Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Episode #184: IEEEVIS Recap
205 Plays4 years ago
Alvitta Ottley and Robert Kosara visit the podcast to talk about the 2020 IEEEVIS virtual conference. We discuss favorite sessions and papers, and our take on the virtual experience.
The post Episode #184: IEEEVIS Recap appeared first on PolicyViz.
Recommended
Transcript
Introduction and Special Episode Overview
00:00:15
Speaker
Welcome back to the PolicyBiz Podcast. I'm your host, John Schwabish. Now you may be thinking to yourself, wait, why is there a podcast this week? There's usually one every other week. And that's true, but I wanted to bring you a special podcast episode because I reached out to Robert Casara from Tableau Software and Elvita Atli from Washington University in St. Louis to talk with me about the recent IEEE conference that took place virtually this year.
00:00:43
Speaker
Obviously, because of the pandemic and I wanted to try to get that episode up as quickly as possible to ride on the coattails of what I'm sure for many people is excitement about the research and the conversations they had a couple of weeks ago during the IEEE conference.
00:00:59
Speaker
So I enjoyed the conference this year, even though it was remote, even though it was virtual and I couldn't stand next to people and talk to them about data visualization. I was also lucky enough to be able to present two of my recent papers, one on teaching data visualization to kids and another one on racial equity awareness in data visualization.
Virtual Conference Highlights and Setup
00:01:18
Speaker
But I saw a lot of other great talks, a lot of other great papers, as did Robert and Alveda.
00:01:23
Speaker
And in this week's episode of the show, we just go through them, talk about our favorite sessions, talk about our favorite papers, and talk about the virtual conference and how it was set up and what we liked about it and maybe what we didn't like about it. So I hope you'll enjoy this week's episode of the podcast. And here is my talk with Robert and Elvita.
Introducing Guests and Their Backgrounds
00:01:47
Speaker
Elvita, Robert, welcome to the show. How are you both? We're doing great. Doing well.
00:01:53
Speaker
Good. I know we're in like a weird time of year.
00:01:56
Speaker
weird world that we're in. So we can just focus on the weather. I think the weather's the way to focus. So thanks for coming on. I'm excited to talk about IEEE, obviously a unique experience this year. We're all in person, but a lot of great content I thought and some great sessions. So I want to get to that and also talk about the virtual nature of the entire thing. But before we get to all that, maybe we could just have each of you talk a little bit about your background, where you're coming from, and maybe also talk about
00:02:26
Speaker
When you're coming into the conference, where you're hoping to sort of go, what you're hoping to learn about your keen interests for this year. So, Alvida, do you want to go first and get us started? Sure. Yes. Thanks for having me. I'm really happy to have this conversation with you. Hi everyone. My name is Alvida Osley and I'm an assistant professor in the department of computer science and engineering at Washington university in St. Louis.
00:02:52
Speaker
I'm also the director of the visual data analysis group there, and we're a group of five researchers. Our research falls in the area of visual analytics. So in general, our research aims to understand how visualizations can support reasoning and problem solving and decision making. We look at things like individual differences, like personality traits, cognitive abilities, and we inspect how these might impact analysis workflows.
00:03:19
Speaker
And finally, we apply machine learning techniques to model and predict people's interactions with interfaces. So this was my supposedly 10th year at Viz. So I've been at this for quite a bit of time now. And I guess going in, my expectation was I actually didn't know what to expect. This was a virtual conference. It never happened before like this. Usually at Viz, I'm excited to see the people that I can only see once per year.
00:03:49
Speaker
Right. Um, but this one, I just went in, I canceled all of my meetings. I set up a out of office and I just attended the virtual phase and I was really excited for it to see how it played out. Nice. Um, okay. So we'll get into that in a second. Uh, Robert, you want to tell folks a little bit about yourself? Sure.
00:04:12
Speaker
Hi, I'm Robert. I'm a research scientist at Tableau Software, now a part of Salesforce. I've been doing this for a while. I've been at Tableau now for over eight years, and this was actually my 20th VIS conference. My first one was in 2001.
Virtual Conference Experience and Challenges
00:04:28
Speaker
And my research interests and also what I'm interested in, I guess, at VIS are mostly about the presentation of data, the communication of data, what people like to call storytelling, which is not
00:04:41
Speaker
actually not a term that people really use anymore, but and I also just kind of hate it, but it's just an interest of mine is like, what, how do you get people to understand the data beyond? So the basic charts and stuff like that. I also run a little YouTube channel, maybe I should mention that.
00:04:58
Speaker
since talking about myself as well here. So that's something I'm doing where I'm trying to show people how visualization works in a more sort of like overall way. So that's kind of my thinking about it. It's not just like the little things, like how well can we see this chart versus that chart, but how does it work? What's the thing that's making it tick?
00:05:16
Speaker
And so my hope was to see more of the sort of communication aspect of it, how do you get things across, or whether there was some work on that. And then of course, as Elvida was saying, to see the people I know at the conference, which didn't really work out that well this year. But that's really the main reason really to go to the conference, because that's where you get to hang out and chat and start new collaborations.
00:05:40
Speaker
Right. So we want to talk a little bit about the virtual setup and what you thought of it. Maybe Alveda, do you want to describe how you, I love that you canceled all your meetings and turn your ad office on. That seems luxurious to me right now, looking back. But do you want to talk about how you sort of navigated through using the virtual system they had set up?
00:06:03
Speaker
Sure. I really like the virtual system this year. So typically, Vis doesn't really have an interface for you to see all the talks and interact with them. So this was very much new and I really liked that.
00:06:19
Speaker
went into my office for the first time in a long time, largely because it was on the weekend when it started and I have a five-year-old, so I was trying not to get zoom-bombed by my five-year-old. And so I had my three monitors that were set up in the office and felt like I was in a studio. I had to get a good camera and a good mic and everything like that. So it's a very different experience overall.
00:06:43
Speaker
But I really liked it. I liked that I could just focus on the talks and I was there in person and present. Whereas in previous years, I would probably be in the hallway having coffee with someone, which is still equally good. Right. Robert, what about you? Did you have the same feeling?
00:07:04
Speaker
So I actually had pretty much the opposite experience. It was well done. I mean, they did what they could, given the fact that it was online. The whole streaming went pretty well. There were some awkward pauses sometimes between the talks and the Q&As, but other than that, it actually worked pretty well for the most part.
00:07:23
Speaker
I got a lot more distracted though by the fact that I was at home and there's lots of things around me here that are shiny and can catch my interest. And so I find it quite difficult actually to focus on the talks. And then there was Discord, which is this chat system, which is a bit like Slack, that was the back channel. And sometimes the discussions in Discord became so interesting that I wasn't paying attention to talks anymore.
00:07:48
Speaker
So I actually missed a couple of talks just because of that because they were sort of running at the same time, but I didn't really pay any attention to them because we were chatting on Discord. And Discord was also a bit of a problem because it doesn't have threads like Slack does. So it was really hard to follow sometimes who was responding to what and what they were talking about. So it took a lot more attention.
00:08:08
Speaker
Then maybe it should have, but it was also really interesting because there were lots of interesting things that were happening that were adding to the talk. So it was, I thought it was both good and bad. It's hard for me to say whether it was net positive or negative. I certainly got a lot out of it too. Yeah.
00:08:23
Speaker
I guess the reason why I think for me it was net positive. I'm also looking at it from my role as the co-chair of Inclusivity and Diversity at Viz. And the virtual conference solved a lot of the problems that we've had for many years in terms of bringing in new people.
00:08:43
Speaker
to come to the conference and the Discord chat, even though it was a little bit cumbersome, it solved the problem that lots of newcomers had at Fizz where you're in a room and you really want to ask questions, but it's really intimidating to stand up in front of a room of hundreds of people and ask a question.
00:09:02
Speaker
So that made their presenters a little bit more accessible to newcomers and then we had videos and closed captioning. So I just felt overall it made the conference a little bit more accessible. I'm hoping that we can take some of these if in the future, if we get back to an in-person, whenever that should be. Yeah, I think that's a great point, especially because in my experience at VisN, I've only been three times maybe.
00:09:28
Speaker
You know, a lot of the online conversation happens on Twitter, and I think you have the same kind of, for some people, the reluctance to have conversation on Twitter, because it is so public. Whereas the Discord seemed to really pick up the pace, and there was a lot going on in Discord.
00:09:42
Speaker
Yeah, definitely. And I think to lead this point about asking questions, I saw this a few times, people saying that they felt much more comfortable asking questions on Discord and wouldn't have gotten up to step up front of the microphone in the room there. So that's certainly a big plus for sure.
00:09:59
Speaker
Yeah. And plus I saw other people answering questions with other links. So that's the other thing, you know, I say, Oh, I have this other paper and, you know, I can give you the link and here it is. So, um, yeah. Okay. So let's talk about, um, content. Um, so do we want to start with the viscom session? So Robert, this is one that you've been co-organizing for a
VisCom Session and Communication in Visualization
00:10:20
Speaker
while. Do you want to talk about the setup and the goal of viscom? And we could talk about that session a little bit.
00:10:25
Speaker
Sure. Yeah. So Viscom, Ben Watson and I started that three years ago, I think. This was the third time this year. And this year also Elvida was actually one of the co-chairs that was Ben.
00:10:40
Speaker
Adriana, Arcea, Alveda, and myself, and so Ben and I are the outgoing chairs, so we're not going to be doing this going further. But the idea was to bring visualization for communication to the Viz conference, because there hasn't been a whole lot of that. There hasn't been much in terms of like,
00:10:58
Speaker
communicative uses of visualization at the research level, which might seem strange to people who are doing this out in like journalism and design and fields like that, because to a lot of people outside of academia, visualization is all about communication. But within academia, it's all about data analysis and insights and all this stuff. So it's not very much focused on the communication part of it.
00:11:20
Speaker
So that was really the goal was to do more of that work and also bring in the people doing that work. So the crypto journalists, the people working for NGOs, the people teaching with visualization, all kinds of things that we hadn't seen all that much of before in visualization. And then if you want me to talk about the actual session this year. Yeah.
00:11:44
Speaker
And so this year what we had, we had two sessions. One was all about health and mostly about COVID-19, as you can imagine. There were quite a few papers talking about different ways of communicating information about COVID. There were also a few people talking about sort of like the
00:12:02
Speaker
the thinking behind it and the ethics, perhaps, of all of this kind of communication, which is also getting more interesting again right now during the aftermath of the election, where people are asking, should we be talking about all these numbers and does it actually help to do that?
00:12:18
Speaker
And then the second session was more broadly about the public, about how how weather gets communicated. So things like hurricanes and things like that. And a bit more like meta points about how visualization works for communication. Right.
Bridging Academia and Practice in Visualization
00:12:37
Speaker
Do you think that more academic data is researchers that attend the
00:12:45
Speaker
full conference, need to attend workshops or sessions like VizCom more, like there needs to be more of the bridging between what seems to be more of the practitioner crowd with the academic crowd. Yeah, yeah, for sure. I mean, there's certainly, and this is being done not just by us through VizCom, there's also the Believe Workshop, which is trying to do that. And I haven't actually watched many of the talks because they were at the same time as VizCom, but that's, there's a lot of good stuff that they're doing and have been doing for a number of years.
00:13:14
Speaker
There's the vision practice sessions that bring in practitioners as well. And they all have different focuses, focus points, I guess, or, or, or kind of directions. Some of them are more about like what's visualization use in business versus what about
00:13:29
Speaker
communication again for viscom. I believe this year, I think was talking a lot about COVID data as well. So public health, I think is a big topic just in general, because it's very important to get information out to people and do that in a way that's ethically responsible and understandable and actionable. And you know, there's lots of questions there that are hard to study in many cases, and that really aren't getting asked enough, I think, in the main sort of conference tracks.
00:13:59
Speaker
I would definitely agree with that and I guess one of the things that has really brought that to the forefront of my attention is I attended a conference on Saturday. So this is right after the visualization conference and it was hosted by a researcher called Veena Das and she's an anthropologist at
00:14:19
Speaker
John Hopkins University, and it was on the topic of concepts and data. And so this workshop brought together philosophers, anthropologists, political scientists, and so
Ethical Aspects of Data Visualization
00:14:30
Speaker
forth. But it was a fascinating conversation about the different ways that we engage with data and the new problems that this might be creating. And it was really great to just see a different perspective and
00:14:42
Speaker
Particularly here, people who are always thinking about how human behavior and the impact of different stimuli on human behavior, but they're looking at visualization and how visualization might be causing lots of different disruptions. And so, overall, that's one of the goals of VSCOM to think about
00:15:02
Speaker
how visualization is being used for communication. To some extent, I do believe that the visualization community needs to be a little bit more introspective. Especially now with COVID-19 and the elections and the way we're representing data, we do need to talk about how our visualizations are impacting human behavior and society.
00:15:27
Speaker
There was also an interesting conversation, I think it might've been during the viscom meetup that we had a couple of days later about the intersection of qualitative folks and quantitative folks. And that was an interesting conversation I thought as well, and relates to Avita what you were just mentioning about sort of other fields and their approaches to data. And I thought that was an interesting conversation about how perhaps
00:15:54
Speaker
Many of us are trained in quantitative methods, but not taking qualitative methods or don't know how to do them or not serious about them or what have you.
00:16:05
Speaker
I agree. And we are lucky enough. I saw this at Viz this year. There were a few papers that included some qualitative analysis. They were largely short papers. But I would like to see longer papers in this. And you're right. There is, I don't know if it's a stigma. I don't know why we're not doing it. But and maybe it's just a selection bias where
00:16:29
Speaker
We are trained as engineers, most of the people in the visualization conference. We are computer scientists and we don't really talk to people, we're introverts.
00:16:42
Speaker
Yeah, I mean, it sounds right to me. We could go back a little bit if we want to talk about some of the virtual parties. Actually, it might be an interesting thing to tell some folks about because there were a lot of different virtual tools that were used as folks who listened to the show are probably not surprised. I do not consider myself an introvert. So I found them not quite as satisfying. Well, probably no one's found them as satisfying as being there in person, but I'm curious.
00:17:07
Speaker
from your perspectives, like you both attended a few of these parties and events. Like how did you feel about the interaction in the virtual world? So I didn't actually attend any of the parties, but I did hang out a few times in this hallway section of Discord, which is the idea that you have. So Discord is not just text, but it's also video and you can have a number of people talk to each other at the same time.
00:17:31
Speaker
And so they had set up these hallway rooms where you could see who was there. And so you could basically join them based on, because you could see who was already there. So you could see whether those are people you knew or not. And I did this a few times, but I found it a lot more challenging than, certainly than in person, because after you get about maybe five people or so,
00:17:49
Speaker
It gets quite difficult to actually really have a conversation when you're not in the same space because you can't read each other's byte language. There's lag, so it's easier to talk over each other. And the other thing that I think also happened this year, so all the other things that virtual is good for in terms of inclusivity, the one thing that it's really bad for is to make new connections.
00:18:11
Speaker
So the usual thing that you do as a professor is you would take your students and walk them around and introduce them to people that have been in the field for a while. And I think that happened just very rarely this year, because it's just that sort of serendipity or just that possibility just wasn't there, wasn't nearly as easy to make happen as it would be in, you know, meat space.
00:18:34
Speaker
Right. Avita, did you have a favorite part of that aspect of the conference of the casual conversations of the party apps or anything like that? Oh, that was my least favorite.
00:18:47
Speaker
So I did not do the hallway thing because I think it's just weird standing by yourself in a virtual hallway or just like jumping into someone's conversation. But I also think that's weird also in person sometimes. So that's a tricky thing to navigate. But I did go to one of the parties and we use this platform called Gathertown.
00:19:10
Speaker
which for people who don't really know what it is, I didn't know what it is at all either. But you have this little cartoon character that you can kind of dress up. And I've only been to one part, I only went to one party, so I don't know how much flexibility there is in terms of the space that people can create. But you essentially appear in a space and it was like a room and
00:19:37
Speaker
You're just there, and there are people there, and you then have to have the awkward conversation. But what I like about Gathertown is that you can just walk away. So basically, if someone's boring, and you don't want to talk to them, in real life, you can't just walk away. But in the virtual world, you could just say, I'm gone, and maybe their feelings aren't as hurt.
00:20:13
Speaker
I think that's right. But we were also in this room where, and I had a group of people that I knew and we decided, you know what, we don't really want people just walking up and getting into a conversation because at least when you're in a physical space, you have these physical cues that someone is coming. You can hear them, you can see them, but in this virtual space, someone just shows up. And so like there was a beach and we all ran toward the beach, but it was,
00:20:31
Speaker
Yeah, it's like, oops! I think there is a way to also respawn yourself, so like...
00:20:42
Speaker
I want to make sure that we let people know, because I also didn't know about Gathertown, but the space looks like a Nintendo 64-bit interface. So it's like these little characters that look like the little characters from the hockey game on the original Nintendo. So it had that very
00:21:05
Speaker
I don't know, basic, simple, old school sort of look and feel to it. But it was definitely interesting. It was certainly interesting. OK, let's get back to the conference. So we've talked about DISCOM. I want to ask you each if you had a favorite session for the week.
Favorite Sessions and Innovative Research
00:21:24
Speaker
Elvita, maybe we go with you first.
00:21:27
Speaker
Yeah, sure. I mean, I have to say it was hard to pick because I liked Barbara Traversky's talk at Vis Psych. And I also liked Sheila Carpenter's Capstone for very different reasons. But if I were to pick a single session that I liked, it would have been, there was a short paper session on theory, cognition and sense making on Thursday morning. And in general, I
00:21:52
Speaker
enjoyed the brevity of the short papers. And this session had a little bit of everything that I'm interested in. So there were a pair of papers on characterizing data insights, and that's relevant to some of the work that I'm doing in machine learning and user modeling.
00:22:09
Speaker
My student, Mel, she presented a paper that used economic theory to evaluate decision making with visualizations. And then there was another paper on personality traits and how that might affect preferences. And then finally, there were two papers in the session that kind of challenged conventional wisdom and
00:22:32
Speaker
terms of how we select data visualization and how we teach visualization. So one paper challenged us to rethink chart junk and they did some interview studies where they got practitioners' perspective on chart junk. And then the other paper kind of posed the question of maybe we shouldn't be optimizing for speed and accuracy when we are designing visualizations. And so
00:22:55
Speaker
maybe other things like memorability, aesthetics, just like overall calling for other ways of evaluating visualization, which is very much in line with the Believe Symposium. Right. Do you find that the virtual setup made it easier for you to choose a favorite session or harder?
00:23:19
Speaker
It was harder because, so at the beginning of the conference, I was sitting in a room in my office and I wasn't distracted at all to the point that I could attend two talks at once. Wow. Good for you.
00:23:34
Speaker
I mean, I think I was sharing the VDS symposium, and I also had Barbara Traversky's talk happening at the same time. And how could you choose between these two? Right. But later on in the week, I was really getting fatigued. And I struggled to just keep up with the talks. And so this was a session later on in the week where the talks were really short.
00:24:00
Speaker
And so it was really easy to just pay attention for a short period of time, take a break and then, and then repeat. Right. Right, right, right. That makes sense. Okay. So Robert, you have a favorite session and then we'll go on to favorite paper. So we're just, we're just winnowing our way down. So Robert, do you have a favorite session?
00:24:18
Speaker
Yes, I'm going to try and not just pick the exact same things as Elvida because I think we actually share a lot of the same sort of ideas and interests and also preferences for short papers because I actually like the short papers too. And not everybody does. I know that some people think that they're not such a great idea, but I actually find it more useful to have a short paper that has a more focused, smaller contribution than
00:24:42
Speaker
having to have a full paper that's trying to stuff tons of stuff into one paper that may or may not be somewhat superfluous to the actual point of the paper. So I think it's actually a good thing to have that. I also like the short talks because I feel like they have to be more focused and that that works better for me. So a couple of things that I'm going to just pick here because there's a little bit of overlap here with Elvida, but I think actually that paper that that Elvida is a co-author on with Melanie Bancillo, I think is the first author. I'm not sure.
00:25:12
Speaker
I'm getting that right. So that paper was very interesting because it talked about how showing the same data to people using different methods changes the decisions they make.
00:25:25
Speaker
And that to me is just the key. If that's what we're finding in 2020, what have we been doing for the last 50 years? So I think that's really, really key to me. And it's also interesting because there's a connection there with one of the best papers, as in the official best papers on communicating and visualizing uncertainty, because they were also looking at different kinds of representations. So they're looking at how you show uncertainty
00:25:51
Speaker
And what do people read from them? How do people interpret them? Can they read them? What do they get from them? And when do they make a better decision? Or when do they give you a better estimate of the actual uncertainty from those? And they also found differences between different kinds of representations. They have these bars and
00:26:08
Speaker
the Gaussian bell curve type thing that you've probably seen. And then you have the dot plots. And so they find some differences between them. But they also find that by adding things that are additional information, like showing a mean or a median, I forget which one, they actually also bias people. And so it's really interesting to see just how much small choices can change how people interpret the data and what they get out of the chart and how they would then make a decision based on that.
00:26:37
Speaker
And many of these things aren't really the things that you would usually talk about when you talk about the choice for making a chart. It's more about like, is this a bar chart or a line chart, or is this whatever scatterplot or something like that. But these little things are actually really important. And I think that's what we're just starting to learn. And of course, the choice of chart isn't just about precision, as we like to think, but it also has to do with how we interpret the data and how we read it and how much we think
00:27:04
Speaker
differences in magnitude are important or not important. So I think there's some really interesting stuff that is happening. So those were some of my favorites. Yeah, those are all great. Do we want to talk about favorite papers? You can pick two if you'd like. Elvita, you want to start favorite paper or papers? Okay, so I have several papers and it's hard to choose because I think that my students would be very upset if I chose a paper that's not theirs.
00:27:34
Speaker
Well, it's like picking your favorite kids. So we could just assume that anyone that's your student, those are your favorites. This is definitely true. So one of my favorite papers is the first paper that actually caught my attention and caused me to actually stop trying to multitask and listen. It was your paper, John. It was on diversity, equity, and inclusion and data visualization.
00:27:59
Speaker
What I really liked about that paper is that a lot of the things that I do and even the things that I teach in visualization, it feels like muscle memory right now, where I try to do this data mapping and so forth. But we don't really stop and think that visualization is also a form of expression. And the same issues that we have in society, in terms of biases and stuff like that, social justice,
00:28:28
Speaker
They also exist when we design visualizations. And so I feel like your talk, it pointed out a few things, like, for instance, by default, when we're having a visualization that is about male and female, we do pink and blue. And as a society, we're now evolving to not see gender as binary, but it's way more nuanced than that. But we don't really acknowledge that in our visualization designs.
00:28:56
Speaker
and the choice of how we present data about different ethnic groups. It's nine out of ten times you see like white, black, and so forth. It's just by default we just keep things in that order. But
00:29:13
Speaker
We need to think more about these things and we need to think about how we express ourselves using visualizations because it is a form of expression. And we also need to think about who our visualization designs include and who they exclude. We don't have these conversations. I feel like this might be the first time at Viz I've heard someone give a talk about these topics and I would love to see more of them.
00:29:37
Speaker
First off, thank you. You know, that work with my urban colleague, Alice Feng has, you know, obviously came out of what's been happening in the States for the last few months. And I think a lot of people, certainly people that I work with, but I think just a lot of people that I know are struggling with what can any individual do in the face of explicit racism? And I won't talk about this too long, but I did just want to just like say a couple of things like,
00:30:03
Speaker
There is rampant racism in the United States that in my position as a white male and my privilege, I am not as aware of as I perhaps should be. I am aware of the anti-Semitism because I have experienced that, but that's a different sort of thing. And I think a lot of people want to know what can they do.
00:30:21
Speaker
And from my perspective, as someone who's working with data and visualizations and doing research, this felt like the thing that I could do, that this was a conversation that needed to happen in at least my little sphere of the DataVis community, which is not really the academic community, but it's something that I felt we, at least in the people that I work with, should be talking about. And Alice and I spent a lot of time thinking about it, and it just kind of accelerated over the last few months.
00:30:46
Speaker
But I think your point is right on, Elvita, that a lot of what we talk about in that paper is really just thinking a little harder and a little more strategic about the words and the colors and the shapes and the icons that we use. I don't think there are any answers to these questions necessarily, but I think it's some more careful thought about this would go a long way to creating a more equitable approach to visualizing data.
00:31:13
Speaker
Definitely. So yes, that was my favorite. And I do have to mention that my student's paper, Jen Ha, she presented a paper in Fail Fest. And I really liked that paper because it was kind of like celebrating our failures. And we try to evaluate a mixed initiative system that adapts to someone as they're clicking around. And
00:31:36
Speaker
no matter what we try, the evaluation just wasn't working. And I think it's just because we had some assumptions about how people would behave and how people interact with visualizations. And they did not meet those assumptions. And so I think that's an important paper and an important future directions that we need to look into.
00:31:55
Speaker
But also another student, Cheyenne, he had a paper where he looked at analyzing people's interactions and try to infer their biases, which is also another important area of work. Great. Robert, you have a couple of favorite papers?
00:32:11
Speaker
Yeah, it's kind of hard to pick, but I'm going to give one more shout out to Alvida's paper about the different kinds of representations because I thought that was really interesting. So that would be perhaps my favorite short paper. And then I also, by the way, John really liked yours in this comment. It was a fantastic contribution there.
00:32:31
Speaker
And then, so I do like the the best info is paper that got the actual best paper award, I think was a really interesting paper and I need to actually really read that in depth. This is the one that talked about uncertainty and how different ways of showing that can change how you perceive that so I think that was a really interesting one, and really well done and really well presented to.
00:32:51
Speaker
And then there were a few papers, those are slightly different, but they kind of go in the same kind of direction, I guess, from the perceptual point of view that measured how people read and remember numbers or the size of bars in particular. There was a paper that was called Truth or Square, and it looked at
00:33:10
Speaker
making a bar chart, basically a single bar, wider or skinnier, and how that would lead to over and under estimation, which is really weird. So there is something about how the aspect ratio of the bar, meaning the width to height ratio, changes how you remember the value.
00:33:29
Speaker
which is really interesting because we can change that for the whole chart. We don't usually, I guess, change it for individual bars. So if there is bias that comes from the fact that in the same chart with the same bar width, the different heights of the bars have different aspect ratios, and those are all remembered differently because of those effects, that actually introduces a bias, which is pretty interesting to see.
00:33:52
Speaker
in a bar chart in particular, because bars are sort of like the beyond reproach, the best chart out there. So that's always an interesting thing to see. And there was a related paper that looked at these categorical boundaries, I guess, where the value goes from zero to half of another bar, when you have two bars next to each other, and then to the same height, where there's this sort of repulsion effect, where if it's exactly the same, you're going to remember that. But if it's just off by a little bit, you're overestimating the difference. And so there are little
00:34:21
Speaker
psychological effects that are really interesting and that they're really only starting to explore and to see that they're there. So that's, that's always fun to see. Yeah, those are both really interesting. So in terms of now that the conference is over, are there particular papers or threads of research that you're interested in seeing what will happen in the next stage or the next phase of research, not necessarily from that particular author, but just in that thread of research?
Future of Conference Research and Structure
00:34:50
Speaker
So I would say that to also echo what Alida was saying earlier about qualitative research, I would hope that there's going to be more qualitative research at the conference in the future. Because there is a little bit, and I think it's getting a little bit more traction, but I also know from people I've talked to who have gotten their stuff rejected, that it's hard to publish this work at the Viz conference.
00:35:13
Speaker
And there are other conferences like CHI that are much more open to it. And so I hope we can get more of that work into the VIS conference and that we also open up the broader thinking about visualization. We've done that with the perceptual psychology work, which was not a thing 15 years ago and is now, I think, very well established. And I think we can do it with qualitative research as well. We just need to want to, I guess.
00:35:37
Speaker
Yeah, I definitely agree with that. And I mean, to add to that, I know like we as a community, the visualization is positioned to address a lot of the societal problems that we have right now centered around data. And I think just in general, I do like this trend that I'm seeing at the conference where you have these new workshops like Vis Psych, Vis X Science, Vis X AI. And of course, there's Malui, which is a machine learning and visualization workshop.
00:36:05
Speaker
But I do like us branching out and talking to different communities because a lot of times we spend a lot of time just talking to ourselves. And I think if we're going to solve really hard problems, we do need to solve them together. Like there's no reason why these separate communities are solving problems on their own. So I do want to see more of these collaborative workshops and
00:36:29
Speaker
Specifically, I do want to see actual actionable things coming out of these workshops and new threads of research that are going to be happening because of these workshops.
00:36:38
Speaker
And so that's a great topic maybe to talk about for next year. And I know Robert, you're transitioning off the VisCom committee and I'll be there on the committee. Within VisCom or more generally, what do you think the DataViz academic community needs to do or needs to have happen to have more of this practical approach or have more of partnerships with practitioners?
00:37:02
Speaker
What is it that needs to happen? Is it more of these kinds of workshops or is it something else or multiple things? I'm guessing multiple things, but what would those things be? Like if you two are in charge of IEEE and had the magic wand, like what would your orders be to come down from on high to sort of encourage these partnerships and this more practical application actionable work?
00:37:29
Speaker
I think that's a hard question. It's a hard question because I feel like there are, like, I don't know why people are not doing the work. That's the thing. So it's hard to solve a problem when I actually don't know what's causing it. Is it that people are doing the work but they're not being accepted? So we are seeing more interdisciplinary work at CHI, for instance.
00:37:53
Speaker
And so, is it that maybe when they try to publish them, they don't get accepted and then that's now a bigger conversation to be had about our identity as their IEEE Vis conference and the types of papers that we want to see represented there. So, my answer is I don't know. I'm not sure if it's that it's not being published, if it's that people are not doing the work. So, it's really hard to answer that question because I don't know what's causing it.
00:38:23
Speaker
So one thing I would add to this, and I'm going to be uncharacteristically positive and sort of like upbeat about the future here, but the VIS conference is changing very dramatically next year. So what they've done is, the VIS is actually supposed to be an acronym. So it stands, it's V-I-S, and it stands, it's really hard to explain, but it's the V stands for VAST, which is another acronym, which I'm not going to explain, and then I stands for Infoviz, so that's information visualization, and then the S stands for CIVIS, scientific visualization.
00:38:53
Speaker
And I could go on for a long time to try and explain what these all mean and how they're different, which for the most part, for most people, doesn't mean anything. But this is a historical thing, how this has developed over the years and over about 30 years or so that it's been around. And this next year, this coming year, what they're doing is they're basically breaking up all these differences between the sort of sub conferences that have been there until this year, even though they've sort of tried to erode them a little bit the last couple of years.
00:39:19
Speaker
But next year will be entirely different. The whole reviewing is going to be different. It's all about the different topic areas. I forget what they're actually called. And it's more of the CHI model. So CHI is this HCI conference, human-computer interaction. And CHI is CHI for some reason. It's just called CHI.
00:39:35
Speaker
But that conference uses that same model. And they seem to have done better in bringing in a lot more diversity, I think, in topics, certainly, and I think also in people. And so I think that this is going to look different next year, that's for sure. And I'm very hopeful that it's going to help a lot to have these little subcommittees and area chairs that can do more of their own thing to bring in more work that just isn't as easy to get in so far. So that's my hope for the future.
00:40:03
Speaker
So I hesitate to ask one more question because Robert ended on a positive note. But I do want to ask one more. So it entails along with Robert what you just said. So in addition to maybe changing how the actual process takes place and how papers are accepted, are there any other changes you'd like to see at next year's
Speculating on Future Conference Formats
00:40:27
Speaker
conference? And I also hesitate to ask whether we'll be in person or virtual. So
00:40:32
Speaker
However you want to answer that, but are there things that you would like to see, in addition to what we've been talking about at the actual conference? Elviet, I don't know if you have anything that you're looking forward to in particular.
00:40:45
Speaker
Yeah, I mean, I see lots of positive changes. So in addition to breaking down these artificial barriers that we've set for our community, which I don't quite understand, even though I've been there for 10 years now. But I am hoping that this virtual conference will open that door. So I'm hoping that someone who wouldn't have normally come to Viz attended the virtual conference this year, and then that might open the door for them to come in person, hopefully.
00:41:13
Speaker
if that's the case next year. But in my capacity as the inclusivity and diversity co-chair, we are generally always trying to expand the type of people that show up at this. So we want more practitioners, we want more people from industry, and frankly speaking, more people of color to come in. And we have like scholarships that are available for people who can't finance it. We have
00:41:38
Speaker
scholarship and grants for people who might need help with child care. So I'm excited for the fact that we had the opportunity to broaden their reach of this and I'm hoping that we can then actually bring them in person. Yeah, that's great. Robert, anything else in addition to what you already mentioned that you wanted to highlight for maybe next year?
00:42:04
Speaker
Sure. So I think next year, in my humble opinion, will have to be at least a hybrid model, because even if we have the wherewithal and the control of the virus and a vaccine and enough people vaccinated and all this stuff,
00:42:20
Speaker
I think it will be very challenging to get people to travel to New Orleans. So it's going to be New Orleans next year to travel to the US and New Orleans from outside because people who need to get visas, for example, or who have like a longer process where you have to book things, you know, months in advance, are you going to know in June that or in July, perhaps that what the situation is going to look like in November? I don't know. So I think it's fairly unlikely that it will be in person only conference.
00:42:48
Speaker
But I do hope that we can retain some of the good parts of the virtual conference and maybe improve some of them. I don't love Discord. I think it's good to have a back channel. I just hope that there's a better system than Discord for that. I also think it's a really good thing to have access to all the talks right away and not like weeks later, as was in the past. There are certainly good things about the online experience.
00:43:13
Speaker
But it certainly has its challenges. So I do hope that some of this can be taken into account and made turn into something useful for next year. And again, I don't expect it to be entirely in person. So it'll be very interesting, especially if it actually is sort of a mixed model, how you integrate people who aren't there in person and not have them sort of like totally separate will be quite interesting. And we'll see how well that works.
00:43:39
Speaker
Well, I like the way we seem to have ended on a positive note. I mean, I love the idea that we can get more diversity and more inclusive work in the people that attend this. And we have this sort of maybe blended model where we can have the best of both worlds. We're able to let people participate in maybe ways that they weren't in the past and bring in more practitioners. I think this is all good. So Robert, I'm feeling conflicted. We're ending on a positive note. Yeah, I'm sorry.
00:44:09
Speaker
Everything's maybe we need a lot more work to do, but I feel like we've done a good job here. Elvito, Robert, thanks so much for coming on the show. This has been really interesting. And thanks for taking time out to chat with me. Yeah, it's been fun. Thank you. Thank you for having me. It's been really fun. Thanks.
00:44:34
Speaker
And thanks to everyone for tuning in to this week's episode of the show. I hope you enjoyed that conversation and I hope you're inspired to go check out some of the papers that we discussed and some of the sessions that we talked about over at the IEEE website and some of the associated websites like VizCom and VizActivities.
00:44:49
Speaker
I posted a few of the papers and some of the links over on the show notes page if you want easy access but I recommend you go over to the IEEE website where you can access all of the material from the conference and even I think coming up in another week or two be able to watch some if not all of the video recordings from the presentations. So until next time this has been the PolicyBiz Podcast. Thanks so much for listening.
00:45:19
Speaker
A number of people help bring you the PolicyViz podcast. Music is provided by the NRIs, audio editing is provided by Ken Skaggs, and each episode is transcribed by Genic Transcription Services. The PolicyViz website is hosted by WP Engine and is published on WordPress. If you would like to help support the podcast, please visit our Patreon page.
Conclusion and Podcast Support
00:45:37
Speaker
I have trouble sleeping at night, eyes took open.
00:45:55
Speaker
It's so hard to make a mess. It's coming around the bend. It's coming around the bend. Too much noise. Too much life. Too much noise.