Notes from AIGA Fresh Dialogue 24: “In/visible” with John Maeda

Notes from the 24th installment in AIGA’s Fresh Dialog lecture series, In/Visible: Graphic Data Revealed.


How do design choices affect the integrity of the data being portrayed? Can information graphics and the designers who create them ever claim pure objectivity? John Maeda, newly appointed president of RISD, world-renowned designer and innovator, will engineer a conversation with three designers who specialize in visualizing information. Steve Duenes guides the New York Times graphics group in print and online; Andrew Kuo cleverly quantifies feelings through his meticulous charts and diagrams; Fernanda Viégas explores the social side of visualization at IBM. Together they will examine a range of current and anticipated trends in visual journalism: judicious simplicity over seductive complexity, data- and story-driven visuals, and the increasing demand for dynamic versus static information visualization.

John Maeda, MIT

  • Growing up, I was into art and science. This book really turned me on: paul rand – designers art
  • went to art school in the 1980s, got away from computers. vacation!
  • Yale art student quote from the 1980s: “i came to learn design, not computers”
  • Started doing experiments
  • “Remember when photoshop switched to layers and got hard?” I had a pixel I couldn’t erase; I went through layer after layer trying to find it. It was piece of dust on the screen.
  • JM started making physical objects: iPod fish, etc.
  • “Museums are great but the world is even better” started making found stuff.
  • Circle triangle, square, right? I didn’t get it until I got older. Circle is smooth. Triangle is the punk rock one. Square is boring and solid. Found stones on Cape Cod (2007).
  • Simplicity became appealing; found the letters “MIT” in “siMplIciTy” (also “coMplexITy”)
  • Career crisis: no “RISD” in “simplicity”. But: found “RISD” in “RaISon D’être”
  • Introduces the speakers, finishes with “Steve, COME ON UP!” (sounding very “Price Is Right”)

Steven Duenes, New York Times


  • What do you do? we’re the dept. that does pie charts.
  • also trying to do other stuff: stock market volatility market volatility
  • sometimes use new forms in familiar guises: a chart tracking court decisions in the Terry Schiavo case looks like a fever chart
  • word frequency – tag clouds of words in DEM/GOP convention speeches, Bush SOTUs


  • Maps: ground zero, bush/kerry contributions by address
  • distinguishing aggregate data by showing Missouri’s Clinton / Obama results. Like the cardioid map of purple counties.


  • mostly 2D vector drawings, no fancy 3D stuff ground zero: adobe dimensions & illustrator
  • not inventing, not going beyond what we know
  • 3D often means guessing and inventing. Pen and ink can be more honest
  • with no warning or change in pace, a (final) slide comes up saying “THANK YOU.” in helvetica. Classy.

Andrew Kuo

  • I went to RISD but I was a bad designer. Too sloppy.
  • started blogging at – wanted to disassociate from art identity
  • he’s the guy that does infographics of indie rock shows. (McCarren Park Pool)
  • “i’ve been doing this x-axis stuff lately”
  • “i was inspired by my friend brian chippendale, he goes diagonal, you read it like a snake” (jb: like cantor’s diagonal argument!) this chart is read on the diagonal
  • “as soon as you extrude something, it confuses people. i like that”
  • “basically this is like a bar graph, done in a really bad way”
  • “enthusiasm is a really big part of what i do” – “7 nights of bright eyes. being a fan is important. it may sound like hell to you, but it was heaven for me.”
  • w x-axis you can say two things at the same time that contradict each other and I really like that
  • meta-reviews: redact by color
  • reviews of txt msgs
  • reducing everything is really liberating
  • animal collective video: really low res, fit for the form. based on a traveling wilbury’s
  • band: hex messages
  • inspired by an artist that I love (sophie call)
  • chart of me getting drunk: happy, blind, grumpy, happy
  • ends with LOLCAT: “‘sup”

Feranda Viegas – IBM

Info Visualization in the 1990s

  • gaudy colors, scientists (by experts, for experts), all men
  • can we make a space where people can share their infovisualization work?
  • 14k visualizations & “a bunch of users”
  • word tree: built a txt analysis tool because so much txt was being analyzed
  • “Most people think of data as numbers, but txt is data too. Text is ripe, its a huge opportunity for types and graphs.”

How are people using the site?

  • user “crossway” made pairs of co-occurrence biblical characters, made a social network
  • made it across the blogosphere, boingboing, etc
  • analysis of results made it back to many-eyes
  • institutions are coming to Many Eyes: sunlight foundation, Lessig’s corruption lecture

Personal Data

  • ME is completely public, but they get personal data all the time
  • 1st example looks just like Andrew Kuo’s work. What’s the difference? Its the difference between fine arts and design (expression and toolmaking?)

Panel Discussion

  • JM: FM, are you gonna put steve out of business?
  • FV: no way, we love steve. the NYT does the craft. its being done by experts. on ME, the idea is “if you bring these tools to everyone, what will they do with it?” I think its a very different thing.
  • JM: Pete, is she right?
  • PD: Its like the difference between Wikipedia and the NYT. We’re telling stories, going out and communicating stories. THey’re a marketplace.
  • JM: FV, will you put AK out of business?
  • AK: yes and no. the more you make these things, the more writing comes into it. if you can be clever, or you can tell a story, that’s where it is
  • JM: SD, what do you mean by “sometimes you have to lobby for that graphic”
  • SD: bribe’em w alcohol. we were talking about a series of maps that we were playing with to describe the result of the 2004 presidential election editors are conservative w charts, not too comfy w something new. If you show a map that displays county by color, its skewed towards large counties and you get the wrong idea.
  • JM: AK, you get away w murder. lobbying?
  • AK: no, its shocking what I get away with. They (don’t?) know what I’m doing, I give it to them, and they all they do is correct my spelling.
  • JM: SD, are there fashion trends? What’s in this year?
  • SD: hard to say
  • JM: FV, are you a scientist?
  • FV: I don’t think of myself as a scientist
  • JM: i saw all the vegetation (trees; plants as metaphors for info vis techniques), but are there other planets up there?
    word tree
  • FV: we wanna get there, if people wanna contribute more visualization techniques, but most of the hurdles are legal
  • JM: you should offer a bounty
  • JM: AK, what’s up w design vs art?
  • AK: artists are embarrassed to admit they went to design school. Design is more worthwhile because you help people. With art you just help yourself.
  • JM: in the AK world, what color is art and what color is design?
  • JM: SD, what makes a good pie chart
  • SD: you never want too many wedges. you never want only one wedge
  • JM: FV, is Many Eyes a science project?
  • FV: no, I think of it as a design project
  • JM: yeah, scientists aren’t allowed to be happy. You were pointing to the humor at Many-Eyes
  • FV: yeah, its really important. people have created diagrams of what color relates to what emotions
  • JM: “a picture’s worth 1000 words, but it takes words to say that” reaction?
  • SD: i’ll show you a pie chart
  • JM: SD, you mentioned adobe dimensions. What’s the deal?
  • SD: we keep an old mac. its like a 3D rendering program for people who’re very slow
  • JM: what’s a digitally manipulated digital diagram?
  • SD: one that suggests more than you do
  • JM: andrew’s work is like visual op-ed. when will there be a whole newspaper of visual op ed?
  • JM: how many people to make ME?
  • FV: 4. No RISD grads though.
  • JM: you all use rollovers. my friend could never stand interactive graphics because he said “its like a dog sniffing for a bone on the screen”
  • SD & AK: what’s wrong w dogs?
  • JM: why scatterplots?
  • FV: matt erickson gave a keynote and said “no more scatterplots” because the readers couldn’t understand them. But wait! its the easiest way to show correlation between two variables. but its not true. people are good at maps, they’re good at time-series. But if yr a pro, you think that scatterplots are easy, but they’re not.
  • JM: AK, what’s yr favorite kind of plot?
  • AK: god, the bag of tricks is really small. i like the circular thing. its really important to be happy about what you make. An example of an AK circular piece
  • JM: SD made the pt about the “honest diagram”. talk more about that. are computers dishonest?
  • SD: if the operator is dishonest. there’re a million ways to represent, that you can distort the data. zero base, make the change look more than it is, etc. So we work through iterations until we feel like its honest
  • JM: there was that moment when I left the computer and was working w ink and reached for the undo key. Is ink honest?
  • AK: no, but getting away is key
  • JM: what’s “honest” for ME?
  • FV: everything is very template-based. we’re starting to let people choose their colors.
  • JM: let’s talk computer programs? will there be advancements?
  • AK: they have it, they’re just not giving it to us?
  • JM: do you feel limitations? Besides using Adobe Dimensions?
  • SD: we don’t feel so limited
  • FV: we’re using java and flash, not illustrator. most of ME is done in java. as soon as we launched, people are like “why java?” “its fast and beautiful, but i hate you. why java?” So we’re moving to flash, but math-intensive stuff like social networks need java.
  • JM: what do people need do be successful like you? who do you hire?
  • SD: we have cartographers, statisticians, illustrators, but they all have a journalistic outlook: they want to have to learn facts and tell a story
  • VF: it would be awesome to find people who’re very good at some craft, but are also willing to cross boundaries and do some inter interdisciplinary work. it makes you think of things in a different way.

Audience Q&A

  • Q: does Many Eyes take dynamic data? or let you share data?
  • FV: not yet
  • Q: SD, to what extent is yr department given investigative leeway and to what degree are you told what to do?
  • SD: we get 80% of what we’re dealing with. people in the department are journalists and should be trying to find stories.
  • Q: Are people on Many Eyes using the wrong schematic?
  • FV: we see that a little. but when people choose the wrong technique, other people will come along and try another technique. But a few times people use a technique in the “wrong way” to great effect. Eg, someone used the time series w categorical data at the bottom and we were like “huh, we didn’t think of that”
  • Q: SD, how do you know if yr designs are working? Do you do testing? And if so, do you see a progression; is the public learning?
  • SD: “we’re totally blind, we have no idea if people understand what we’re doing” We get some feedback from the peer review workflow in the office. In the past few years we’ve been doing more web work, and so we’re doing a little more user testing.
  • Q: How does Information Design lead to other disciplines?
  • AK: Don’t worry about it. See things as pure objects.
  • SD: if yr a bad journalist, you wont get data, so to me that’s the source
  • FV: I can’t think about these as separate disciplines. We don’t yet know what “good” is, we’re just scratching the surface. What does it mean to think about these moving things as something that you can really inhabit. We’re still in the beginning of that.
  • JM: yr either talking about line — i did this first — or yr sharing
  • Q: Do you take notes at rock shows?
  • AK: i have to. its like steve said: the data is real
  • Q: SD, what do you think about the donut versus the pie?
  • SD: if you look at european newspapers they use the donut more, but we’re patriotic, so we use the pie.
  • Q: FV, have you thought about privatizing ME?
  • FV: that’s the #1 request we get from companies that want to visualize private data. we’re just a small research group and we don’t have the bandwidth for a project
  • Q: Are there copyright issues w redesigning data?
  • FV: as long as you don’t use it for commercial ends, you can use it any way. The only thing we ask is to credit.
  • Q: do reporters in other departments turn to you for help understanding data?
  • SD: a little, there’re resources out there and we develop resources. its pretty common to get sent spreadsheets and ask for a visualization. usually they come when looking for a lede, “here’s the data, where’s the lede?”
  • Q: SD, what’s the average profile in yr dept? AK, have you collaborated?
  • SD: backgrounds are all over the place
  • Q: 3 elements: accuracy, precision, and keeping interest? which is the most important? [QUESTIONER wanks on about the difference between accuracy & precision]
  • AK: i’m concerned with not-accuracy. I want to get back to the last question: I don’t collaborate. its important for me to be the producer. When people ask me to do a chart of their ex-boyfriends, it doesn’t work.
  • FV: back to accuracy & precision: we want to spark conversation, that’s the first thing we’re here for. But accuracy is important when you have a public site that anyone can use . Its sort of like Wikipedia. So having mechanisms to let other people check is really important.
  • SD: Accuracy is the most important. But i’m contractually required to say that for the NYT.
  • JM: to close: artists & creatives are good at dealing w ambiguity. People with this skill are important more than ever as the world gets more and more complex. Go out and show.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s