06 April 2007

What is an infographic?

This year's Malofiej competition is notable because for the first time ever, an interactive graphic was the only winner for the "best in show" award.

This sparked a lot of conversation about "what is an infographic?" As infographics go online, the line between infography and software begins to blur. Where does infography end and interface design begin?

About a week ago I had an interesting conversation with Juan Velasco, Graphics editor of National Geographic, where we tried to answer some of these questions. It's the beginning of a manifesto -- can you help us make it better?


1. It's a visual explanation that helps you more easily understand, find or do something.
2. It's visual, and when necessary, integrates words and pictures in a fluid, dynamic way.
3. It stands alone and is completely self-explanatory.
4. It reveals information that was formerly hidden or submerged.
5. It makes possible faster, more consistent understanding.
6. It's universally understandable.

Please add your comments and thoughts.

Keep in touch! Sign up to get updates and occasional emails from me.


MAA said...

Great post. Possible 7. It is engaging.

Anonymous said...

I suggest another points:

a) It reachs the left side of the reader's brain in a fast and effortless way. So the reader could take a desision faster based on the emotions and intuitions awoken by the infographic.

b) It's focused on the points that really matter

c) It's target oriented.

d) It's enjoyable for the reader (at least more than a 30 pages brochure without images and writen with Arial 9 point at single space)

e) An infographic could be a "functional art" masterpiece (why not? You (Xplane) and Chris Ware had made several works that demonstrate this point)

I'm a big fan of XPlane. I've been watching you since 2001 (I get a poster that you made to HP Mexico about e-government) and I'm glad to see how much you've grown.

Create Infographics is a challenge that requiere the use of both sides of your brain. The work you have done is great.

Congratulations for your success and your new office in Spain.

Unknown said...

Thank you both!

Anonymous said...

I would perhaps temper the suggestion about universality. I work on infographics in a bilingual country and find the language barrier a constant challenge for mixed graphic and language schemes (and some iconographic only schemes, even). I've found greater problems internationally, as infographic elements that seem inoccuous can turn out to be a disaster (the "a-ok"/thumb-and-forefinger sign that is common in North American stock photography suggests something quite different in Latin America).

I've thought about this with respect to Edward Tufte's work. He touts the universality and created that wonderful wordless graphic for Bose's international instruction manuals. He also emphasizes universality in his latest book. In his course, he qualifies by saying that the main principles of design he espouses are universal, but the specific expression may or may not be. Indeed, many of the graphics in his books are not at all self-explanatory universally (either in our age or in our culture; otherwise, he wouldn't have to explain them; and he's big on the principle of annotate everything).

If universality is a criteria, I think many culture- and client-specific designs would be excluded.

I also think there is a need to not just define information graphics, but define what it is not. The information graphic that won the award is a bit of a fad at the moment: raw data that moves and shows comparison, maybe with a bit of annotation. It puts comparative statistics in the hands of interested people: yay for that. I partly work as a statistician and good stats infographics require an extraordinary amount of forethought and checks with knowledgable colleagues. Too often the wrong (or opportunistic) inferences are drawn. And Tufte too warns about inferential errors with statistics (such as the cherry-picking problem).

All this to say that I think that information graphics are more refined/digested/considered than the type that was nominated. Interaction is becoming a useful tool in inforgraphics as a way to overcome the limitations of the resolution of the monitor (as an example). But not all interactive graphics that don't constitute an interface can call themselves an infographic. Those that are not should be defined.

So, on this note, I would emphasize the notion of teaching. Doesn't your company's Web site mention education as part of your infographic value proposition?

Unknown said...

Good points about universality -- but isn't the whole point of visual communication that it can be understood around the world?

The O-K (and other things) sign is a good case in point. But would you make the case that that gesture is an infographic?

This bears further discussion, because I think that some gesture/word combinations would qualify and others would not.

For example, a video of a chef giving cooking instructions which included gestures, pointing, examples etc. would in my mind qualify, whereas a simple "OK" gesture would not.

It's a good point that the purpose or intent is important. The primary intent of an infographic should be to inform, explain or educate.

With regard to "what an infographic is not" -- you seem to be saying that the NYT graphic "is not" -- but I think "more refined/digested/considered" is not an exact enough definition to work as a true criteria.

I and others are struggling to define the difference between an infographic and an interface -- more discussion on this topic will be much appreciated.


Anonymous said...

Anonymous, your point is interesting.

Of course there are different meanings for a gesture made with a hand (thumbs up or middle finger raised) depending of the place you live. But in this case we are speaking about the use of a specific code (regional code).

Anger, joy, pain, sadness, love and hundreds of emotions and behaviors can be represented visually. Readers of around the world can understand those representations because humanity share a basic and common languaje (primal code composed by physical expressions, face gestures, color appreciation...)

Every visual message (created to communicate or express an idea) is composed by several layers of meaning that must be decoded by the reader using their own cultural background and experience. Each layer added to the message increase their complexity and demand an extra effort to the reader (That's the reason why some poeople think that Picasso and Pollock painted like some 4 years old childrens, but for another people they are fine art masters (myself included)).

Every message has a target. So the universality of the meaning of the message is necessarily linked the reader.

An infographic must be created with the explicit purpose of communicating in a clear and concise way. In order to achieve this goal we could use the very first and primitive layers of meaning. But we don't have to underestimate the Reader's power: human brain is capable to understand, reject, accept, learn and adopt new codes if its necessary. That's how language works.

So I guess that universality is reachable if you create a good message that really interacts with the reader.

Saludos desde México.

PS. Excuse my awful english. I hope I've expressed correctly my points.

Anonymous said...
This comment has been removed by the author.
Anonymous said...

To follow-up on my previous post about universality given your comments ...

I suppose my point is: (a.) some infographics aim to be understood around the world and achieve that goal to differing degrees; and (b.) some infographics aim to be understood by all the members of a target audience. I would not want to exclude from the definition any infographics aimed at a geographically based audience and that use locally understood symbols.

My point about gestures is that: if you are to create an infographic depicting people in various situations, as many of the XPLANE graphics do, the gestures and encounters may or may not be understood around the world. We know from research that facial expressions and emotions are more-or-less universally understood--fine. The situations/encounters are very context dependent an may be widely understood (such as a business context given the spread of international business) or may not be (cooking? I'm not so sure, beyond "throw that ingredient into the pot" and the like; diets, cooking implements, ingredients, and practices vary quite a bit around the world). And the gestures can be very problematic, despite the spread of such things as comic books and movies around the world.

Let me provide an example. Say I create an infographic and one of the nodes (as you put it in your workshops) is the idea of the international business greeting, exchange of contact information, and establishment of a mutually beneficial relationship. One figure is handing another a business card with one hand. The other figure is registering acceptance with the a-ok sign. In North America, I'm sure this would be interpreted as a friendly exchange and the intended point might get across. In, say, Singapore, it would not: one-handed hand-over is a sign of disrespect; a-ok sign may be interpreted as a sign of reciprocal disrespect. Now, these may have been a poor choice of gestures that illustrate my lack of cross-cultural awareness. But I'm not so sure there are many such encounters (beyond the purely pedestrian) that would work so well across the world. I'm having a devil of a time with this in my own work--and so the easy solution is generic postures and annotations, which then have to be translated.

Putting iconography into the mix can help or hurt. Airport signs, computer icons, weather icons, and basic typographical symbols (checkmarks and such) are fairly universal. Other icons (including symbols on traffic signs) less so.

So, I am agreeing with you that some visual elements work universally (or close to it) and others do not. My point is simply that I wouldn't want to exclude more localized infographics from a definition. Perhaps the idea of connecting with an intended audience (which I take to be gabriel's point about targetting).

I would take a stand, for the sake of argument, and say that the NYT graphic is not an infographic because it does not have a message to convey or have meaningful and obvious juxtapositions to offer.

I think the useful comparison here is Hans Rosling's Gapminder World graphic (gapminder.org). This graphic uses motion to convey one or more messages. It is interactive insofar as it allows people to see the messages from a few different angles and at a chosen pace. But meaningful inferences are made clear by the juxtapositions in the data (assuming the person is reasonably capable of interpreting the chart); it is *leading*, to some extent. You're then free to draw your own conclusions based on these (empirically based) inferences.

In the NYT graphic, I'm not so sure that there is much in the way of meaningful message or pattern here. It seems like a data dump with an interface more akin to what is fashionably called an "executive dashboard" (or information dashboard, as Stephen Few puts it in his O'Reilly book). Here the inferences are almost entirely the audience's job; that is, the analytical work is up to the audience. And, presumably given the wide range of likely inferences, the conclusions drawn would also vary quite a bit. As few puts it, good information dashboard designs *allow for* useful comparisons between indicators, but are quite agnostic about the actual data.

So instead of "refined/digested/considered", I'd restate as: some analysis has been done to create coherent and meaningful messages and comparisons. There is a notion of *analysis* and *intent* in an infographic that would not be present in an interface, which is simply about an *arrangement* of information that allows for efficient interpretation (cognition) and manipulation.

Your list of "inform, explain, educate" seem to be a good list of intents. Perhaps also "persuade" when it comes to subjects that are contested and someone is trying to make an argument.

Cheers, Peter.

Unknown said...

I have been thinking a lot about this "universally understandable" issue. I agree in principle, but it seems that if we give in on this point we miss the really big opportunity.

Mathematics, for example, can make the claim to be a globally consistent, universally understandable language. Visual language has the same opportunity.

Now, mathematics is not always understood by everyone, because to some degree it must be learned.

Some aspects of visual language can have multiple meanings (some of the gestures mentioned, for example).

But on reflection, it seems to me that this is true in any language. Intonation, tone and context can change the meaning of a word.

Additionally, one word can have multiple meanings -- for example, "plane" can refer to an airplane, a flat geometric plane, or a metaphorical "higher plane." When spoken there are even more potential meanings, e.g., "plain."

This does not mean that it isn't still universally understandable. I am pretty sure I could tell the difference between someone who was angry at me and someone saying "OK" based on context, even if the gestures were identical.

For us as infographers to do the best work possible, we should be aware of as many of the possible meanings as possible -- however, after much thought, I am now ready to argue that "universally understandable" is critical to the larger mission, and education is the key to resolving potential misunderstandings.

Unknown said...

To go a bit further with this:

I don't think the "universally understandable" criteria excludes the things you are talking about Peter.

Universally understandable is not the same as universally understood. Any language can be misinterpreted.

It might sound like a hedge or a cheat, but I think it's fair to say that multiple meanings and interpretations are still possible, even with a universal language.

JJeffryes said...

You need to look at what language you are using.

If you are using North American middle-class white midwesterner gestures, then that is the visual language you are using. The same gestures will have different meanings somewhere else, even in the same country.

If you want to use a truly universal visual language, then you can only use "words" that are truly universal. A square is a square everywhere. But using square to mean solid, strong or boring is not universal. Simplified graphics may not even be seen as the same thing everywhere. The typical North American lightning symbol is different from the symbol used for lightning in most other cultures, for instance. Other typical symbols, like pointing fingers, arrows, red for danger, etc. may have entirely different meanings based on culture and are not universal.

Mathmatics is universal because it is based on the underlying principals of reality, not culture. Two is two everywhere. Likewise geometric, math-based symbols are universal. A square has four sides everywhere.

But visual symbols with meaning beyond geometry are inextricably tied to culture, because that is where they come from. Their meaning derives from culture, not physics or math, thus that culture cannot be stripped from them without removing their meaning.

Unknown said...


I think visual language is actually more universally understandable than mathematics. People who can't read or write, for example, still understand gestures.

Even a dog can be taught to point, for example, but teaching a dog to do even simple math is pretty difficult.

For example, you say "The typical North American lightning symbol is different from the symbol used for lightning in most other cultures, for instance."

Maybe so, but it's still universally understandable. More so than these words I am writing, and certainly more easily understood -- by anyone -- than most math.

When you say "using square to mean solid, strong or boring is not universal" you're not talking about visual language, you're talking about spoken language. When people use square in that way they don't draw a square, they say the word.

I've never suggested that spoken idioms like the one you mention are universal. At XPLANE we have a monthly "all hands meeting" -- which in the US means everyone in the company attends. Here in Spain it just doesn't translate, it sort of means something like "everybody send your hands to the meeting." But a picture of a meeting could probably convey the concept more easily.

"Other typical symbols, like pointing fingers, arrows, red for danger, etc. may have entirely different meanings based on culture and are not universal."

I don't know if I agree with that Josh. I am pretty sure you could go anywhere in the world and, if you pointed at something, people would understand that you meant "that thing."

"Mathmatics is universal because it is based on the underlying principles of reality"

"Two" is not a real thing, it's an abstract concept. In actual fact it refers to nothing. You can't point to a "two" in the real world, you can only point to a symbol that represents it, or to two "of something."

But a rock is reality that you can point at. And a picture of a rock, for example, refers directly to a reality; it represents or "points at" a real thing.

"Visual symbols with meaning beyond geometry are inextricably tied to culture, because that is where they come from. Their meaning derives from culture, not physics or math, thus that culture cannot be stripped from them without removing their meaning."

Yes, I agree. Of course visual symbols are tied to culture -- so are all symbols, including letters and numbers.

The more literal a representation, the more exact, and universally understandable, it can be.

But we probably need more discussion on this; I never suspected that the universality of visual language would be such a hotly debated topic!

Let's keep this conversation going.

Anonymous said...

Continuing on the "universally understandable" issue ...

Let me shore up my earlier position by saying that I agree that information design ought to aspire to universalism. Is a normative goal, I think it is a good one. I also think that information design has greater opportunity to communicate across cultural and linguistic lines than the written language. This is because language is more culturally specific than pictures. And I can think of several examples that would count as universal. We're in agreement here, I think.

I would also suggest, as I mentioned with respect to Tufte, that there are universal principles that should guide the design of information design. Tufte has six big principles in his Beautiful Evidence book (chapter 5).

I would also say that certain forms and styles are universally recognizable. For example, certain graph styles and ways of illustrating or painting people are pretty much universally understood.

And some human qualities are universally understandable too. Facial expressions are an example that I've mentioned.

I'm more sceptical of the idea of a universal pictoral language, a la Otto Neurath. That is because the more abstract and iconographic you get, the more the system has to be learned like a languaged or system of mathematics. It could be globally consistent but not widely recognizable.

So, that said, I would add the following.

I think that there are a few dimensions that can hurt or enhance the universal applicability of an infographic.

First, there is LITERAL-VS.-FIGURATIVE imagery. So we're in agreement on this point. The more literal--the more realistic--the easier it is to grasp across cultures.

Second, realism is not enough if the objects used are NOT RECOGNIZABLE (OR DON'T EXIST) across cultures. I can use photo-realistic pictures in an infographic, but if the item is completely foreign, the communication is undermined. For example, I was in Singapore and saw two infographics about the durian fruit: one was about how to prepare it and one was about how you're not allowed to bring it on the subway. Being a Canadian, my first thought was: "What am I supposed to do with the pine cone? That's a big pine cone? Wait, what is that?" So when infographic designers use local references in infographics, universalism is undermined.

For example, many infographic designers use a picture (or illustration) of their national legislature as a symbol for government in their infographics. Most of these don't look anything like Capital Hill or Westminster, thus leaving North American audiences scratching their heads. As a test, try to find an infographic about U.S. government processes that doesn't use Washington imagery as a symbol for government. (Such usage is lazy and clichéd, but certainly popular.)

Third, there is the different way cultures TREAT CONTEXT. English is a very literal language (even though, as you say, many English words have multiple meanings). Other languages, such as Japanese, are more contextual. A "yes" can mean "no" in Japan depending on the context. So when we create information graphics, the context we build into the graphic (or leave out) can have a major impact on the extent to which it is understood universally.

Fourth, there is a tension between MANIFEST-vs.-CONCEPTUAL. Highly conceptual ideas are more difficult to express visually. I suspect that XPLANE deals with this challenge all of the time because it works so much with business concepts (e.g., organizational culture). I think that many of the social sciences are in the same boat. This is where infographics require a great deal of ingenuity.

The beautiful infographics in National Geographic and Scientific American mostly represent the workings of the physical world. I've never seen a volcano in person, but those infographics are very good at helping me understand. The social world, on the other hand, is trickier. Highly abstract concepts, trickier still. They can be broken down into components (nodes and links) and annotated, but nonetheless, the meaning may not be clear to everyone--and certainly relies on the skill of the information designer. And here universalism suffers.

Fifth, there is the tension between NOVEL vs. CONVENTIONAL. Many ways of representing something visually have spread around the world. Maps are the best example. It is probably no coincidence that they are looked upon as one of the most universal forms of infographic. But some visual forms are not so widely recognized. Many relatively recent graphing techniques have yet to catch on across the world.

I think these tensions cut to the issue of "universally understandable", as in understandable across the world (not just misunderstood in the sense that there will always be people who don't catch the meaning of a particular infographic).

After considering these tensions, I would certainly agree that infographic designers should be aware of multiple meanings. The worry that I have is that it's so difficult to know when it doesn't work across cultures. I think that infographic designers are perhaps a bit overconfident in this area. There are no easy ways to test, as far as I know.

Which suggests that infographic designers need to design in a particular way to achieve universalism. I think managing the tensions mentioned above would be part of it. But I suspect that is not a comprehensive list.

Cheers, Peter.

Unknown said...

Peter, thank you for this wonderful and thoughtful commentary.

It's a rich and deep subject we're discussing here and I have a few thoughts to add: First, for those who don't know them, here are Edward Tufte's six principles:

1. Document the sources and characteristics of the data

2. Insistently enforce appropriate comparisons

3. Demonstrate mechanisms of cause and effect

4. Express those mechanisms quantitatively

5. Recognize the inherently multivariate nature of analytic problems

6. Inspect and evaluate alternative explanations

In summary,

"Information displays should be documentary, comparative, causal and explanatory, quantified, multivariate, exploratory, skeptical."

Secondly, I have a simple observation regarding the universality of visual language:

For better or for worse (many would say worse), the US has made a fortune in the entertainment industry by exporting its culture to the world. As a result, North American culture is widely understood (which is different from being endorsed or even liked) throughout the world.

Therefore, in many conversations I have had on the subject, it has become apparent that most references to North American culture are understood more or less universally.

Certainly this begs the ethical question: should we, as designers, employ North American conventions? There is no doubt that it smacks of hegemony and imperialism. However it can be helpful to remember that, when in doubt, many North American visual conventions are understood far more globally than many North Americans realize.

I've had conversations about this where people explicitly said to me

"I'd rather have you communicate with me visually -- even if it's obvious you don't understand my culture -- because the alternative (words alone) is so much poorer."

Also, translations of words are much less exact than many people realize.

I've seen many people avoid visual communications due to a fear of unintentionally offending their audience. At the same time those people felt comfortable sending messages in text form. These messages are often badly translated, and in fact many come out the other end of the "translation pipe" as complete nonsense.

As an additional thought, I am attempting to compile all these thoughts into a single article/document. If you would like to collaborate with me on a Google document, let me know by email at dgray (at) xplane (dot) com and I'll send you an invite.

Anonymous said...

Thanks for the nod.

I think it is a good idea to consolidate the discussion into a document to try to get some clarity. Count me in.

After further thought, I have another tension to list.

The sixth tension would be the CULTURAL BAGGAGE that certain imagery might have in a place. Historical circumstances can give an image a local meaning that is not acknowledged elsewhere. For example, the image of a police officer can mean "law and order", "hero and protector", "party crasher", or "thug and oppressor" depending on local history and prevailing ideology. People in relatively new democracies in Eastern Europe and South America respond differently to such imagery than, say, Scandinavians. Admittedly, much depends on the image itself (e.g., a picture of riot police would signal something more controlling). I suspect that mundane items would be more universal, whereas provocative items are more likely to come with baggage.

Certain things like colour can also be imbued with cultural baggage (red in Chinese culture often has a different connotation than in North America). I wouldn't want to overstate the colour issue, however (I have a new book on corporate-identity design from Taiwan that arranges examples chronologically so that you can see changes in colour fashion across time; the pallets vary considerably).

I would also stress that the tensions I listed can merely influence shades of meaning, while not necessarily undermining understanding completely.

Moreover, the tensions can counter-act each other. For example, Olympic sporting events are often denoted using abstract icons. These are used on infographic maps in the Olympic village. The abstraction would suggest less universalism. Yet, many of these icons are depicted using bubble-headed figures, which are very conventional--thus enhancing universalism. Many have been used for several decades, thus adding recognition value. The recent Olympic winter games chose to move away from the bubble-heads and create icons that is more stylish and abstract--adding novelity. But the context in which the icons are used (and other corresponding imagery and text) enhances universal recognition. Thus, in sum, it's not such a simple equation. Information designers have to strike a balance.

I certainly agree with your comments about the translation of language. Most of what I write is translated into French. Difficulties often result. For example, the term "public policy" doesn't have an exact analogue in French. It usually gets translated as "politics", which has a very different meaning. Two languages may not have corresponding terms.

On the issue of American influence, I would agree in general ... but I think that Globalization leads to a more complicated mesh of influences.

Here is a baker's dozen factors that have implications for both information and graphic design, as well as world-wide spread of cultural meaning:

(1.) International Business Culture (US and Western Europe), which has spread graphing and charting techniques, as well as such things as marketing visuals. It has also brought with it certain norms (such as styles of dress: tie, business suit, etc.) and rituals (handshakes, etc.). There are regional adaptations (bows and two-handed business card exchanges in Asia, as I've mentioned), although some are variations on a theme.

(2.) Graphic Stories, such as comic books (US), bande dessinée such as Tin Tin and Asterix (francophone Europe), and manga (Japan).

(3.) Cinema and Television (mostly US), which spreads understanding of certain lifestyles and such, as you mention.

(4.) Internatonal Fashion, particularly Eurolux (mostly Europe) and fashion related to pop music, that spreads certain ways of dress, graphic design techniques and styles, and the like.

(5.) Consumer Products, such as electronic devices and computer interfaces. Here we get a spread of such things as computer icons, design styles, instruction manuals and such.

(6.) Print Publishing, especially magazines and newspapers that feature infographics. Many magazines from the US (National Geographic) are translated into many languages. Others, such as the Wall Street Journal and the Economist, are available in every business center as is. Others still emulate the style and conventions of leaders in the field (e.g., the new Dubai Express is full of emulation, example click here)

(7.) Internationalization of Professions (various) which spread techniques around the world within certain fields of practice (via conferences and such), with those related to the visual arts particularly well connected.

(8.) Travel and Expats that expose people to different cultures at relatively low cost, aided by cheap jet travel and the mobility of labour. Airport signs and other methods for getting around language barriers come from the travel/transporation industry.

(9.) International Events such as the Olympics, World Cup, and cross-national leagues. These spread sports symbolism in much the same way as the travel industry.

(10.) Internet

(11.) Inventions and Good Practices. Maps would fit here. So would traffic control methods (traffic lights, signs, and such; although variations apply). Indeed, there are a lot of ancient technologies that have spread around the world.

(12.) Architecture, such as generic building styles and internationally renowned landmarks.

(13.) International Art, which at first blush has a strong European influence. But remember that artists are often influenced by styles from other continents. For example, Picasso was influenced by African art he found in Paris. And Hokusai's The Great Wave was a huge influence on late Nineteenth Century European art.

I think that this list shows a strong U.S. influence, but not an exclusive one.

I think that these factors are in flux and influence world understanding and taste much like the Jet Stream influences weather. This suggests that there will be lags (between regions, between hub-cities and the rest, between in-groups and out-groups) that need to be taken into account to ensure universalism. (Lags would be part of the novel-vs.-conventional tension) Also like the Jet Stream, these factors also spread seeds that take root and grow.

Cheers, Peter.

Antown said...

i think infographic is a design that explain an information with a picture.

infografis untuk menjelaskan informasi melalui gambar sehingga pembaca akan lebih mudah untuk memahaminya.


Anonymous said...
This comment has been removed by a blog administrator.
siam said...

Thank you very much! @Dave

Scott Alan Miller said...

I must agree that universality in infographics is not truly possible. Looking towards clear communications, broad acceptance, obvious context are all good and important but being truly universal is not really possible. I think that a more important goal is to avoid unnecessarily narrow understandability.

To the comparison to universally accepted math notation I would cite the following example where even fairly basic math is not broadly universal (even within a single language speaking population) and even the understanding that notational context is necessary to interpret it is not widely known making the situation very difficult.