WEBVTT 1 00:00:00.200 --> 00:00:06.919 So you remember flattening the curve right back in twenty twenty, 2 00:00:07.040 --> 00:00:11.119 that simple line graph became this, well, this incredibly potent image. 3 00:00:11.199 --> 00:00:12.000 Yeah, it really did. 4 00:00:12.039 --> 00:00:16.480 It was everywhere exactly and it represented this statistical argument 5 00:00:16.519 --> 00:00:20.079 basically for why we all needed to, you know, drastically 6 00:00:20.199 --> 00:00:21.920 change our behavior during the pandemic. 7 00:00:22.000 --> 00:00:25.519 It's pretty interesting how a visual representation of data could 8 00:00:25.559 --> 00:00:30.000 become such a like a shared understanding almost instantly. 9 00:00:29.640 --> 00:00:32.799 Right, and renod Letur pointed out something interesting there. He said, 10 00:00:32.840 --> 00:00:35.759 the actual virus, the trace of it, was really only 11 00:00:35.840 --> 00:00:40.079 knowable to scientists. For the rest of us, its impact 12 00:00:40.200 --> 00:00:44.079 was understood through well data, through statistics like that graph. 13 00:00:44.200 --> 00:00:48.359 It's fascinating, isn't it. How quickly a fairly abstract idea 14 00:00:48.479 --> 00:00:51.960 just shown visually became such a central thing in public discussion. 15 00:00:52.000 --> 00:00:55.799 It really highlights the power these data visualizations have. 16 00:00:55.960 --> 00:00:58.520 Definitely, how they shape how we see things, how we 17 00:00:58.560 --> 00:01:00.880 react to the world. Yeah, and that's really what we're 18 00:01:00.880 --> 00:01:03.560 digging into today. We're not just looking at how visualations 19 00:01:03.920 --> 00:01:06.599 help us understand things, you know, on the surface, but 20 00:01:06.719 --> 00:01:09.599 taking a much more critical look. It's kind of a puzzle, 21 00:01:09.680 --> 00:01:12.680 isn't it. We're just swimming in data these days. Oh yeah, 22 00:01:12.760 --> 00:01:15.400 yet our ability to really interpret it, to make sense 23 00:01:15.439 --> 00:01:19.239 of it, maybe even question it. Well, it hasn't necessarily 24 00:01:19.319 --> 00:01:19.760 kept pace. 25 00:01:20.000 --> 00:01:22.280 That's a really important point. And you see this rise 26 00:01:22.400 --> 00:01:27.159 in skepticism even towards you know, well established science, plus 27 00:01:27.239 --> 00:01:28.920 the whole filter bubble issue. 28 00:01:28.799 --> 00:01:31.079 Right, how that shapes what we even see. 29 00:01:31.120 --> 00:01:34.359 Exactly, and the sheer volume of data itself can just 30 00:01:34.400 --> 00:01:37.439 be overwhelming. It makes it harder to understand, harder to 31 00:01:37.519 --> 00:01:38.879 interpret effectively. 32 00:01:39.120 --> 00:01:42.519 So okay, our mission today is to really unpack how 33 00:01:42.599 --> 00:01:45.159 these data visualizations work. We want to look at the 34 00:01:45.159 --> 00:01:48.799 assumptions underneath the decisions that often get hidden behind the scenes, 35 00:01:49.120 --> 00:01:52.439 and crucially why it's so important to approach them with 36 00:01:52.480 --> 00:01:53.159 a critical eye. 37 00:01:53.319 --> 00:01:56.480 Yeah, and maybe even think about not just critiquing what exists, 38 00:01:56.519 --> 00:02:01.120 but how we could create visualizations that offer different perspectives, 39 00:02:01.200 --> 00:02:02.599 challenge things exactly. 40 00:02:02.719 --> 00:02:06.040 And for you listening, this is really about feeling more confident, 41 00:02:06.120 --> 00:02:09.560 more informed, without getting totally lost in all the numbers 42 00:02:09.560 --> 00:02:10.000 and charts. 43 00:02:10.400 --> 00:02:13.080 Right, understanding these nuances, it can be a kind of 44 00:02:13.120 --> 00:02:17.400 shortcut to deeper insight, a way to cut through the 45 00:02:17.439 --> 00:02:19.319 noise and figure out what's actually being said. 46 00:02:19.400 --> 00:02:21.560 Okay, So let's start with this idea of the veil 47 00:02:21.599 --> 00:02:29.039 of neutrality. It often seems like DEBTA visualizations are just neutral, objective. 48 00:02:29.400 --> 00:02:31.840 Yeah, that's a common perception. You see a clean chart, 49 00:02:31.840 --> 00:02:34.639 a polished graphic, and you think, okay, the data speaks 50 00:02:34.639 --> 00:02:35.159 for itself. 51 00:02:35.199 --> 00:02:37.120 But that's often where it gets interesting, doesn't it. 52 00:02:37.120 --> 00:02:41.479 It really is because these seemingly objective visuals can actually 53 00:02:41.639 --> 00:02:48.520 hide the very human choices behind them and the broader context, 54 00:02:48.639 --> 00:02:50.400 you know, the political, the economic context. 55 00:02:50.439 --> 00:02:53.240 They are made in things like algorithms too. They seem neutral, 56 00:02:53.240 --> 00:02:55.639 but they're built on tons of decisions made by. 57 00:02:55.520 --> 00:02:59.800 People precisely, which leads to those two extremes you sometimes see. Yeah, 58 00:03:00.159 --> 00:03:03.319 while on one hand you get visualizations used almost like weapons, 59 00:03:03.680 --> 00:03:07.439 pushing specific agendas, often by powerful groups, and then on 60 00:03:07.479 --> 00:03:09.719 the other extreme you get this sort of blind distrust. 61 00:03:10.120 --> 00:03:13.120 People immediately reject anything visual that doesn't fit what they 62 00:03:13.120 --> 00:03:13.719 already think. 63 00:03:13.960 --> 00:03:16.280 So it's a real tension how they can inform versus 64 00:03:16.280 --> 00:03:19.960 how they can persuade or just reinforce biases. 65 00:03:20.039 --> 00:03:25.360 Yeah, the challenge is to encourage like careful examination, rigorous critique, 66 00:03:25.800 --> 00:03:28.319 not just knee jerk acceptance or rejection. 67 00:03:28.199 --> 00:03:30.560 And that careful examination. It needs to go beyond just 68 00:03:30.879 --> 00:03:33.599 does it look nice or is it easy to read? 69 00:03:33.919 --> 00:03:37.520 Right? Absolutely, there's often this big emphasis, especially in design circles, 70 00:03:37.879 --> 00:03:43.919 on making things efficient, clear, excellent visual displays, which is valuable. 71 00:03:43.520 --> 00:03:46.680 Sure, but it can overshadow the really fundamental question. 72 00:03:46.560 --> 00:03:49.400 Exactly, like how is this data gathered in the first place, 73 00:03:49.479 --> 00:03:52.199 who decided what to include or exclude? And how is 74 00:03:52.240 --> 00:03:55.560 it being presented in a way that might subtly guide 75 00:03:55.599 --> 00:03:59.000 how we interpret it. Even someone like Hans Rosling, who 76 00:03:59.080 --> 00:04:04.240 is brilliant, absolutely brilliant at using visualizations to correct misconceptions 77 00:04:04.240 --> 00:04:07.800 about global health, he didn't always push that deeper level 78 00:04:07.800 --> 00:04:10.199 of critical thinking about the data's origins. 79 00:04:10.280 --> 00:04:12.000 Okay, so let's make this concrete. Let's look at a 80 00:04:12.000 --> 00:04:15.960 specific example, David mccanless's Colors and Culture visualization. 81 00:04:16.399 --> 00:04:18.759 Right, that's a good one. On the surface, it looks 82 00:04:18.759 --> 00:04:19.519 pretty straightforward. 83 00:04:19.680 --> 00:04:23.480 Yeah, the title says it all colors and culture, the 84 00:04:23.600 --> 00:04:27.040 meanings of colors around the world. It's supposed to show how, 85 00:04:27.040 --> 00:04:31.199 I say, red means good luck here, but maybe danger 86 00:04:31.319 --> 00:04:32.360 or debt somewhere else. 87 00:04:32.800 --> 00:04:37.800 And McCandless positions himself as this, you know, data journalist, 88 00:04:38.199 --> 00:04:42.040 someone telling stories with data using spreadsheets and design. 89 00:04:41.800 --> 00:04:44.319 Tools, which sounds great, right, and a world full of 90 00:04:44.399 --> 00:04:46.720 data having someone make sense of it as appealing. 91 00:04:46.399 --> 00:04:50.519 Definitely, But when you start looking closer at that specific visualization, 92 00:04:51.240 --> 00:04:53.600 some questions pop up about its foundations. 93 00:04:53.959 --> 00:04:56.480 Yeah, this is where it gets really interesting. Where did 94 00:04:56.480 --> 00:04:59.079 the information actually come from? Because it doesn't seem like 95 00:04:59.079 --> 00:05:02.000 it's based on you know, deep academic ethnography or anything. 96 00:05:02.120 --> 00:05:05.399 No, it seems to rely heavily on well tertiary sources, 97 00:05:05.839 --> 00:05:09.959 specifically Wikipedia and what he calls the general Web, which 98 00:05:10.040 --> 00:05:11.759 probably means Google searches. 99 00:05:11.439 --> 00:05:14.319 Things like that, so readily available, but maybe not the 100 00:05:14.360 --> 00:05:17.839 most rigorous or nuanced sources. 101 00:05:18.079 --> 00:05:22.360 That's the concern. It raises questions about thoroughness, potential biases 102 00:05:22.399 --> 00:05:23.560 in those sources. 103 00:05:23.439 --> 00:05:26.519 You know, and maybe that explains some of the well, 104 00:05:26.560 --> 00:05:29.560 the inconsistencies in how things are categorized could be. 105 00:05:29.680 --> 00:05:31.439 I mean, you look at it, You've got continents like 106 00:05:31.560 --> 00:05:34.360 Africa and Asia, but then also America, which is in 107 00:05:34.439 --> 00:05:36.879 the comment in South America and Eastern Europe. 108 00:05:36.920 --> 00:05:40.920 Right, subcontinents were regions. Then you've got religions mixed in Hindu, Muslim, 109 00:05:41.319 --> 00:05:45.120 one indigenous group, Native American, and just two specific countries 110 00:05:45.240 --> 00:05:46.600 Japan and China. 111 00:05:46.680 --> 00:05:51.040 Yeah, it feels a bit arbitrary. There's no clear consistent 112 00:05:51.120 --> 00:05:53.959 logic behind the categories, like where's Western Europe? 113 00:05:54.000 --> 00:05:56.839 And the amount of information seems really uneven too, like 114 00:05:57.000 --> 00:06:00.639 America has loads of entries, but South America, the inside 115 00:06:00.680 --> 00:06:02.800 ring has hardly any Exactly. 116 00:06:02.879 --> 00:06:05.600 That imbalance suggests that data collection might have been skewed, 117 00:06:05.959 --> 00:06:08.800 maybe just reflecting what was easiest to find online for 118 00:06:08.879 --> 00:06:09.600 certain regions. 119 00:06:09.639 --> 00:06:12.839 So it makes you wonder what was the main goal here? 120 00:06:13.000 --> 00:06:16.560 Was it really about accurately showing the nuances of color 121 00:06:16.680 --> 00:06:17.959 meaning globally. 122 00:06:17.800 --> 00:06:21.000 Or was it perhaps more about creating a visually appealing 123 00:06:21.160 --> 00:06:23.639 infographic the form over the function. 124 00:06:24.040 --> 00:06:27.600 Maybe it raises that important question, doesn't it. We often 125 00:06:27.639 --> 00:06:32.240 assume a visualization is authoritative, but its validity really depends 126 00:06:32.319 --> 00:06:34.319 on the rigor of the underlying data. 127 00:06:34.560 --> 00:06:36.519 Absolutely always question the source. 128 00:06:36.839 --> 00:06:40.279 It's crucial, Okay, thinking bigger picture. Now we touched on 129 00:06:40.600 --> 00:06:44.160 who creates these things, and there's often a power imbalance 130 00:06:44.199 --> 00:06:44.839 there isn't there? 131 00:06:44.920 --> 00:06:50.680 Yeah? Definitely, State actors, big corporations, they generally have way 132 00:06:50.720 --> 00:06:54.720 more resources to collect, store, analyze massive data sets. 133 00:06:54.480 --> 00:06:59.199 Which means individuals, citizens, community groups, we're more likely to 134 00:06:59.199 --> 00:07:02.160 be the subjects of data collection rather than using data 135 00:07:02.160 --> 00:07:03.800 ourselves for civic stuff. 136 00:07:03.959 --> 00:07:07.360 That's often the case. So based on that access issue, 137 00:07:07.480 --> 00:07:09.600 what kind of power dynamics does that create in the 138 00:07:09.600 --> 00:07:10.959 world of data visualization. 139 00:07:11.199 --> 00:07:13.839 Well, it suggests that people with the resources get to 140 00:07:13.879 --> 00:07:16.480 frame the narratives, right, They control the data flow and 141 00:07:16.480 --> 00:07:17.680 how it's presented. 142 00:07:17.360 --> 00:07:19.519 Exactly, which is why there's such a strong argument for 143 00:07:19.560 --> 00:07:21.639 boosting data literacy more broadly. 144 00:07:21.399 --> 00:07:25.000 Giving people the skills to understand, interpret it, even create 145 00:07:25.040 --> 00:07:26.319 their own visualizations. 146 00:07:26.759 --> 00:07:30.720 Yes, it's a key step towards addressing that power imbalance, 147 00:07:31.279 --> 00:07:35.079 making people more active users, not just passive consumers or 148 00:07:35.120 --> 00:07:36.720 subjects of data. 149 00:07:36.759 --> 00:07:40.040 Okay, so we have this illusion of neutrality, this power dynamic. 150 00:07:40.720 --> 00:07:44.240 Now let's dig into how the visualizations themselves can sort 151 00:07:44.240 --> 00:07:47.920 of hide the work, obscure how they were made. 152 00:07:48.240 --> 00:07:53.040 Right. It's interesting how a really clear, polished, excellent visualization 153 00:07:53.600 --> 00:07:57.319 often makes the messy reality of its creation completely invisible. 154 00:07:57.399 --> 00:08:01.519 All the uncertainty, the arguments, the cleaning, the data filtering. 155 00:08:01.600 --> 00:08:04.519 Yeah, all that preparatory work that debates the choices, it 156 00:08:04.639 --> 00:08:07.480 just vanishes behind this slick, user friendly interface. 157 00:08:07.560 --> 00:08:09.759 It's like a magic trick. Almost it looks so clean, 158 00:08:09.920 --> 00:08:12.560 so authoritative, you forget all the human judgment calls that 159 00:08:12.600 --> 00:08:13.199 went into it. 160 00:08:13.240 --> 00:08:15.360 And it's not just visualizations, is it. It's similar with 161 00:08:15.399 --> 00:08:16.560 algorithms totally. 162 00:08:16.879 --> 00:08:20.079 Kathy O'Neill's book Weapons of Map Destruction talks about this right, 163 00:08:20.160 --> 00:08:23.560 how algorithms, often rolled out without much oversight, can bake 164 00:08:23.639 --> 00:08:26.000 in and even amplify existing inequalities. 165 00:08:26.279 --> 00:08:29.240 She gives that stark example of the teacher fired by 166 00:08:29.279 --> 00:08:33.360 a flawed algorithm. The algorithm didn't account for key context, 167 00:08:33.840 --> 00:08:36.879 and the hidden human choices in its design had devastating 168 00:08:36.919 --> 00:08:38.279 real world consequences. 169 00:08:38.679 --> 00:08:42.080 So the point isn't that every visualization should look messy, 170 00:08:42.279 --> 00:08:45.519 but we need to be aware of this, this rhetorical effect. 171 00:08:46.039 --> 00:08:49.279 How hiding the decisions can create this false sense of 172 00:08:49.840 --> 00:08:52.159 absolute truth precisely. 173 00:08:51.840 --> 00:08:53.440 And we can actually learn a lot here from a 174 00:08:53.480 --> 00:08:58.840 related field, critical cartography map making. How so well. Critical 175 00:08:58.879 --> 00:09:01.360 cartography emerged when PEP people started questioning what they call 176 00:09:01.399 --> 00:09:04.879 the cartography of progress. This was mapping driven mainly by 177 00:09:04.919 --> 00:09:07.320 technical goals, seemingly neutral. 178 00:09:07.080 --> 00:09:10.399 But ignoring the underlying ideologies like why map this, how 179 00:09:10.440 --> 00:09:11.639 map it? Who benefits? 180 00:09:11.799 --> 00:09:15.759 Exactly? Just like decisions behind visualizations can be hidden. The 181 00:09:15.840 --> 00:09:18.200 bias is embedded in maps were often overlooked for a 182 00:09:18.240 --> 00:09:21.759 long time. Critical cartography brought those questions to the. 183 00:09:21.720 --> 00:09:25.240 Foe, and Halpern argues visualization goes even further. It doesn't 184 00:09:25.240 --> 00:09:28.399 just reflect the world, it actually shapes how we think 185 00:09:28.440 --> 00:09:28.879 about it. 186 00:09:29.080 --> 00:09:35.399 Yeah, Calpern's argument is quite profound. She suggests visualization literally 187 00:09:35.480 --> 00:09:39.639 trains our perception, confines what counts as rational or reasonable, 188 00:09:40.039 --> 00:09:43.360 and even transforms how we think about governing populations. This 189 00:09:43.480 --> 00:09:47.720 idea of governmentality who Co's concept increasingly tied to data 190 00:09:48.120 --> 00:09:51.559 calculation and neoliberal economics in the twentieth century. 191 00:09:51.799 --> 00:09:55.559 Visualization makes things visible that aren't immediately obvious. 192 00:09:55.120 --> 00:09:58.240 Right, and she talks about this shift towards experiential vision, 193 00:09:58.519 --> 00:10:01.759 trying to get us to almost think like a machine. 194 00:10:01.240 --> 00:10:05.240 Like those old IBM exhibits at worldfairs, multi screen things 195 00:10:05.279 --> 00:10:06.600 bombarding you with info. 196 00:10:06.919 --> 00:10:10.919 Kind of The underlying idea, maybe was that presenting information 197 00:10:11.080 --> 00:10:14.960 visually dynamically could bypass old ways of understanding and create 198 00:10:14.960 --> 00:10:17.480 this direct, intuitive grasp of complex systems. 199 00:10:17.559 --> 00:10:20.879 Now, most histories of data is presented as this straight line. Right. 200 00:10:21.000 --> 00:10:24.279 Progress innovation leading to better innovation is that the whole story. 201 00:10:24.360 --> 00:10:27.639 That's the conventional narrative, often a linear path, like everything's 202 00:10:27.679 --> 00:10:29.200 heading toward one perfect endpoint. 203 00:10:29.279 --> 00:10:32.639 Think of timelines, Yeah, the timeline metaphor itself, right. 204 00:10:32.879 --> 00:10:35.360 It's so ingrained, it's hard to think of time differently. 205 00:10:35.919 --> 00:10:39.720 But the philosopher Ari Bergson, writing way back, critiqued that 206 00:10:39.879 --> 00:10:43.759 very idea. He said, interesting narratives often move in multiple directions, 207 00:10:43.960 --> 00:10:45.159 they aren't just straight lines. 208 00:10:45.679 --> 00:10:48.039 He was writing around the time of Murray's early motion 209 00:10:48.200 --> 00:10:51.559 capture work, wasn't it and he felt that focus on 210 00:10:52.000 --> 00:10:56.240 measured clock time ignored our personal experience of time, his 211 00:10:56.360 --> 00:10:57.799 idea of duration. 212 00:10:57.799 --> 00:11:00.919 Exactly, and even then people were you using charts to 213 00:11:01.039 --> 00:11:04.960 argue for this idea of progress, this acceleration of science 214 00:11:05.000 --> 00:11:08.279 and art. It shows how historical accounts often have built 215 00:11:08.279 --> 00:11:10.720 in biases. Emphasizing linearity and. 216 00:11:10.679 --> 00:11:13.559 Speaking of biases, it's crucial to remember that Western ways 217 00:11:13.559 --> 00:11:15.159 of visualizing aren't the only. 218 00:11:14.919 --> 00:11:19.159 Ways, absolutely not. Take the kipu from ancient Andean cultures. 219 00:11:19.360 --> 00:11:21.799 It's a fascinating non Western system. It's not just about 220 00:11:21.840 --> 00:11:25.039 abstract numbers. It involves embodied data, physical knowing, and. 221 00:11:25.080 --> 00:11:27.919 David Turnbull's book Maps Where Territories highlights others. 222 00:11:28.039 --> 00:11:31.200 Yes, lots of examples challenging our typical Western map concepts. 223 00:11:31.919 --> 00:11:35.600 Varican drawings used as land claims, okay, Aboriginal bark paintings 224 00:11:35.600 --> 00:11:39.720 from Australia Marshall Islands, stick charts, stick charts. Yeah, they 225 00:11:39.720 --> 00:11:42.279 weren't for navigation like our maps. They were more like 226 00:11:42.320 --> 00:11:45.840 teaching tools, helping mariners learn to feel the ocean swells 227 00:11:45.879 --> 00:11:48.559 and currents. Navigation was a physical skill. 228 00:11:48.679 --> 00:11:49.120 Wow. 229 00:11:49.320 --> 00:11:52.639 Even Inuit coastal maps carved from wood. They had this 230 00:11:52.879 --> 00:11:55.799 tactile physical quality. You'd feel the coastline. 231 00:11:55.840 --> 00:11:59.559 So these aren't just abstract pictures. They're tied to physical experience, 232 00:11:59.559 --> 00:12:01.000 different way of knowing the world. 233 00:12:01.320 --> 00:12:05.759 Okay, let's shift. Let's look at some specific visualizations explicitly 234 00:12:05.799 --> 00:12:10.440 made for critique or reform. The Brooks slave ship diagram. 235 00:12:10.960 --> 00:12:12.000 That's a powerful one. 236 00:12:12.080 --> 00:12:15.600 It is incredibly powerful in surveillance studies. It's become really central. 237 00:12:15.720 --> 00:12:18.399 It argues for putting race and blackness right at the 238 00:12:18.399 --> 00:12:19.960 start of surveillance history. 239 00:12:19.759 --> 00:12:22.159 More so than Bentham's Panopticon, even. 240 00:12:21.960 --> 00:12:25.000 In many ways. Yes, the Panopticon was about subtle control, 241 00:12:25.080 --> 00:12:28.960 the possibility of being watched. The Brooks diagram depicted the 242 00:12:28.960 --> 00:12:32.080 horrific reality of a slave ship, a policy of terror 243 00:12:32.120 --> 00:12:33.480 that defined people as cargo. 244 00:12:33.919 --> 00:12:36.799 And those tiny figures on the diagram, they aren't just 245 00:12:36.879 --> 00:12:39.799 identical blobs, are they. They seem to have different gestures, 246 00:12:39.840 --> 00:12:43.759 some looking back right. Simone Brown, a scholar in this area, argues, 247 00:12:43.799 --> 00:12:46.879 those details suggest to kind of resistance even within that 248 00:12:46.960 --> 00:12:51.600 horrific representation, And unlike the Panopticon meant to be copied, 249 00:12:51.879 --> 00:12:54.399 the Brooks diagram was shown as something so awful it 250 00:12:54.480 --> 00:12:55.799 demanded its own elimination. 251 00:12:56.120 --> 00:12:59.159 Its power is in revealing that legacy. 252 00:12:58.840 --> 00:12:59.919 Yeah, the ongoing legacy of s. 253 00:13:00.480 --> 00:13:03.480 Then there's Florence Nightingale, her polar Area diagram from the 254 00:13:03.519 --> 00:13:06.000 Crimean War, another fascinating case. 255 00:13:06.360 --> 00:13:10.759 Nightingale's chart is famous for his visual innovation, the Rose diagram, 256 00:13:11.159 --> 00:13:13.679 but it's crucial to see it as a situated critique, 257 00:13:13.720 --> 00:13:17.039 a feminist critique really. How So, she used statistics to 258 00:13:17.120 --> 00:13:21.720 directly challenge the dominant, almost romanticized view of soldiers dying 259 00:13:21.759 --> 00:13:26.080 heroically in battle. Her data showed overwhelmingly that most deaths 260 00:13:26.080 --> 00:13:29.159 were from preventable diseases due to terrible sanitation. 261 00:13:29.360 --> 00:13:31.720 So it wasn't just presenting numbers. It was an argument 262 00:13:31.960 --> 00:13:34.759 a direct challenge to the status quo, pushing for reform 263 00:13:34.799 --> 00:13:36.720 in military medicine. 264 00:13:36.360 --> 00:13:40.039 Exactly, and she strategically sent her report with the diagram 265 00:13:40.240 --> 00:13:44.480 to people in power. A very effective piece of persuasive visualization, 266 00:13:45.159 --> 00:13:48.000 though we should also note it was still produced within 267 00:13:48.039 --> 00:13:50.639 a certain Eurocentric colonial context. 268 00:13:50.759 --> 00:13:55.360 Okay. And then a very different example the Great Trigonometrical 269 00:13:55.440 --> 00:13:57.200 Survey of India. 270 00:13:56.759 --> 00:13:59.919 Right the GTS, done by the British East India Company. 271 00:14:00.840 --> 00:14:04.799 This really shows how visualization can impose a conceptual framework 272 00:14:04.879 --> 00:14:05.919 to support control. 273 00:14:06.039 --> 00:14:07.679 It wasn't just mapping what was there. 274 00:14:07.639 --> 00:14:11.360 No, it was about creating an imperial space, imposing a 275 00:14:11.399 --> 00:14:14.679 British way of seeing India onto the land itself, making 276 00:14:14.720 --> 00:14:18.519 it seem manageable, categorizable for the colonizers. 277 00:14:18.120 --> 00:14:21.799 Turning complexity into something seemingly ordered and controllable through the 278 00:14:21.799 --> 00:14:22.720 map precisely. 279 00:14:22.919 --> 00:14:25.279 And this went along with the population census, seemingly just 280 00:14:25.360 --> 00:14:28.639 counting people, but it actually created new identity categories and 281 00:14:28.679 --> 00:14:30.360 became a tool for bureaucratic control. 282 00:14:30.600 --> 00:14:32.919 Arjent Oppaduri talks about this right number is creating the 283 00:14:32.919 --> 00:14:34.519 illusion of bureaucratic control. 284 00:14:34.840 --> 00:14:39.799 Yes, though interestingly, over time those very categories and ways 285 00:14:39.840 --> 00:14:44.120 of counting were sometimes turned back against the colonial power 286 00:14:44.399 --> 00:14:45.519 by the people being counted. 287 00:14:45.600 --> 00:14:48.399 Okay, fast forward a bit W. E. B. Dubois and 288 00:14:48.440 --> 00:14:52.120 the American Negro Exhibit at the nineteen hundred Paris Exposition. 289 00:14:52.720 --> 00:14:55.120 That sounds incredibly innovative, it. 290 00:14:55.080 --> 00:14:58.879 Really was remarkably prescient. Duboy and his team created about 291 00:14:58.919 --> 00:15:03.919 sixty and drawn charts, graphs, maps, and drawn one. Yeah, 292 00:15:04.320 --> 00:15:07.799 and they weren't just formally inventive, reimagining visual forms using 293 00:15:07.799 --> 00:15:11.519 color coded maps spiral graphics. They're also incredibly powerful in 294 00:15:11.559 --> 00:15:16.120 their message. They aim to directly counter racist stereotypes about African. 295 00:15:15.720 --> 00:15:19.879 Americans, focusing on things like population, land ownership, income education, 296 00:15:20.279 --> 00:15:21.840 especially in Georgia Exactly. 297 00:15:21.879 --> 00:15:24.799 The goal was to demonstrate with data that any perceived 298 00:15:24.879 --> 00:15:28.600 racial differences were due to social conditions, systemic barriers, not 299 00:15:28.759 --> 00:15:29.799 inherent inferiority. 300 00:15:29.960 --> 00:15:33.039 So at a World's Fare that often glorified colonialism, he 301 00:15:33.159 --> 00:15:36.879 presented this authoritative, data driven counter narrative. 302 00:15:36.600 --> 00:15:40.440 A powerful statement about progress and resilience despite immense hardship, 303 00:15:40.600 --> 00:15:41.799 really groundbreaking work. 304 00:15:42.080 --> 00:15:45.759 And then we have auto and marine neurath and isotype. 305 00:15:45.879 --> 00:15:48.519 Another hugely influential visual system. 306 00:15:48.679 --> 00:15:52.639 Isotype, the International system of typographic picture education. Yeah, often 307 00:15:52.679 --> 00:15:56.480 seen as the foundation for modern infographics, but their goal 308 00:15:56.600 --> 00:16:00.480 was bigger than just nice pictures. It was educational, deeply social. 309 00:16:00.720 --> 00:16:04.679 They wanted to empower ordinary people, help them understand complex 310 00:16:04.759 --> 00:16:06.960 social and economic stuff in modern cities. 311 00:16:07.159 --> 00:16:10.240 Exactly, they had this department of Transformation in their museum. 312 00:16:10.879 --> 00:16:14.120 Marine Noirath played the crucial role of the transformer. 313 00:16:14.279 --> 00:16:16.399 She took the raw stats and figured out how to 314 00:16:16.399 --> 00:16:21.960 frame them, select organize before Gerdarn's created the iconic pictograms. 315 00:16:22.120 --> 00:16:26.080 Precisely, it's a process very much like mapping, discovery and structuring. 316 00:16:26.639 --> 00:16:30.840 Now isotype has been criticized, maybe for oversimplifying sometimes, but 317 00:16:30.919 --> 00:16:34.200 its influence is massive. Think of the London Underground map, 318 00:16:34.600 --> 00:16:38.799 sacrificing geographic accuracy for clarity of connections. That owes a 319 00:16:38.799 --> 00:16:41.399 lot to Neurat's focus on conveying essential information. 320 00:16:41.759 --> 00:16:44.360 And they had that distinct way of showing quantity right, 321 00:16:44.440 --> 00:16:47.320 repeating standard units instead of just scaling one image up. 322 00:16:47.559 --> 00:16:51.039 Yes, like their chart on Great War casualties, each little 323 00:16:51.039 --> 00:16:55.039 soldier figure represents a million soldiers, color coded for wounded 324 00:16:55.120 --> 00:16:57.559 or killed. Very clear systematic. 325 00:16:57.679 --> 00:17:00.759 That systematic approach influenced later designers too. 326 00:17:00.960 --> 00:17:05.000 Definitely, people like Yan Cheechold with his new typography, Canare 327 00:17:05.119 --> 00:17:09.839 and Calvert's UK road Science, all focused on legibility efficient 328 00:17:09.960 --> 00:17:11.119 information transfer. 329 00:17:11.359 --> 00:17:14.599 It's a good reminder that even seemingly simple visual languages 330 00:17:14.640 --> 00:17:18.519 have philosophies behind them. Okay, let's talk about putting different 331 00:17:18.599 --> 00:17:19.519 data sets together. 332 00:17:19.880 --> 00:17:23.839 Yeah, combining data sets can definitely reveal interesting connections, like 333 00:17:24.279 --> 00:17:27.480 plotting your commute time against transit spending, my show or relationship. 334 00:17:27.599 --> 00:17:30.240 Yeah, but there's a potential pitfall, which is well. 335 00:17:30.079 --> 00:17:32.319 As you layer more and more data, you can start 336 00:17:32.359 --> 00:17:34.799 to lose sight of the specifics of how each individual 337 00:17:34.839 --> 00:17:37.799 data set was made. The focus shifts to the connections 338 00:17:37.839 --> 00:17:41.000 and the context. The potential biases of the original data 339 00:17:41.079 --> 00:17:43.880 can get sort of blurred. 340 00:17:43.519 --> 00:17:46.519 Which brings us to this idea of data as assemblage. 341 00:17:46.680 --> 00:17:49.000 Kitchen and Lareo talk about this right. 342 00:17:49.400 --> 00:17:52.359 They see data not just as raw numbers, but as 343 00:17:52.440 --> 00:17:57.680 part of a complex sociotechnical system includes the tech, the politics, 344 00:17:57.720 --> 00:18:01.400 the social factors, the economics, all interconnected. 345 00:18:01.119 --> 00:18:03.799 Like a recipe. It's not just the ingredients, but the chef, 346 00:18:03.920 --> 00:18:05.720 the oven, the cultural context. 347 00:18:06.160 --> 00:18:09.279 That's a great analogy and the controversy around Michael Mann's 348 00:18:09.359 --> 00:18:13.880 hockey stick graph showing global temperatureize is a perfect example 349 00:18:14.000 --> 00:18:15.599 of a data assemblage in action. 350 00:18:16.039 --> 00:18:19.680 Right, That graph became iconic and face huge backlash. 351 00:18:19.759 --> 00:18:22.839 Yeah, that single visualization and the science network behind it 352 00:18:22.880 --> 00:18:26.480 became a major target for corporate and political interest trying 353 00:18:26.519 --> 00:18:29.519 to downplay climate change. They tried to attack the analysis, 354 00:18:29.559 --> 00:18:31.960 split the data, anything to undermine it. 355 00:18:31.960 --> 00:18:34.599 It shows that facts derived from data need constant work 356 00:18:34.640 --> 00:18:37.720 to maintain their validity within these networks of power and interest. 357 00:18:37.920 --> 00:18:40.440 Exactly, data doesn't exist in a vacuum. 358 00:18:40.680 --> 00:18:43.960 Now, when we talk data, we often mean numbers quantitative, 359 00:18:44.319 --> 00:18:48.400 but there's also qualitative data, text, images, descriptions. How do 360 00:18:48.440 --> 00:18:49.799 they interact in visualization? 361 00:18:50.240 --> 00:18:56.720 So conventionally quantitative is countable, mathematical, qualitative is descriptive, needs interpretation. 362 00:18:57.319 --> 00:19:02.119 Often qualitative data gets turned into quantitative data by categorizing it. 363 00:19:02.079 --> 00:19:04.880 Like coding open ended survey answers into numbers. 364 00:19:04.599 --> 00:19:07.880 Right, which allows for analysis, but you inevitably lose a 365 00:19:07.880 --> 00:19:10.599 lot of the original richness and nuance. It's a trade off. 366 00:19:10.759 --> 00:19:15.160 You also mentioned different levels of measurement indexical data like fingerprints, 367 00:19:15.200 --> 00:19:19.119 attribute data like eye color, and metadata data about data. 368 00:19:19.319 --> 00:19:22.519 Yeah, all these distinctions matter for analysis and Underlying all 369 00:19:22.559 --> 00:19:27.119 this are different philosophical worldviews about reality itself. Creswell talks 370 00:19:27.119 --> 00:19:32.599 about five positivist, post positivist, constructivist, advocacy, pragmatist that shape 371 00:19:32.640 --> 00:19:35.519 how we even ask questions and represent findings. 372 00:19:35.400 --> 00:19:38.039 And trying to fit the messy world into niaque categories. 373 00:19:38.279 --> 00:19:39.640 That gets tricky fast. 374 00:19:39.440 --> 00:19:39.759 Doesn't it? 375 00:19:39.759 --> 00:19:39.920 Oh? 376 00:19:39.880 --> 00:19:42.799 Absolutely? Heck William can't use the simple word book. What 377 00:19:42.960 --> 00:19:46.839 is a book? The author's text, the physical object, a manual? 378 00:19:47.039 --> 00:19:50.240 The edges are blurry, and applying categories to society, like 379 00:19:50.440 --> 00:19:53.160 the gender pay gap, gets even messier. You have to 380 00:19:53.160 --> 00:19:57.880 define work pay gender, and those definitions themselves are loaded 381 00:19:57.880 --> 00:20:02.480 with assumptions, often reflecting dominant ideas that might exclude unpaid 382 00:20:02.480 --> 00:20:04.079 work or non binary genders. 383 00:20:04.359 --> 00:20:07.880 Right, categories are useful, essentially even for thinking. We group 384 00:20:07.960 --> 00:20:11.480 things with shared properties, but the frameworks we impose are 385 00:20:11.519 --> 00:20:14.960 always context dependent, never perfectly capturing. 386 00:20:14.559 --> 00:20:18.480 Reality, which leads to Ian Hacking's idea of dynamic nominalism. 387 00:20:19.079 --> 00:20:22.119 Naming and categorizing can actually make up people. 388 00:20:22.279 --> 00:20:24.880 It's a fascinating concept. Hacking argues that the act of 389 00:20:24.880 --> 00:20:29.480 classifying people defining categories like sexual orientations or psychological conditions 390 00:20:29.920 --> 00:20:33.960 shapes those very identities in society. The labels often created 391 00:20:33.960 --> 00:20:36.240 by experts influence how people are seen and how they 392 00:20:36.279 --> 00:20:37.599 see themselves. 393 00:20:37.160 --> 00:20:40.960 So visualizing data about groups can actually reinforce and solidify 394 00:20:41.000 --> 00:20:43.319 those labels, affecting policy and perception. 395 00:20:43.680 --> 00:20:46.559 Yes, it's a feedback loop, and Lorio extends this to 396 00:20:47.000 --> 00:20:51.200 making up spaces. She argues institutions like Statistics Canada through 397 00:20:51.240 --> 00:20:54.039 things like the Atlas of Canada, do something similar for geography. 398 00:20:54.279 --> 00:20:57.480 They create and reproduce a specific way of imagining the 399 00:20:57.599 --> 00:21:01.839 territory and its people, like Charles Booth's poverty maps, influencing 400 00:21:01.880 --> 00:21:04.440 how poverty was understood and measured exactly. 401 00:21:04.640 --> 00:21:08.720