Of Poets and Visualizations

Le debat sur les langues, Charles Huot 1793

The Great Debate

In college, two classmates fought vigorously their viewpoints:
  • One stated that poems had to have a strict form, follow strict rules
  • The other, that no rules could be applied to poems and could have any form

Points on both sides were relatively convincing. For example, in French, there are names and rules for many forms of structured poems. Specifically, one might be familiar with haiku. In classic English language form, it has 3 lines, 17 syllables arranged as 5, 7 and 5 syllables.

In French, this type of construct has rules from a single word / 1 line, to 2 to 16 lines, each having a name:
  1. (no 1)
  2. distique
  3. tercet (or haiku)
  4. quatrain
  5. cinquain (or quintil)
  6. sizain
  7. septain
  8. huitain
  9. neuvain
  10. dizain
  11. onzain
  12. douzain
  13. (no 13)
  14. quatorzain (or sonnet)
  15. (no 15)
  16. seizain

In all, there are hundreds of forms with even more rules. But then again, you could also have poemes en vers libres (freeform poems). Or poemes typographiques (typographic poems) like l'acrostiche & le calligrame (like this one from Pierre Albert-Birot)

In the end, it is totally possible to have a poem that is purely rules or without apparent structure.


The reason I bring up this thread is that the same thing applies to information and data visualization. Like written language, it is a form of communication, an exchange of information between individuals through a common semiotic.

Another similarity is that language & information visualization can convey facts, ideas, beliefs and emotions. And they can have different audiences.

Even when we agree on the type of visualization (say, a bar chart), we have a considerable latitude based on: 
  • what we want to convey
  • who we want to reach

Compare Albert-Birot's poem above & W.E.B. Du Bois "bar" chart


There are also specific reasons to make visualizations, beyond communicating with others, as we will see in a follow up post: "Visualizations: Explanatory, Exploratory, Diagnostic"

Francois Dion
Chief Data Scientist, Dion Research LLC