My name is
Jack Moffett. I am an Interaction Designer with over ten years of experience. According to
Herb Simon, that makes me an expert, so I must have something worth sharing. I have started this venture as an exercise to spur critical thinking about my chosen profession. I hope that others may find it thought provoking as well.
DesignAday will present a brief thought about Design every weekday.
The New York Times published a great visualization this week representing three years of Kickstarter projects. It places each project in one of eleven categories and then plots the amount pledged on the x-axis with time on the y-axis. Particularly interesting projects, such as the Pebble watch, are called out and explained. It also compares the number of live or successful projects in each category to the average financing of them. This shows something that I find really intriguing. The categories of Film/Video and Music have had the most projects—almost eight times as many as the Design or Games categories. Technology is the smallest category. But projects in these three categories—Design, Games, and Technology—brought in over three times as much money, on average. Design, in fact, has the highest average financing, including the Pebble, which is the highest funded project to date, and the TikTok, which was the first project to approach $1 million.
This visualization makes very clear the opportunity available to enterprising designers.
One of the questions in my survey asked what languages respondents were proficient in. I categorized the languages by type and turned the responses into a treemap with the help of Many Eyes.
As you can see, client side languages are the most popular, with HTML and CSS leading the way. That is no surprise. JQuery is the most used JavaScript library. I think it significant that JS libraries are getting more attention than JavaScript itself. I thought that Objective-C might turn up a little larger due to designers wanting to build their own iApps. I was surprised that ActionScript wasn’t more prominent, but it had the same number of respondents as Java.
This semester, I will be teaching my information visualization course for fourth time. It has been two years since I taught the course, and in that time, three noteworthy books have been published on the subject.
Manuel Lima’s Visual Complexity: Mapping Patterns of Information is an absolutely gorgeous collection of network visualizations. I have yet to read the book, but from flipping through it, I could see that the first three chapters and the last two contain the majority of the written content. The juicy middle two chapters are a gallery of beautiful and complex visualizations with very short descriptions. Consider the book to be visualization porn; if you are into data visualizations, this book will definitely turn you on.
Then there is Beautiful Visualization: Looking at Data Through the Eyes of Experts, a collection of essays edited by Julie Steele and Noah Iliinsky. I’ve read the first chapter of this one, so I can’t give it a review yet, but with contributors ranging from artists and designers to scientists and statisticians, I expect it to be well worth reading. There is less eye candy, and they are generally smaller, than in Lima’s book, but most spreads have supporting examples.
Visualize This: The FlowingData Guide to Design, Visualization, and Statistics by Nathan Yau is the book I’ve decided to use this semester as a companion to The Visual Display of Quantitative Information by Edward Tufte. Yau’s book is not one to pick up for pretty pictures. It is a practical guide, giving the reader an overview of where to find data, how to acquire it, how to transform it into a useful format, and how to render it using a variety of technologies and tools available to anyone with a computer and an internet connection. This is exactly the type of instruction that I’ve known was missing from my course, and I’m anxious to work it into my assignments.
I’m looking forward to sharing my students’ work with you. Stay tuned.
Lindsey Estep started out with a simple venn diagram as the basis of her map and built out from there. The initial three circles forming the diagram are:
- Creative Research: primarily focused on conceptual and visual consideration
- Engineering Research: significant testing and study, frequently requires prototypes, pragmatic
- Social Research: centered around research about the audience/user, including emotions
There are color-coded areas that contain the cross-over domains, and then the gray center is made up of the four domains that significantly utilize all three, with labels placed on the side they are weighted towards. Listed around the outside can be found organizations, journals, and luminaries for each area within the diagram.
I must say, of all of the diagrams I’ve encountered that try to map the design landscape, this is one of the most elegant. The categorizations and relationships depicted are accurate, and nothing seems forced. I’m extremely pleased with the way it turned out.
Aaron Geiger is one of my graduate students this semester, but he is actually a Master of Journalism student. He’s been taking the course because he needed an elective outside of his department, and he is very interested in design. This was actually the perfect course for him, as it was primarily reading, writing, and discussion.
It is no surprise, then, that Aaron took a much different approach to mapping the design landscape than did the rest of my students. Here is the process he followed:
- Collected data using snowball effect on established, credible websites that define different areas of design. For instance, after selecting “Industrial Design”, there were options (links, suggested views) to “Peter Behrens”.
- Selected three different definitions of each area of design, and weighed amount of times each word was used.
- Word usage was coded with a number depending on amount of times each word was used.
- Data entered into a spreadsheet, then visualized using Gephi.
- In Gephi, data was plotted and then visualized by algorithm.
The map is organized by the major disciplines, each assigned its own color. Every entity in the map (e.g. person, school, subject) is represented by a circle. The size of the circle corresponds to the number of times each word connected with the original discipline definition. Font size matches circle size. Lines show connections between circles, and the thickness of the line represents the number of threads connected with the snowball sample.
The design of the final artifact needs some work, as it was generated by software, and Aaron doesn’t have the design chops to redraw it. The results are interesting, none-the-less. I’m particularly surprised by the seeming insularity between the disciplines.
One of the capabilities of an application I’ve been working on for two years now is the generation of graphs as PDFs. The UI allows the user to choose a type of graph, select several parameters, specify which axes they go on, and set a number of options. We are using JFreeChart and JasperReports to generate the graphs. I was designing a new type of graph for the next version of the application based on the standard open-high-low-close (OHLC) chart used for financial reporting. I have no need to show an open or close value, but I need to show the high, low, and average values. I was hoping to create something more like a box plot, and while JFreeChart doesn’t do box plots, it will create candlestick charts. So, I worked with a developer to explore the various options provided for styling the candlestick chart. To make a long story short, the best we could come up with was to use the high and low “wicks” as intended, but set both the open and close variables to the average. That gives us a vertical line from high to low intersected with a horizontal line for the average. I can style the length of the horizontal line, but that’s the extent to which I can customize the display in any useful fashion.
Time and time again, I end up working with these tools that make it really easy to implement what would be relatively difficult and time consuming, but they are hobbled by a seemingly mindless inflexibility. How difficult would it have been to allow the thickness of the vertical stroke to be styled separately form the horizontal stroke? Why not allow a horizontal stroke to be placed at any data point on the vertical line? With just a couple extra options, the tool would be able to support many variations on the basic chart types. It’s as if the creators never imagined that someone might want to do something differently than the way Microsoft Excel does.
Andrew Sheldon approached his visualization primarily as a timeline. Starting in the 1400’s with the invention of the printing press, he presents events in each of the major design domains right up to Steve Job’s passing. Events are color-coded and linked to their domains with lines. Domains are sized based on typical salaries, and arranged so as to show interrelations by overlap. Color bars on the inside of the circle reinforce the introduction of each domain. Design methods are listed and explained on the left using color coding to attribute them to the domains that use them.
There could have been a better selection of events with more time for research, and the treatment of the methods was not given the same attention as the rest of the map, but those are minor drawbacks that can be easily fixed with another iteration. It’s a well conceived piece and quite beautiful.
