Speaking at this year's Thought For Food
Last week on July 23rd, by invitation of Mrs Kitty Leering (thank you!), I spoke at the Thought For Food Academy opening ceremony and it was one of the most important experiences I had this year (maybe the last 2).
It was a talk/exercise on Data Visualization (one of my biggest passions) but it meant much more to me than just the talk itself.
First of all I had never done something like that before, second it wasn't just any talk at just any event, and third it was in english.
I mean, I can write and sometimes I even find myself thinking in english but I don't speak it daily so, yes, it was an extra item on that challenge. But I'm getting way ahead of myself here so let me take a step back and tell a little story of how I got there in the first place.
Helping out
I met Denise Grimming, an awesome producer from Rio, many years ago when I was a fresh-out-of-college-freelancer helping design an exhibition of Pre-Columbian Art at the CCBB (this cool venue here in Rio) called "Por Ti América". Over the years she contacted me whenever she needed any help design wise and, although most of the jobs didn't have anything to do with what I was doing at the time it as always a pleasure to help her out. She contacted me this year for some help again and we (Café.art.br) just had too much on our plate and the project didn't involve our thing — information design — but there was no way I'd let her down. So I got my good friend, Bruno Ventura from Disarme Gráfico, to help out and I'd supervise now and then.
We met Denise and Kitty at the Museum of Tomorrow, Rio's most important museum today, and learned that the job was to help produce not only all the design collateral but also the scenography and lighting for the main event at the Museum. It was a perfect fit for Disarme's talents so I was more than happy to have gotten them onboard. We were accompanied by designer Joana Croft, the designer from Oupas! Design who was responsible for the scenography (we later learned that they are THE masters of cardboard — check them out).
After checking out the venue, taking notes and pictures with Joana we went to lunch. I usually don't do that but recently I've been interested in talking to as many different people as possible so I decided to go. Over lunch Kitty and I talked about TFF — What it was exactly, what was it's purpose, how they defined their goal of "How do we feed 9+ billion people by 2050?", how it was cool that they framed it as a very open question and how that could lead to a wide range of different answers, etc… Really awesome stuff, you know? It got me really excited about TFF and at the same time feeling sorry it didn't involve what we do.
So what do you do?
At the end of our lunch, over coffee, she asked what was it that I did and I think I probably said "data visualization" and without catching my breath almost certainly started blabbing about design, data and technology and how we combine them to tell stories and shed light on complex subjects and all the things I say about what I do because I get overexcited about it... (breathe).
And that was it.
Everyone said goodbye, I went back to the studio and sent our company presentation/portfolio to Joana and got back to work.
A few days later…
I was driving to the studio as I do every morning when a notification popped up on my phone. It was an e-mail from Kitty. I pulled over to check it out: she said Joana had forwarded our portfolio to her and that she liked our work a lot and asked if we'd be interested in contributing to a workshop on the Big Data and GIS track at TFF Academy. I thought "HELLLLL YEAHHHHHH!!! OF COURSE!!!" and properly answered a:
"Yes, we’re in. :) It would be our pleasure to work with you and TFF. I’ll put together a few possibilities for the workshop exercise and run them by you in the upcoming weeks…"
She then introduced me to Brittany Dahl from ESRI Australia who was responsible for part of the workshop (and was wonderful helping us out — thanks, Brittany!) and we talked about what they would do and what was the best way we could fit in and make the transition between the different parts of the workshop run smoothly.
First drafts
It took me a few days to put some ideas together and the best one involved the Data Portrait concept from MIT Media Lab. Basically it's a visualization of some kind of personal data so I thought it would be cool to use TFF participant's data and show them how that data can be translated visually — and drawn by hand. To explain what that meant and heavily influenced by designer Giogia Lupi's work I came up with this:
So in this example I made up each "portrait" represents a person and each of its elements represent an answer they gave to a question. Reading the key you figure out the first one is female, likes to wake up early and mountains, has 3 siblings a masters degree and is from South America while the second is male, likes the ocean and to stay up late, has no siblings, a PHD and is from Europe.
I purposely didn't specify the inner workings of the workshop because I believe collaboration is the best tool for design projects so I wanted Kitty's and Brittany's input. I figured it would involve designing the portraits, or at least some of their visual elements but was pretty open to even trashing it if the idea didn't fit with their purpose.
Getting friends to help out
Before they could answer I asked a dear friend, Julia Giannella, to take a look at the idea, give me some feedback and come onboard to help me get the workshop ready to test with some people before the TFF Academy event. She was awesome enough to not only say yes but to say we should have something set up to gather the data we'd use and also to send me a description of a class she taught for her PHD that was similar to the Data Portrait idea and involved the Quantified Self concept. A class which she masterfully called "Selfie de Dados" (hint: that's "Data Selfie" in Portuguese). WOW1.
I also got my friend and business partner Henrique Ilidio to take a look as well. He was already full of client work but depending on where the workshop would head he was the one I trusted with the visuals. He gave some very cool feedback about how we could make it a fun workshop, plus tips on the design and data collection processes.
Feedback — Breakfast of champions
Brittany answered first saying she liked the idea of designing the portrait with participants in the workshop. "Each individual can then have ownership of what they create at TFF — and, I can see this being a great way to share information during the conference!"
She went on, perfecting my initial idea: "the data portraits could be stuck to their name badges, so that when they talk to other people after your workshop, they have something insightful to share!". WOW2.
She also had the same idea as Julia about setting something up to gather the data and offered to set up a survey. How cool was that?
So Kitty finally answered with one of the best e-mails I got this year. It went something like this (I did take the liberty of condensing it):
"I LOVE this!!! Could we do this as a start exercise on the first day?"
To which I thought "HOOOOLY SHIIIIIT!!! YEAAAAAAHHHHHH" and showed the e-mail to Julia, Henrique and the whole team with a huge smile on my face. Then, of course, very professionally replied:
Wow! We’ll be more than happy to do both.
Then she, just like Julia, Henrique and Brittany before her, added to my idea telling us how she once did a networking game where every single participant received a badge with an animal that was chosen based on their profiles. They then had to talk to each other to find out things they had in common in order to figure out why they were given that specific animal. That's WOW3.
Talk + Workshop
Of course the workshop still needed to be taught so we now had to finish the Portrait idea and come up with another workshop. I immediately got to work on the Portraits talk while Julia took on the new workshop idea.
Julia came back with a new idea for the Big Data and GIS workshop. It was a hands-on activity on data visualization for spatial data analysis. Further information and link for the presentation slides can be found here. Let me just tell you: her hand-drawn slides were amazing!
Hours and hours later, and in collaboration with Henrique and Julia, I had the mechanics and questions down and a basic idea of what I was going to say. It was a talk but people still needed to feel they owned their Data Portrait so, instead of getting 180 people to design their portrait in 30 minutes — which would be impossible — we would have the designs ready beforehand, gather the data live during the talk, process it through software to make printable files, print and make them into pinback buttons. Easy peasy. :)
Visuals
I didn't focus on the visuals at all until the very end not because they're unimportant but because (1) the TFF visual identity looked awesome (courtesy of designer José Silva, from Origo Design Studio) and we had already decided we'd use it on the portraits, (2) Henrique was the one that was going to take that on and if there's any designer I fully trust its him, and (3) because in this project it made no sense at all to get to them first.
In our design process, what usually happens is that the visuals are the very last thing to be considered. Not all the time, but most.
Wrap-up
After we had the visuals down, the data collection working, the variable data process set up and the button makers hired we tried out the talk a bunch of times until the words felt right.
There was just this one thing: "Data Portrait" didn't sound right. Too formal, too precise, too not TFF and too not us. Then I remembered Julia's class "Selfie de Dados" and knew we got it.
The "Data Selfie" was born.
Below are the words I was supposed to say on my talk. The stage isn't my natural environment (yet) so I think I kind of did a lot of things different when I was live but kept the message.
The TFF Data Selfie talk
Hello everyone and good morning!
Thanks Thought for Food, Christine Gould and especially Kitty Leering for the invitation to speak. It’s a great honor to be here today.
I'd also like to thank Brittany Dahl from ESRI Australia for all the help getting the survey up and running. And lastly thanks to my team, Henrique and Julia for helping me get everything ready for today.
For those of you who don’t know me — which is all of you I guess — I am Leandro Amorim.
I’m a graphic designer, and head of Café Information Design, a studio that specializes in interactive data visualization. I’m also partner in Tapronto Creative Coding, a web development agency that's also focused on data-driven applications.
And being in charge of these two businesses puts me in a very special place — right between design and technology — which I think is a pretty cool place to be in at this day and age and I’m going to try and take advantage of that in todays talk.
More recently I was elected member of the board of directors of the Brazilian Graphic Designer’s Association, ADG Brasil, and I’m trying to use it as a platform to get more designers interested in technology and data. So, that’s it for credentials.
Now for the more important stuff: My passions!
Like everyone, I’m passionate about many things but the three that I think got me here today are:
1. Design (love everything about it)
2. Data-driven stuff, and the range of possibilities that data opens up, and
3. Sharing knowledge
I deliberately don’t use the word “teaching” because I feel it’s too one-way and formal for my taste but if you prefer to use it, it’s all good.
But we’ve got little time so enough about me.
As you all know Thought For Food is about community.
It’s about engaging, empowering and uniting people towards this one huge common goal.
So I’m here today to show you one of the ways we make data and technology, which might seem like very cold and distant things, a little bit more human. And use them to help stimulate interaction and conversation. To get everyone talking.
And the way we do that is by adding a pinch of design.
We’re calling it the Data Selfie.
I’m gonna to go into some background in where that came from so I'll mention some names and projects but they'll all be on my last slide so please don’t worry about writing anything down now, ok?
This Data Selfie exercise you're about to take part in is based on the “Data Portrait” concept by Judith Donath and Fernanda Viégas from the MIT Media Lab and includes some insights from the "Selfie de Dados" class taught by Julia Giannella - who, by the way is right there, and will do a cool data workshop in a few hours.
More recently, the Data Portrait concept was beautifully revisited by designers Giorgia Lupi and Stefanie Posavec in their book “Dear Data”. It’s a year long diary filled with very non-traditional data visualizations — all done by hand — on the back of postcards they mailed each other weekly.
If you’re interested in the less traditional side of data visualization get the book, it’ll really open your mind to new possibilities. And no, I’m not getting paid to say this. It’s just really good and really fun.
But back to our talk, we tried our best to boil all this inspiration down to the Data Selfie.
One important thing to know beforehand is that we won’t represent your face.
We call them Selfies in order to shift the way we think about data. While the terms "portrait" and "data visualization" evoke precision, clarity and objectiveness, a "selfie" on the other hand, is meant to to be less complex and to convey something that's meaningful about your personality and your state of mind.
This shift in perspective also affects our approach as designers. We might take some liberties we wouldn’t when working in a more traditional dataviz project, we could maybe flirt with art a bit… Who knows! It’s all good.
With all that freedom the Selfie can then become this very non-traditional graphical representation of data, and the way we get there is through something called Visual Encoding.
I don’t want this talk to get boring so I came up with an oversimplified way to quickly explain what that is: Take for example a non smoking sign. It’s information (data) that's been visually encoded. Now take a QR Code. It’s also information and it's also been encoded visually.
BUT… One needs to be understood fast and across different places and cultures, the other one well… there’s just no way you’ll understand it. It’s way too complicated for our brain to identify the patterns and transform them into something that carries meaning.
The Data Selfie intends to be somewhere in between these two. Not so easy to understand as to be commonplace and boring, but also not so hard, otherwise it wouldn't be able to pass along any information.
Got it?
So when we visually encode data we build what we call a Visual Dictionary, which is like a key to a map or a graph that you need to use in order to figure out how to read it.
For our exercise today we’ve already encoded every piece of data into our visual dictionary. What we need now and what you can all help us with is the data!
The data will be gathered through an online survey you will all fill out in a minute, then we’ll clean the data and process it through software using the visual dictionary we designed in order to build everyone’s unique Data Selfie.
Please grab your phones…
…and go to the following URL (unfortunately it's down now but you can see the questions here).
If you need wi-fi use the password given to you by TFF staff earlier today. And if everything went right you should be looking at this form.
Quickly fill out the form so we can get moving. Each entry will represent an element of your Data Selfie so please be sure to fill everything out!
Of course, some of you might not have your phones with you. In that case we gotta help each other out here. If you don’t have a mobile device on you, please raise your hand. Everyone sitting beside someone with their hand up please lend them your phone once you are done so everyone can fill out the form. Just hit refresh and fill out the form again.
Now that everyone’s set, let’s go through the process of designing the Data Selfie…
…from setting it's goal to the design decisions that went into visually encoding each variable.
The first thing we had to do was to clearly define our goal. Our goal was defined as "to use data, technology and design to spark a conversation”.
Now what questions from our survey will more likely spark a conversation? The more important a question is, the more identifiable and memorable it’s visual representation should be. So we analyzed the importance of each question, and ranked them.
Of course the rank will vary from individual to individual. For example I love to talk about people’s views of the future therefore I’d rank this high on my list and the next guy might not like that at all so he’ll rank it last. There are no right or wrong answer here, just different decisions that will lead to different outcomes. And as designers we understand that in order for this to be broad the decision-making process can’t be personal so we got friends and colleagues to help us with the ranking and this is what we came up with.
Our first and most important question is: Are you available for a conversation?
This is the most important question for it can make or break our goal. We have 3 possible answers (variables) and we decided to be bold right off the bat: we won’t let anyone get away with “Nope, ain’t gonna happen” without at least giving it a second thought. For that reason we only encoded the first two variables to give any of you who marked the third some time to think it over.
If you decide you still don’t want to talk then it’s ok, just hide your Selfie and no one will bother you!
This was pretty simple and we encoded the variables with color. "Chat away!" as a blueish green and "Maybe, depends" as a pinkish purple.
Our second question is: Which would you consider your area of expertise?
This is kind of a compound question when combined with the third one:
Which other area would you like to learn more about?
They are both great conversation fodder so we ranked them high on importance. The variables are the same for both and the two different questions hint at some kind of an order. Like one is my background and the other a possible future… These conceptual hints are really important when deciding how to encode the variables as you'll see later.
There are 7 variables and we defined a color for each one plus a shape to contain the colors and different positions for the shapes: one in the background and one on the foreground.
Our fourth question is: How do you feel about the future of food and agriculture?
Another interesting question conversation-wise. We have 3 variables, representing very simple concepts so we decided to broaden our visual repertoire and design minimalistic icons. A plus sign (+) for "Optimistic", a circle or zero (0) for "Concerned" and a minus sign (-) for "Pessimistic".
Our fifth question is: What continent are you from?
Maybe it might spark a conversation or two but it’s not that compelling so it's ranked towards the middle. Also 7 variables to be represented visually and we already used color so this time we decided to use something else: shapes, one for each continent.
If there’s anyone from Antarctica here, you got a special treat: an icon. Plus that's very interesting so we gotta talk about that over lunch.
Our sixth question is: How long will you be staying in Rio?
This is a prioritization question. For example "You leave tomorrow so let’s talk now!", or "Oh, you leave in 6 days, let’s grab coffee the day after tomorrow — no rush."
For our Selfie we decided to use the size of a shape — a circle — to represent the 3 ranges of days: a small one for "2–3 days", medium for "4–7 days" and a large one for "Indefinitely! ".
Now for our last question we have: What’s your timezone?
It might have no apparent meaning but we thought it might be fun to try to cross-reference continent and time zone in your head to try and guess where someone is from. We also thought of a cool graphic representation for it.
We designed a kind of clock face with time zone zero at the top, positive time zones to the right and negative to the left and we grouped the time zones in 2 hour blocks. We also positioned the continent shape relative to the time zone range but that's giving away too much info on the Selfie so you’ll see how that looks in a minute.
So you might be thinking: What ever happened to "What best defines your attitude right now?", the first question on our survey?
Well, it’s always great to start off a survey with a smile so we put that in but it’s actually irrelevant for our goal because the Selfies will only be available after 12! We decided to drop it. It's one of those decisions I mentioned earlier we must make.
OK! Now that you know how we designed the visual representations for every variable, let’s see how they all come together in the TFF Data Selfie.
This is my personal TFF Data Selfie:
And this is a fictional character we created just to show you a contrasting Data Selfie: Mr John Green.
You'll see that each Selfie will have its own unique flavor. We did the math and found out there are 68.796 different combinations, plus an almost infinite number of names!
I’m reaching the end of my talk…
…so I’d like to just say a few words of advice if I may. On design, on data, and maybe on life too…
1st advice: Clearly define your goal.
That will keep you focused and make all your work much easier plus it will back any tough decision you might need to make. And trust me: you always will need to make tough decisions.
2nd advice: Don’t use traditional visualizations just because it’s what’s in Excel.
And by Excel I mean your toolkit, whatever that may be. Don’t let your toolkit dictate what you can and cannot do. Use your brain and your hands, use pencil and paper, scissors, glue, whatever, to come up with new and interesting stuff.
Make mistakes, but preferably make them slowly. Softwares make mistakes too fast for us to actually learn something in the process. Try to use drawing as an extension of your thought process. It'll really make a difference.
3rd advice: Don’t use non-traditional visualizations if a bar chart does the job.
If you think it through and figure out that a bar chart will do the job then, by all means, please use it.
Just think first, ok? Always think first.
4th advice: Always seek beauty.
You're probably thinking that's weird advice. But, think about this: when everything else is equivalent between two choices, beauty always wins.
That’s why we used the TFF Visual Identity designed by our friend José Silva over there, as the base for the Data Selfies’ look. We just knew that — eve if we did everything wrong — they’d be beautiful. But we did good, trust me!
5th and last advice: Really talk to people.
Like for ever, but especially today, take some time either during or after lunch to look at each other’s Data Selfie. Spot what you have in common as well as each difference and then use these bits of information to start a conversation.
And I don't know if it's just in Brazil or it's a worldwide thing but conversations keep getting shallower and shallower so try to make it a deep conversation if you can!
It will not only and enrich your TFF experience but your life as well.
And in the end… that’s what I truly hope that the combination of design, technology and data will do:
Enrich all of our lives.
Thanks so much everyone! And thanks again Christine, Kitty, Brittany, Henrique and Julia!
You all have an awesome day!
Thought For Food
Engaging and empowering the next generation of innovators to solve the complex and important challenges facing our food system.
https://thoughtforfood.org/
Café.art.br
Information Design Studio. That's us ;)
www.cafe.art.br
Judith Donath & Fernanda Viégas
Data Portrait concept
https://www.mitpressjournals.org/doi/10.1162/LEON_a_00011
Julia Giannella
ESDI — Selfie de Dados
www.juliagiannella.com/selfiededados
George Lupi & Stefanie Posavec
Dear Data
www.dear-data.com
