Last year this got mixed with overviews, but the lecture notes are useful.
Different readings this year.
Tangles knot of questions:
- Why do it?
- What does it work?
- What is it good for (or not)
- Why vis instead of X?
- How do we do it?
- How to design it?
- How to realize those designs?
- When is it good?
- What are the different flavors of it? (since these may have different answers to the other questions)
Different perspectives on how to come at these questions
What is Visualization?
do we really care about a nitpicky definition?
- not really unless we are engaging in some form of turf war
we are spending a class on it
- me: the creation of imagery meant to communicate
- data (data visualization as a separate piece, what isn’t data?)
- don’t want to mix in the idea of what is a good visualization (it is readable, …)
- kosara: non-visual data, image, readable
lots of words – have connotations within certain communities
- data visualization
- information visualization / infovis
- statistical graphics, charts, … – to me this implies standard charts. more about who than what
- graphs – CS sense vs. statistical chart sense
- infographic
- data graphics
Visualization is general, infographics are specific. Visualization is context-free, infographics are context-sensitive. Visualization is (largely) automatic, infographics are hand-crafted.
Neither are objective, and both require hand-tuning and understanding to get right.
Readings
- Classifications chapter (from Designing Data Vis)
- Kosara: defines vis
- Kosara: names of vis
- Kosara: many word for vis
- Kosara: vis. vs. infographics
- smashing magazine dos and don’ts
- tufte (and maybe the article about him)
- few (the commentary is quite valuable)
a lot of reading (mostly short for 1 day)
Where are these people coming from:
- Tufte: historian (lesser degree, designer)
- Smashing Magazine: designers/artists
- Few: tries to come from all directions (perception, history, practical, design, …)
- Kosara: academic, tries to be far reaching and connect to others
Some key characters (mentioned in Few)
- Playfair
- Bertin – Semiology of Graphics
- Cleveland and McGill – Graphical Perception
- Card, Schneiderman, Mackinlay – early infovis manifestos
Note: interaction is missing from lots of these (especially Tufte and Few)
Few:
- over-simplifies the history
- too hard on pie-charts (he has his particular mission)
- has a clear idea of what is good, which makes the other questions easy for him
- we should always judge a visualization’s merits by the degree to which we can easily, efficiently, accurately, and meaningfully perceive the story that the information has to tell
- (beware: he’s a pundit and is trying to sell his particular style)
- basic perception principles (nice since it mixes gestault with lower levels)
- Rensink (psychologist)
- Robbins (pundit) – why don’t people buy our books (why are there bad vis)
- Kosara (academic vis) – interaction and his recent papers
Designing Data Visualizations:
- starts with a (trite) list of why (leverage capabilities, inspire, …) – doesn’t say why it does this
- classifications (what I really liked from this)
- complexity – an important point. simple things are simple
- infographics vs. visualizations – hand drawn, specific, aesthetic, data poor (but do these go together?)
- explore vs. explain
- informative vs. persuasive vs. art
- designer – reader – data
The Smashing Magazine how to Make an InfoGraphic
Theme: make is visual, different, unique, flashy. It’s OK if its hard to read.
Are they serious?
Tufte at his Best and Worst
defines things based on what he thinks is good – then rhetorically mocks anyone who doesn’t achieve his goals.
Rants about what you shouldn’t do:
- solar radiation vs. stock price is “silly theory means a silly graphic”
- then says that people in MN get cancer since they eat smoked fish
blind assertions (meinard is the best), …
but great, historical examples – and a clear message (which some people follow as gospel)
great sense of history – no sense of how to explain how we should do it
no real narrative or organization – just one example after another
Thoughts to put in people’s heads
It is worth thinking about what “good” means
Do you have a specific message (either explicitly or not)?
Effectiveness of communication
Aesthetics – does this only have to be the domain of “art”
Specific vs. General Solutions
New York Times vs. my personal experiment
It’s not worth arguing about what a visualization is: there are many different perspectives and we can learn something from all of them.