Published: 16 July 2017
Gerry Gaffney: This is Gerry Gaffney with the User Experience podcast. My guest today is the Knight Chair in Visual Journalism at the School of Communication at the University of Miami. He has a B.A in journalism and a Masters degree in Information Society studies from, and excuse my pronunciation, the Universitat Oberta de Catalunya in Barcelona. He has recently defended his PhD thesis which is entitled "Nerd Journalism; How Data and Digital Technology Transformed News Graphics." He wrote The Functional Art: An Introduction to Information Graphics and Visualisation and, most recently, the excellent The Truthful Art: Data, Charts and Maps for Communication, which I recommend very highly.Alberto Cairo, welcome to the User Experience podcast.
Alberto Cairo
Well thanks so much for having me. It's a pleasure.
Gerry
And how did your thesis defense go?
Alberto
Well it went well. I was actually joking with some students the other day that it seemed to be that I was the person in the room that had taken the defense most seriously because all the members of the committee were… I mean they asked very good questions but they took the defence sort of casually so that the questions were quite easy. But I prepared for that presentation, I believe more than I have ever prepared for any other presentation in my life, and I have been doing talks about graphics for almost twenty years now, so that's something I guess. It went well.
Gerry
So you will be the other Dr Alberto Cairo.
Alberto
Oh, well there are many Doctor Alberto Cairos out there and this is another joke that it sounds really weird to be called Doctor now for me. Actually some of my friends are trolling me about it and, yeah, but what I tell them is that, I mean the real Alberto Cairos out there are medical doctors so there is my dad, for example, my dad is Professor Dr Cairo, his name is not Alberto but he is Doctor Cairo. But there is a famous Dr Alberto Cairo, who is an Italian doctor who treated patients in Afghanistan for many years and saved many lives, a true hero and he has a TED talk that I encourage everyone to watch. So just got to the TED talks website, look for Alberto Cairo and just be very aware that that is not me. That is not what I do.
Gerry
OK. Early in the book, the new book, you say that the term "infographic" has been hijacked by PR advertising and marketing people and you talk about infographics being used as "puerile posters used as click bait". How can we address this problem?
Alberto
Yes, well I need to provide some context to that part of the book because, I come from the world of journalism and news design and news graphics and I have been in this industry since 1997, and the term infographics is a term that we have always used to refer to data-rich, information-rich, displays that explain complex topics to readers. So if a reader says, if you think about these beautiful posters that the National Geographic publishes on a regular basis showing you, for example, cutaways of buildings or the internal function of the anatomy of a whale or you know a complex sophisticated topic like population changes or things like that and they do that through data maps and charts et cetera and they accompany those with textual explanations of what you are seeing, we call that an "infographic," that is what we used to call an infographic. The things is that around the 2010's more or less the PR and marketing industries started using the term "infographic" to refer to a data-poor displays like info posters that are in general, with exceptions, there are always exceptions, right? But in general they are a highly simplistic, they are over simplications of data and it bothers me a little bit that the term basically has changed meaning. And I used to refer to my own work as "infographic design" and nowadays I'm a little bit wary that that term will be misunderstand because most people came to equate infographics which is this kind of info posters that I don't really, I don't really enjoy and I don't really promote.
The other part of your questions is how can we address this problem? Well, first of all I don't if it is a problem. People should be free to use, to basically redefine the meaning of words the way they like and language is a living being, right? And it's constantly evolving and the meaning of a word today may not be the meaning of that same word ten years from now. So it's not really a problem although yeah I mean it bugs me a little bit, right? A word that I cherished and enjoyed and I liked years ago being used for something completely different. I don't think that the problem is the word itself, the problem may be more the fact that there are many people out there producing graphics that are perhaps too simplistic and that's one of the things that I try to fight against in the book, explaining that a good infographic, a good data visualisation should never be conceived as a simplification of data but as a clarification of data, which is something completely different. This is a distinction that I took from my friend Nigel Holmes who is a famous infographic designer and this is a distinction that he makes. He says, you know, infographics should never simplify, they should clarify. And by clarifying we all mean that, both Nigel and I we mean that sometimes in a graphic intended to explain something you need to reduce the amount of information but sometimes you need to increase the amount of the information that you show in order to put the information in context or to explain that information clearly to people. So this is the problem that we need to address, I believe.
Gerry
I guess that's something that occurred to me several times as I was reading the book as well. I mean in the UX profession we tend to talk about making things as simple as possible and several times in the book you do talk about this topic of the need to communicate being paramount. So there's a little bit of a tension there, isn't there, between simplification and clarification?
Alberto
There is a tension in there, but look, if you go to some of the best books about design out there, this is not a novel idea. I mean one of my favourite books about design is The Laws of Simplicity by John Maeda. In "The Laws of Simplicity" there's a wonderful quote in which Maeda says; "Simplicity is about subtracting the obvious and adding the meaningful". So simplicity is not just about simplification, it is about clarification; sometimes you need to reduce but sometimes you need to increase in order to clarify, right? It all depends on the nature of the information in the nature of the story that you are trying to tell. If the data and the story that you're trying to convey are complex, the representation will need to be necessarily complex as well. It should not be overly complex though. There is always a threshold there, right? I like to refer when I explain this concept to a quote that is commonly attributed to Albert Einstein, and I don't know if it is apocryphal, it may be, but apparently Einstein once said, referring to scientific theories; "Everything should be made as simple as possible but not simpler." Well that's the key thing; it should not be made simpler, right? Because if you make it simpler than it should be, then you are over-simplifying.
An example that I put to people to understand this concept is that sometimes when you're reporting data, people tend to… Let's suppose for example that you are going to report, I don't know, the unemployment rate in a country, in the United States or Australia or whatever and you report just the national average, right? A national average is a simplification of the data. It's a summary of the data set. You have all the unemployment rates in all cities and all neighbourhoods in Australia or the United States and instead of reporting every single data point, you calculate the mathematical average, the mean in this case and then you just report the mean. Well, is the mean a good representation of the data? It all depends, it depends on many factors. Imagine for instance that the unemployment rate in the United States is 5%, right? That's the average. If the underlying data shows you that the minimum unemployment rate in a country or in a state, in the United States is 4% and the maximum is 6% then the average, 5%, is a good representation of the data. It's a simplification but it's also a clarification because the range of values is really, really low. But imagine for example the minimal value is 0.1% and the maximum value is let's say 12% and you only report the average which is 5%. Then that average, that 5%, is not a good representation of the data because the range is so large. You also need to show the range, you need to show the minimum and the maximum just because those minimum and maximum contain very valuable information for the reader. So in this case reporting only the average will be an over-simplification but it will not be a clarification of the data.
Gerry
And of course if you've got small sample sizes among your data sets then that gets even messier doesn't it?
Alberto
Well it could be, yes of course. I mean that's another thing that we need to discuss when we do graphics, right? We need to report what that methodology is, the sample size et cetera. And there are many, many problems that I discuss in the book about the use of smaller sample sizes such as the fact that, you know, the smaller the sample size or the smaller the population, the larger the variation of the data will be obviously and the larger the error may be also as well.
Gerry
To backtrack a little bit, Alberto, how did you get interested in journalism and visual communication in the first place?
Alberto
Yes, alright so it's a little bit of a convoluted story so I'm going to try to summarise it. So I have always been interested in communicating information. Since I was a kid for example I started reading newspapers, print newspapers when I was 8 and I am 43 now so I have always been reading newspapers. I always enjoy them. I always enjoyed radio. So I got into journalism school when I was 18 and my original purpose was to become a radio journalist. I actually did internships in radio stations, reading the news in Spain. But then in the last year of my journalism degree in the B.A that I did in Spain I got a job offer, I got an internship offer, and it came from a professor. So this professor received a request from a local newspaper that was looking for a journalism student who could also draw a little bit. And I've always been a fan of comic books and you know visual communication in general. I have always enjoyed visual communication. My dad taught me how to draw a little bit. I mean I'm not a great artist but if you ask me to sketch something out, I can do it. I mean I can sketch something out, I'm not a Leonardo Da Vinci but, you know, if you ask me to draw a cow, it will look like a cow and not like a dog. So I could draw a little bit so this professor recommended me to this department, this infographics department in a local newspaper called La Voz de Galicia, and I got there as an intern knowing basically nothing about visual communication. They just got me and they educated me. They started giving me books and teaching me how to use the software, explaining basic concepts of visual communication. I learnt on the job and then later on, well I educated myself, obviously I started reading more, practising more, making many mistakes and trying to solve them. So it was a trial and error process that I greatly enjoyed and that took two decades up to this point. And then later on I got interested in, ten years into my career I started getting interested in not just pictorial representation of information like drawings of actual things but abstract representations of data, what we could call data visualisation. So I started getting into that work of reading and studying about statistics and about the digital representation of data, learning about cartography et cetera et cetera and then writing about it because the books that I have written so far, both "The Functional Art" and "The Truthful Art" are summaries of what I have learnt in the past twenty years through this process.
Gerry
Your love of journalism comes out repeatedly during the course of the book but do you think it's a profession that's under threat? Is there a future for it?
Alberto
Well that's a complicated issue. I think that all the discussion about the future of journalism has several dimensions. One of the dimensions is the future of the organisations that promote or that practice journalism and the other one is the future of the craft itself. They are connected obviously but they may not be as connected as people believe. I think that traditional media organisations are obviously under a lot of pressure, particularly newspapers, and not all newspapers have been able to transition nicely to the digital world. Many of them are shrinking in size and suffering a lot. We are seeing some signs of hope, particularly in certain large national-level organisations. In the US, for example, we could cite The New York Times or The Washing Post which in my opinion, even if they are still making cuts in the workforce, they are still doing very good journalism and they are hiring also people in areas like visualisation and data et cetera. But the newspapers that are suffering the most are, at least in the U.S, are often mid-level newspapers like regional newspapers, for example. Those are the ones that are suffering the most and many of them will disappear unfortunately.
On the other hand, there is also a new breed of online only data organisations that are doing good journalism and I would not say that they are thriving, I don't know what their numbers are, but certainly they are surviving and they are hiring and they are doing good work et cetera. I could quote, I could mention Vox.com or FiveThirtyEight or ProPublica, which is a non-profit media organisation here in the U.S and they seem to be doing quite well. Will those new organisations make up for the demise of traditional news organisations? I don't know. But the fact is there is a transition going on there so that's part of the conversation, the organisations or the companies.
The other part of the conversation is the craft, and I don't think that the craft is under threat. Quite the contrary is true. The tools that are existing nowadays, digital tools enable anyone basically to become a journalist and as I advocate in "The Truthful Art," "The Truthful Art" is basically a manual, a handbook of data journalism and visualisation for the general public because I do believe that anyone and everyone could, and I would dare to say ought to, learn to think like a professional and honest journalist, like, you know, we need to start communicating more honestly with each other. We need to start, for example, verifying the information that we find before we share it in social media. We all can learn how to communicate effectively using words. We all can learn, this is what my books are about, we all can learn how to communicate effectively using charts and using graphs and using maps. So potentially I can see a future in which more and more people will embrace values that were traditionally linked, connected or belong to the world of journalism. And the reason why I say this, and there is an example that I explain in "The Truthful Art," the reason I say this is that throughout my career I have found many people who do practice journalism without calling it journalism. So in the prologue of the book I believe, for example, I talk about a computer science student from Brazil, I am regularly from Spain but I worked in Brazil for some years, so I was reading the newspaper one day in Brazil and I read this story about a group of computer science students, Brazilian computer science students, who had created a digital interactive tools for the citizens of the city of Sao Paulo to see which areas, which neighbourhoods of the city got flooded more often and the tool was like this big map of the city and you could click on it and zoom in and zoom out and see, you know, which one got flooded more often or less often et cetera. I immediately reached for my phone and I phoned the Human Resources department, I was working for a media organisation at the time, I called Human Resources saying I wanted to interview this person, bring this person here because I want to interview this person. So they called this guy, he came over for an interview with me and his first question was; "Why am I here? I mean you're a journalism organisation, you're a news organisation and I'm a computer scientist. What am I doing here?" I said; "Well because you are doing journalism, what you are doing is journalism." Journalism is basically the craft or the task of getting information, shaping information, valuable information, and then putting that information at the service of the public, right? Verifying the information first obviously but then shaping it in a way that the general public can benefit from. That's what journalism is in a nutshell.
Gerry
And there are some people doing really, you mentioned that particular example, but there are people doing fascinating and often very beautiful work, and one of the things I particularly liked about the book is that it's very, very rich in examples. For example, that Italian designer, I can't remember her name, but she had that postcard project with her colleague in New York.
Alberto
Yeah that's Giorgia Lupi and Stefanie Posavec and the project is called, ah, I can look it up, Dear Data. "Dear Data," yeah, Giorgia Lupi and Stefanie, yes.
Gerry
And if anybody hasn't seen that, I certainly encourage them to have a look at it because it's just a beautiful piece of, it's an aesthetic and I guess, from a communication perspective, it's just a marvellous piece of work, isn't it?
Alberto
Oh, yeah it's a fantastic piece of work. So what they did, just to explain to your listeners, is that, Stefanie Posavec, she's a designer living in Britain right now and Giorgia, Giorgia Lupi is another visualisation designer who lives in New York, and they decided to begin this project that consists of gathering information from their own lives like quantifying things that they do every day, the restaurants that they visit, the books that they read, I don't remember exactly the details but they quantified all that and they drew data visualisations based on those data on postcards and they send those postcards to each other for an entire year and then they collected all those postcards in a book in the form of a book and a website called "Dear Data" which is absolutely wonderful.
There are many creative people out there. One of the great things about the present time is the amount of talent and creativity that we can see. I am really excited about the present actually.
Gerry
Now I often ask people who have written books who their intended audience was and I had that question in mind until I got to the last few pages of your book and you said who the intended audience was; do you want to tell us?
Alberto
Yes, so well the intended audience was myself, I mean it's like alright, so I kind of joke in the book that I write my books for my students mainly first, for journalists and graphic designers second. But ultimately what I try to do when I write my books is to try to remember how I was and what I didn't know myself ten or 15 years ago. So I try to go back in time like Dr Who in the Tardis, I go back in time and see myself and remember what I didn't know and I should have known ten or 15 years ago. And then I try to write in a way that myself could have understood ten or fifteen years ago. I think that's a very valid approach for any writer who writes a popular non-fiction or explanations of complex topics because you, yourself ten or 15 years ago is basically your students, the students that you are teaching in your classes, right? So yeah.
Gerry
You talked about the book earlier on as being a manual or a guide and just to I guess get very hands-on and practical for a couple of minutes, you do describe in the book four steps for finding the right graphic form; can you perhaps give us a brief description of those four steps?
Alberto
Yes, maybe three, maybe four depending on the book. It's actually a process. So the process that I follow to choose the best ways to represent the data… First of all they begin with an understanding of what data visualisation is about. Data visualisation is simply the craft of mapping the data onto the spatial properties of objects. Okay? That sounds a little bit abstract but I can give you examples. You can begin with you know ten numbers, for example, or one thousand numbers and then you try to represent them proportionally using object properties such as the length of an object, the height of an object, the area of an object, the angle of an object, the thickness of an object, the shade or colour of an object. The key thing is to try to represent those numbers proportionally through those aspects or features of objects. We call those things or those methods of representation, we call those things methods of encoding. The whole idea behind data visualisation is visual encoding, that's what it means. Now the key to understand though after that is that not all methods of encoding are equally effective depending on what you want to show. So the key thing in the process or thinking how to represent your data also includes thinking about what your graphic is for, what it is that you want your readers to do with your graphic. And in the book I explain, for instance, that for example when my students, and when I talk about my students, I talk about myself 15 or 20 years ago. When they find that data set that has a geographic component, for example, unemployment rate per region, or something like that, they rush and they create a map, right? A data map with shades of colour showing where there is more unemployment or less unemployment. Is that correct or is it incorrect? It depends, it depends on what you want to show. You need to ask yourself what is this for? What is the data map good for? A data map like that, a map that uses shades of colour which is often called a choropleth map, a choropleth map is used to spot geographic patterns in the data, large concentrations of unemployment, you know dispersion, concentration et cetera, et cetera. If that is what you want to show, you know, concentration or dispersion of the variable that you are plotting, then by all means use the map because the map is the best way to show that concentration or that dispersion in terms of a geographical area. But what if the purpose of the graphic were to, for instance, compare very accurately you know the regions of Australia or New Zealand or of the United States or whatever it is that you're plotting, comparing the regions to each other; seeing the first region, the second one, the third one, the fourth one and so on and so forth and then compare them to each other in terms of unemployment. Then the map would not be effective because the purpose of the graphic is not to show geographic patterns in the data, the purpose of the graphic is to compare things and to rank things. Therefore instead of a map, perhaps you should use a graphic that uses for instance length or height to represent the data, like a bar graph, right?
So these are the key things, the key component, the key element in choosing your graphic forms, asking yourself what the graphic is for and then finding the method that better fits the purpose of the graphic. We need to think about the purpose of the graphic.
Another thing is also, another component of this process is thinking about the audience. So because it is not the same to create a graphic for specialised audiences like statisticians or scientists et cetera than creating the same graphic for the general public. And what I mean is that for instance when you're going to create a graphic about a complex topic for the general public, you may want to shape the information a little bit differently or, more importantly, you may want to add more of a textual explanation to the graphic, like explaining what it is that people should see, giving people more clues, right?
If you're going to do that graphic for a specialised audience, right? For example think about a statistician doing a graph for other statisticians, that statistician can take a lot of things for granted because his or her intended audience already has some pre-existing knowledge that will enable them to read that graphic correctly. That pre-existing knowledge doesn't exist in the brains of the general public. TTherefore we need to make up for the lack of that pre-existing knowledge by adding more explanations.
So that's more or less the components that I explain in the books to choose these graphic forms.
Gerry
And of course when we're reading graphics or accessing graphics or visualisations, we need to be quite critical as well, don't we, and think about, I guess, the politics underlying it and the assumptions? I mean just to throw a question without notice at you; early on in the book you've got number of road deaths or something and it was by, you know you were suggesting the appropriate measure was by the number of vehicles in that state or in that area and I was thinking a more appropriate measure might be actually the number of residents in that area seeing that many of the people affected by road trauma are in fact non-drivers.
Alberto
Yeah it could be. That's a perfect… so the point I make in that part of the book is that that specifically is about the comparison between absolute or raw measures versus relatively measures, right? Like rates and things like that. And obviously in a case like that I think that both things are important, you want to show the raw numbers of deaths in traffic because each one of those deaths is a person, so you need to show the raw number of deaths. But at the same time you also need to show the relative measure. Now that relative measure could be anything, as you say, from the number of deaths per number of vehicles, number of deaths per one thousand people or something like that, just to take into account non-drivers. It could be anything of that but you would need to calculate that relative measure.
But that is an example. In the first part of the book I talk about certain, very simple techniques that we can use to approach graphics critically, not only absolute versus relative but also asking yourself about whether the graphic is using or displaying the right number of, the right amount of information, which is what I explained before with the dichotomy between simplification and clarification. Is the graphic including an enough amount of data to interpret this story correctly? That's what we need to ask ourselves. And in order to do that, media organisations or individuals who create graphics should disclose or we ought to disclose, disclose our sources, link to our sources and to the raw data sources so people can double-check the data if they want to obviously.
Gerry
I found Chapter 6 which is called "Exploring Data with Simple Charts" and Chapter 7 on "Visualising Distributions" to be very enlightening. In particular, you really do de-mystify statistics and I know I'm not alone in always having disliked and in fact been somewhat afraid of statistics and it seems such a shame, doesn't it?
Alberto
It is a complete shame and believe me that's a very common experience. I used to dislike statistics myself. I love mathematics. I was, I must say I enjoyed mathematics when I was in High School but then I completely forgot about mathematics and statistics when I was in college. I took a course in statistics and I didn't like it and I think it was because for some reason the instructor was not able to make the connection between the statistical methods that we were learning with how those methods apply to the real world. What I usually explain to people is that, and I think that this may entail a large, a very big change in the way that math is taught in schools at all levels, one of the things that I believe that we don't do well, educators in the area, is we tend to approach mathematics as a set of algorithms to solve problems and mathematics is more about, I believe, and in mathematics I include statistics also, you know although some statisticians will have some beef on that; they will say that statistics is not strictly mathematics, right? But mathematics and statistics, we should teach more numeracy. Numeracy is the ability or the skill of thinking rationally and critically about numbers. And numeracy is reasoning. It's not just applying, you know, raw equations to data, it's more about thinking about what those equations mean, what those methods are for and what they are useful for. And that is what I try to distil in the book a little bit, right? It's like the book includes a few formulas here and there and then I provide a lot of mentions of further books that people should read to learn more about the statistics because my book is maybe just a beginning, it only scratches the surface. But at least I try to explain things in a way, again, that I could have understood myself 15 or 20 years ago, right? This is how I would have liked my instructors fifteen or twenty years ago to have explained elementary statistics to me when I was in college.
Gerry
Something that I think will resonate with people in the UX world is your focus on finding stories. Now perhaps this is too big a question to cover today, I don't know, but I wanted to ask you how does one find a story in data and then present that story in a digestible format? I guess you've already addressed the second part of that, but how does one find the story in the data?
Alberto
Yes, well there is, alright so the answer is obviously an entire book, right? So, but it all begins with questions, right? So you begin with a data set, you download a large data set for instance of county-level data and that data set includes anything from unemployment rates and income levels and obesity rates et cetera. That data on its own means nothing. What you need to think about is what it is that you want to find there, what it is that you want to explore, right, whether you want to explore each variable on its own or if you want to explore the connections between the different variables that you have on your data set. You need to begin with some questions, right? So those questions could be, we could call them conjectures, right? And based on those conjectures you can, you could even create scientific hypotheses, right? I hypothesise that, you know, the connection between this variable and that variable will be such and such and now let's go to the data and see whether that conjecture stands or not, right?
And then what you can do with the data is not just look for the evidence in the data that confirms obviously your conjecture but also think about the possible ways in which your conjecture could be refuted. You know, you have to think about what variables or what numbers or what values could potentially refute that conjecture of yours and then try to find them and see if they are there because if they are there you will need to change your mind obviously, right? And then what you can do, I know that I'm not being very specific but in the book I talk about exploratory data analysis which is a very fancy term, EDA, created by a statistician back in the '60s and the '70s, his name was John Tukey, a very big name in the world of statistics who wrote a book with the same title and explained techniques that are simple, very simple techniques that you can use to explore data and that's what I tried to explain in "The Truthful Art." You begin with the data, first of all you calculate numerical summaries of the data like averages, means, medians, modes, you know quartiles and quintiles and deciles et cetera to explore the data and then you start visualising the data. You start visualising the range of the data, the shape of the data. You can also visualise the connections between the difference data sets using graphics that let you see the correlation between different kinds of variables.
So there are different methods in there that you can use to find potential stories in the data. But this process is only the beginning because one of the dangers about data is that it is very easy to find things that are the product of just random variation or error or things like that. So even when you think that you have found these interesting stories in the data, you should not publish them immediately. This is part of that journalistic thinking that I mentioned before during the interview. What I usually do in my process when I am doing graphics, after I have explored the data, is to write notes. I find interesting stories in the data like outliers or possible correlations between data and I write down; "There is an interesting connection here, there is an outlier here, why is this happening?" So I ask questions to myself. And then what I do is to consult with sources. So I go to expert sources. You know, if I'm doing data, if I'm working with data about education, I go to experts in education. If I am working with data about demographics, I go to the demographers or political scientists or whatever and I say you know, "Here is what I have found in the data. Can you help me figure it out, what is going on over here?" Or "Do you have any idea some things that I could look for in this data?" And in that process is when you can start shaping a story. It's in the combination of the exploration of the quantitative data and the very qualitative process of asking people that know much more than you do about the data that you're manipulating. And then you can start doing your visual story if you want, only after you have verified the findings.
Gerry
A bit of a question without notice, Alberto, it occurred to me several times in the book, I mean you talk in particular about interactive infographics, the types of things done for example by The New Times and you know they're very, very rich and enriching.
Now simple infographics can be made accessible by adding textual explanations but once you get something that's very rich and interactive it's difficult, isn't it, to keep it accessible to people for example who might be visually impaired. You do mention sonification at one stage in the book.
Alberto
Yes, yes and that is one of the concerns that I have in the future; how to make graphics more accessible to people, to people in general, to the public. There are maybe techniques, different techniques that may make graphics more accessible. Sonification may be one of them. The use of videos for sometimes for people also, right? For instance, in the books, in both books I talk about the relationship between presentation and exploration in visualisation; some graphics are intended to present information and some graphics are intended to let you explore the information.
Well those two strategies can be combined. You could begin, for example, with a short video explanation of the main stories behind the data, right? So people can get excited about the data that they are about to see then a short explanation about how to interact with the data that you're presenting to them and then you let them explore.
So there are always workarounds. Those problems we have not found them all obviously but it's an area that more research is needed in the future and that's exciting, I guess.
Gerry
Yeah, I found a funny mistake in your book, Alberto.
Alberto
Oh there are many, many of them, even the tonnes of people who read it and helped me copy edit it and... alright, so what is the mistake?
Gerry
[Laughs.] It's referring to mammalian brain weight.
Alberto
Mammalian brain weight?
Gerry
Yeah you talk, there's a graph under a heading called "Data Transformation". I hope you have, you've seen this one have you?
Alberto
I have not found that one, right so I will write it down.
Gerry
I have to tell you, page 261 the Y-axis shows brain weight of different animals and instead of it being in grams it's in kilograms. So a human with a 1700 kilogram brain. [Laughter.]
Alberto
I will, I will write it down. I will write it down. Yeah there are many little mistakes here and there and it's actually amazing how easy it is to end up with typos in a book even if you have read the book like ten times even, you know the book was read by three statisticians who are friends of mine and they copy-edited it very heavily but and then it was read by like 10 or 15 other people who also were kind enough to send me corrections. But in the end, always, always, you will always end up with mistakes that can only be addressed in a potential second edition I guess.
Gerry
That's right. It's very final print, isn't it? I mean once that physical book is....
Alberto
Exactly and as you well know, the first thing that happens when you open up your own book on a random page, you always stumble upon a typo.
Gerry
That's right.
Alberto
That happened to me on my first book. It happened to me on my second book, I say; "Damn, they have printed already thousands and thousands of copies of these things and this typo is here." Anyway.
Gerry
Perhaps following on from that, there's a real sense of joy and fun in the book and it's something I certainly didn't expect approaching this type of book.
Alberto
Yeah, well that's, again, that's because I, remember that I'm writing to myself twenty years ago, right? So I like, I enjoy, alright, so I like reading a lot, I enjoy reading a lot and I enjoy reading people who can write clearly about complex issues, right?. So others and funnily the best writers about complex issues are often not journalists, they are often scientists, right? So I don't know, books about biology or revolutions, for example, explaining the science of revolution by natural selection, the best books out there are often written by biologists themselves, right? Richard Dawkins or Jerry Coyne or people like that. The best books about physics, for instance, they are written by physicists who are able to bridge the gap between their specialisation and the general public and they are able to distil and condense and summarise their knowledge in plain language; plain language, that doesn't mean that they are speaking down to people. They are just using language that everybody can understand; they are not dumbing down the message. They are just making the message more accessible by making it devoid of jargon but without losing depth and that's the critical thing and it's one of the hardest balances to find; how do you convey complex messages without oversimplifying them but at the same time avoiding specialised jargon or, if you need to use jargon, how to explain it clearly so people can understand it.
So I try to use, you know, writers that I have always enjoyed reading as examples. I copy, right? I copy shamelessly, my style. I don't plagiarise, obviously, but I take those people as inspiration of how to communicate, how to communicate clearly. And then I try, as much as possible, I try to convey my own enthusiasm for the field. I do believe that, you know, data and numeracy and statistical thinking and reasoning and then the visualisation of evidence and data is a craft that, again, everybody can learn and everybody can take advantage of and it's something that makes our lives better. I mean it lets you discover things that you may not find otherwise and how wonderful is that, to discover new things of our surrounding world, right? I believe that there is no more, there is no higher joy than the discovery of new knowledge. So perhaps that transpires a little bit in the writing.
Gerry
I just read a review or a short article in the latest issue of New Scientist about somebody called Mary Somerville who wrote a book on the connection of the physical sciences that was published in 1834 and apparently was enormously popular and one of the very first populist science books and, for what it's worth, just an aside.
But Alberto, for someone like myself perhaps with a limited gift for visualisation, what should they do to develop or improve their ability to create or direct visualisation work?
Alberto
Well the first step is to get a little bit acquainted with the language that is used in visualisation; so learning a little bit, obviously reading a couple of books about visualisation. The second one is to learn one tool, a tool or two and there are many free tools out there nowadays that one can use to do graphics. And then begin small, right? It's like don't try at first to do something with the same level of quality and complexity as The New York Times because that's the long term goal, right? Begin with a bar graph, with a line chart, with a series of maps, something like that; those are not that hard to, they are not hard to create with current software tools. And so start practising, start making mistakes and then putting them out there so people can critique them and give you suggestions. One of the great features of the visualisation community is that it's very open, it's very welcoming. It's very critical but it's very critical in a very constructive way so if you Tweet at me, for example, with a graphic that you have done and if I have the time I would take the time to sit down for five minutes or ten minutes and give you some suggestions. So that's something that I do and not only myself, there are hundreds of people who do the same thing, perhaps not hundreds but certainly dozens of people who do the same thing online. And the key thing is to start working and start practising and after learning one tool or two and then certain principles of data visualisation and also data thinking, that's a very important component of the craft, the numeracy part, the thinking about the numbers critically and then learning a little bit of writing, that's something that is really important in visualisation, writing, learning how to write a good title, a good introduction, a good explainer, a good annotation here and there; that also helps a lot. Visual communication or more precisely, data visualisation, is not just about visually representing data, it is also about writing the text that puts the data in context or that explains to people what it is that they are seeing or what it is that they should be looking for in the visualisation that you have done.
Gerry
I'd love to keep talking to you, Alberto, but I realise we've used up the time that you promised me but obviously this is a huge topic and it's great to hear somebody who's so passionate about the work they do and I would certainly highly recommend the book for anyone who's even vaguely interested in this area. It's truly inspirational and a beautifully written book and my only complaint about it was that you know the physical book is a bit small. It could have been a bit bigger for me to read some of the examples. But it's just a marvellous piece of work.
Alberto
It's a challenge with publishers so it's like.
Gerry
Yeah for sure.
Alberto
Because if we make the book bigger, it would be a coffee table book, not a book that people will read, because it's too large. If we make it too small, then the examples are too small. So the trade-off was to make it mid-size and then just put all the URLs in there so people can see the examples in their own computers. So it's not a perfect trade-off but as all trade-offs, it's the best thing that we could do unfortunately. The points of discussion that we had, when I started these series and the third book will have exactly the same size obviously. But it was a discussion that we had and the conclusion was let's do something mid-size because it's not a perfect solution for anyone but it's good enough solution for everyone so, anyway.
Gerry
Yeah and I'll remind listeners that it's called The Truthful Art: Data, Charts and Maps for Communication by Alberto Cairo and really if you're at all interested just go out and buy it.
Dr Alberto Cairo, thank you so much for joining me today on the User Experience podcast.
Alberto
Thank you so much Gerry, thanks so much for having me and your kindness.