Lecture 04: Why Visualization?

From Helen Purchase:

I don’t know a great deal about design studies, but when I teach the concept of design to my HCI students, I focus on the concept of design space: any artefact can be described by a single point in multi-dimensional space, where each dimension is a choice that has been made. A choice along one dimension may constrain the choices along another (e.g. choosing a small size for a button constrains the possible font size for its label). This is very much like the rational model described in the Wikipedia ‘design’ page you refer to – and while real designers may not follow this rational design process, I find that it is useful to tell the students about the concept of design space, so that they realise the HUGE scope there is for different designs for the same interface functionality. And design rationale is justifying why the artefact design point is exactly where it is (and not anywhere else!)
So: my approach to ‘design studies’ is that of careful decision making, while appreciating the full range of decisions that can (and must) be made – and the more carefully these choices are made, the ‘better’ the overall design.
And in IV, we still have to make careful choices: these may be colour, shape, round-vs-square edges, means of moving data around… some of these choices may be made because they are necessary for data interpretation (a trivial eg. you would not have all the slices in a pie chart the same colour!), but other choices may be made for ‘aesthetic’ purposes: because, in the eye of the designer, they are better/nicer/more elegant than the alternatives offered in the design space.
And I suppose what many IV papers miss out on is any justification as to why the visual design has been chosen the way it has been: my guess is that this may be because the designer has not thought at all about the visual design, and has simply made arbitrary choices (and I suppose we can ask ourselves whether this matters or not…)

• A Design: is a plan / set of choices (intention) in order meet goals
• Design (verb) is the act of making those choices

it is difficult to define design without trying to define good design

in hind-sight, it’s easy to see if we made good choices or not

designs may be good for some purposes, and not others

Good Design vs. Bad Design vs. Not-Designed (orthogonal? – triangle or square)

luck (or brilliance) is the 4th corner of the square

Good Design (verb) = process that makes it likely that you are in the upper right corner of the square

Sampling

Use 1D events (analog to snow)

overdraw, binning (histograms), kernel density estimates, pareto chart, rotation (look at spaces), other designs

in hindsight – you can know what’s right

Piazza Comments

The Ware chapter was extremely clear and engaging (he clearly considered his readers’ cognitive processes in the construction of the chapter!).

see design ideas apply everywhere!

Narrative (known vs. unknown vs. help in constructing)

Multiple visualizations as the way around too much data – detail on demand paradigm

it’s difficult for me to conceptualize a better way to layout a wikipedia article, for example, but perhaps it’s a worthwhile discussion.  – start with easier ones, or hard ones

Presentation: the whole picture of communication – and the context of a visualization (Tufte and Ware both emphasize presentations)

These readings had me wondering about the role of good visualization in the everyday workplace. What I mean by that is that why it would be great if every major had to take a visualization course, it’s not realistic. Does that mean that info vis specialist should be in every company or is the role of the info vis community to train the general population to make better visualizations through the building of tools or the constant bombardment of good visualization examples?

Disagree with Tufte. “Annoying” – he gets worse

But what does ‘intuitive’ mean

Standardization

Why bother defining all this (some bad papers)

Eye movements based on questions