Intentional Distortion

Course evaluation forms

No class next week (Spring Break)

Paper due March 25th (discussed today)

Feedback on first paper March 18th



Distortion Function

Next Paper


10-20pp. hardcopy
Relevant illustrations

The topic

Choose an effective organizing principle for "three-dimensional visualizations"
Choose examples that demonstrate the effectiveness of your principle

"Three-dimensional": data or visualization?

Deliberately not limiting the domain of the paper

Almost certainly incorporates interaction

Purpose of "organizing principle" is to help make useful decisions: Possible questions:

Is this a good visualization?
Is it suited to the data?
Does it perform its task?
Is it robust in the face of errors?
How does it compare to alternatives?

An organizing principle

Card, Mackinlay, Schneiderman's model

Driven by perceptual science
Useful for experts, difficult to apply for novices
Ranking of factors by data category simple and useful


Performance metrics

Possible taxonomies

A taxonomy based on data

Category: Quantitative, Ordinal, Nominative

Quant: floating point, rational, integer

Ordinal: ordered enumerated type (like C), hierarchies (partial ordering)

Nom: element of a set, categories

A taxonomy based on rendering

Category: Scatter, bar, line, area
Example: Choosing graph from MSFT Excel

Performance metrics

Tufte: Less ink == better visualization

Oversimplification, but concrete and specific
Measuring clarity: presenting important data and nothing else

User study-based performance rankings

Arguably the "right" way to organize visualizations

Correlation with concrete standard

E.g., given data ranking and type, check Card's ranking
Suggests automated improvement suggestion



Intentional Distortion

Categories of Distortion

In the visualization (e.g., perspective)
In the data (e.g., elision)

Dimensionality of Distortion

1D, 2D, 3D, Temporal

Distortion Functions

Piecewise, perspective, warps


Distortion in the visualization

Consistent data, distort picture
Possible in all perceptual tasks

Distortion in the data

Increased detail

Distorted visuals

Cleveland and McGill: empirical verification


Suggests distortion of position most useful tool, followed by length

Cf. distortion papers in text

Part 4: Focus + Context
E.g., perspective in cone trees, perspective wall, hyperbolic trees

Ordinal Distortion

Position, position, position followed by density, saturation, hue, texture

Log-scale axis as ordinal distortion

Also quantitiative
Consider percentile rankings

Log-scale density, saturation distortion

Hue is cyclical, but still works: log of angle

Ordering is important: Increase perceptual space in ranking

Nominal Distortion

Position; then hue, texture; then connection, containment

Log-scale distortions not applicable

Not without creating artificial order to nominal categories
Interaction can be used here: user ranks importance

Hue, texture: Contrasting colors, orthogonal color axes in different color space

Contrasting colors: Choose one axis, other two mixed form contrast (e.g., blue implies red and green make orange)
Careful of meaning: Red "means" danger


Distort more effective attributes for more emphasis

Position is uniformly the most effective attribute

Quantitative: Length, angle, slope
Qualitative: Density, saturation
Nominal: Hue, texture, connection, shape

Log-scale distorts orders of magnitude

Linear distortion distorts values

E.g., perspective

Distorted data

Sampling rates

E.g., numerical flow simulation


Original Furnas fisheye view

Ellipses in outlines, code

Removing subtrees


Choosing focus of fisheye

Identifying priorities


Browsing can lead to decisions on importance

Get rid of unimportant data (Tufte)

Reveal patterns obscured by false correlations

Browsing: Undirected search

Filtering: Directed search

Browsing: Serendipity

Filtering: Detection



0D: Either/or, multiple views

1D: One-axis linear or logarithmic

2D: Two-axis, perspective

3D: Spatial warps

Temporal: The Matrix :-)