that how that gets transposed how how how data this quantification. I slept for 9.7 hours last night. I slept for 8.7. This is me being aspirational. Believe me, I don't say that long.

Speaker3:
But how? What happens when we convert that into a picture and being reflective about that I think is okay. So so I'm always an advocate of getting your hands dirty and getting in there and starting to mess around with some of that stuff and and I think are in Illustrator are great places to start. There's a book called. Visualize that I think by Nathaniel you guys can edit that if if it's not correct it's it's not a new book but it's a book I use in my classes. It's Nathaniel is a really fantastic information designer, visualization designer who has worked for The New York Times. And it's a series of exercises that takes you through some of these practices of of representation and modeling and within our and illustrator. And that's a great book to just play around with and look at it. It has some of this design language in it, but also some of these practical technical activities that I mean, don't get me wrong, they're wicked fun, right? They're like I said, writing a little a little bit of script and then turning that into a picture is is really exciting. You know, I get all into the weeds of like, why is it problematic? And but, but there's a power to that basic act that I try to stay as true to as I can.

Speaker2:
Jamie, thank you for naming some of those resources. Unfortunately, we are at time for this conversation, but thank you so much for joining us in general today. It's been a.

Speaker3:
Pleasure. Yeah, this has been wonderful. Thank you, guys.

Speaker1:
We want to thank Jamie again for joining us today for this wonderful conversation. And as usual, it's time for our debrief of our immediate reactions and takeaways. So, Dylan, I know that this research is near and dear to your heart. What are some of your immediate reactions?

Speaker2:
Yeah, as you just mentioned and as I mentioned at the top of the episode, what one of the things I'm working on right now is visualization work and also personal health informatics work on mental health, especially around suicide, bereavement and suicidality. And one of the things that I'm looking at and probably the questions behind the questions for me today around app development and around the market and regulation and all of that stuff, is that one of the things I'm interested in is how do we make mental health tracking apps more accessible and also how do we streamline their effectiveness? Because right now there are a lot of apps out there partially to either make a quick buck or just because there is an open research space there or just space there in general in the market. And it's unclear how actually effective or accessible or helpful in general those kind of apps are. And so I think even in our community and the human computer interaction community and also in the industry community. For me, a huge question is, well, how do we do this? Well, in a way that actually encapsulates the human experience, the whole human experience, and doesn't make assumptions about what that human experience is, so that these apps and other technologies can actually help people and that our research takes that into account. So I think that's the big thing that this interview gave me some tools to begin to think about those questions a little bit more directly. I love what Jamie does and some of her qualitative research, really thinking through that design lens to be like, Well, okay, let's not just do like a spreadsheet or just like a survey in the same way, but like let's take a line and draw out your life over the past few years using that line and those kind of, I think, real innovations drawing from that design background. But just what are you thinking about right now?

Speaker1:
Honestly, my takeaways were pretty similar to yours, and I totally agree that line visualization use case was fascinating and actually it made me think about what my life would look like as a line. And I think around like deadlines and finals, it might just look like a scribble PhD students will get that. But I also I very much agreed about the speaking about apps and how this sort of research is implemented in real technology. This is something that I've been thinking about a lot in the last year in general, because I've spent half of this last year doing internships in industry. So I've been thinking about how we apply research to real products, and especially when it comes to ethics, which part of ethics research is actually making it into an online live setting in front of real users and to shift features? And when we were talking about standardization with Jamie, I was just reminded of I think she actually might have said this in one way or another, this, this idea that there is no perfect solution to anything when it comes to ethics. That's just like the nature of ethics and doing responsible tech work.

Speaker1:
There will never be a solution that works for everyone or that benefits everyone. But there is a way to design technologies that are much more informed and that are much more transparent about why they chose to design or to be designed in the way that they were. And I think this is totally applicable to this use case where we have a standard about data visualization that could be used in these kind of high stakes scenarios where people are using it for really important mental health tasks like self tracking in the bipolar use case. And I'm imagining that there's never going to be a design that makes everyone happy who experiences bipolar. But there's probably better ways to be more transparent in those apps about the design and to try to gather this information in a way that's human centered and actually working with real users rather than, as Jamie said, making assumptions about what those users actually need. So that's sort of where my mind was going. It seems like it's a little bit similar to you. We're thinking about like how this research is applied in the real world.

Speaker2:
Yeah. And I think accessibility is a theme that was prevalent throughout. And for me, I think this was one of my last questions is around data literacy because just I think you and I, we are steeped in data. We're trained to read data and to navigate how to collect data and then how to present it in various ways, whether it's industry for you and some academic research for me right now and some of the projects that I'm working on. And so for me, like data literacy is, is huge because sometimes I think we take for granted not just you and I, but bigger communities in research. We take for granted that people can read these charts, but really that that power issue of who gets educated and how to read charts, and then are these charts actually being used and understood by people who need them most? And that, I think, is a very live question and a very thorny topic of, well, how do we how do we solve that? And so I think you're right, there aren't perfect solutions to that. But I do think that we as a community need to be a little bit more intentional in understanding, even on like the user experience perspective like. Are these useful, right? Like are these useful in the first place? And these being like data visualizations but also apps and other things that are in the milieu of some of the topics that we talked about today.

Speaker3:
Is that I.

Speaker1:
Pronounced that word.

Speaker2:
I don't know, I don't speak French. I'm a Spanish speaker. I, I just like to say it like that.

Speaker1:
Yeah, no, I completely agree. I mean, data literacy is something that I often take for granted. And you just brought me back to this, like, PTSD space after taking the act where I discovered that the entire science section of that standardized test was just interpreting visualizations of data and charts and graphs, which is honestly like really awesome that I had the opportunity to learn that at a young age to take that test. But also like that's definitely not something that is part of standard curriculum around the world. And a lot of people are not trained in any way to interpret data, whether it's through visual visualization or just in general. And one final thought that I really I did want to say that was just like an important takeaway from this, too. That is lingering for me is Jamie's thoughts about how data is a reduction of our selves, our humanness. And I could talk about this for hours. I'm not going to go in depth, but I'll just say at a high level, I'm thinking about how data is. So it is so incredibly difficult to attempt to encapsulate the complexity of humanity through a data point and through a number. And I'm just going to leave it at that, because I could go down that rabbit hole for so long. But I think that we're at time. So for more information on today's show, please visit the episode page at Radical AI dot org.

Speaker2:
And as always, if you enjoyed this episode, we invite you to subscribe, rate and review the show on iTunes or your favorite pod catcher. We especially right now invite you to share our episodes because the Twitter algorithm, it doesn't seem to be liking us right now in our sharing. So share our Twitter as well, but really share those episodes. I think we mentioned this often, but word of mouth is the way that these episodes get shared. So please do share them with your friends and colleagues, anyone who you think might be interested in the content, our guests or I guess in. In us in the podcast generally great. Catch our regularly scheduled episodes the last Wednesday of every month with possibly some bonus episodes in between. And again, join our conversation on Twitter at Radical iPod and as always, stay radical.

Sonix is the world’s most advanced automated transcription, translation, and subtitling platform. Fast, accurate, and affordable.

Automatically convert your mp3 files to text (txt file), Microsoft Word (docx file), and SubRip Subtitle (srt file) in minutes.

Sonix has many features that you'd love including automated translation, upload many different filetypes, powerful integrations and APIs, automated subtitles, and easily transcribe your Zoom meetings. Try Sonix for free today.