Hanspeter Pfister, who teaches a perennial information visualization course down the street at Harvard will be doing a two day course through Harvard’s Continuing Education program:

Data Visualization: Conveying Information through Visual Representations

Date: January 12–13, 2011
Time: 9 am to 5 pm
Location: Harvard University
Tuition: $1,900. After December 20: $2,200
Class size is limited.

The amount and complexity of information produced in science, engineering, business, and everyday human activity is increasing at staggering rates. This program introduces you to visual representation methods and techniques that increase the understanding of complex data. Good visual interpretations of data improve comprehension, communication, and decision making.

What you will learn

• How visual representations help in the analysis and understanding of complex data
• How the human visual system processes and perceives images
• How to critique visualizations and identify the design principles used to create them
• Good design practices for visualization
• Various visualization approaches for different data types

Topics covered

• Design principles
• Perception
• Color
• Statistical graphs
• Maps
• Trees and networks
• High-dimensional data
• Visualization tools

Who should enroll

Professionals or academics who need to analyze and present complex information in an easily digestible manner benefit from this program. The program is open to anyone who is interested in the visual analysis of data. You should have a basic knowledge of how to use computers and the Internet.

Other information

You are encouraged to bring a wireless laptop.

Faculty

Hanspeter Pfister is Gordon McKay Professor of the Practice of Computer Science in the School of Engineering and Applied Sciences at Harvard. His research lies at the intersection of visualization, computer graphics, and computer vision. It spans a range of topics, including scientific visualization, point-based graphics, appearance acquisition, GPU computing, and 3D displays. Pfister also offers his semester-long courses online at Harvard Extension School, where he won the Petra T. Shattuck Excellence in Teaching Award in 2009.

Before joining Harvard he worked at Mitsubishi Electric Research Laboratories as an associate director and a senior research scientist. Pfister has a PhD in computer science from the State University of New York at Stony Brook and a master’s in electrical engineering from the Swiss Federal Institute of Technology, Zurich.

Questions?

Contact harvardprofdev@dcemail.harvard.edu.

Should be great!

Number of Processing users, every four weeks, since 2005:

Long version: this is a tally of the number of unique users who run the Processing environment every four weeks, as measured by the number of machines checking for updates.

Of note:

• In spite of the frequently proclaimed “death of Java” or “death of Java on the desktop,” we’re continuing to grow. This isn’t to say that Java on the desktop is undead, but this frustrating contradiction presents a considerable challenge for us… I’ll write more about that soon.
• There’s a considerable (even comical) dip each January, when people decide that the holidays and drinking with their family is more fun than coding (or maybe that’s only my household). Things also tail off during the summer into August. These two trends are amplified due to the number of academic users, however other data I’ve seen (web traffic, etc) suggests that the rest of the world actually operates on something like the academic calendar as well.

About the data:

• This is a very conservative estimate of the number of Processing users out there. Our software is free — we don’t have a lot to gain by inflating the numbers.
• This covers only unique users — we don’t double count the same person in each 4-week period. Otherwise our numbers would be much higher.
• This is not downloads, which are also significantly higher.
• This is every four weeks, not every month. Unless there are 13 months in a year. Wait, how many months are in a year?
• This only covers people who are using the actual Processing Development Environment — no Eclipse users, etc.
• Use of processing.js or spinoff projects are not included.
• This doesn’t include anyone who has disabled checking for updates.
• This doesn’t include anyone not connected to the net.
• The unique ID is stored in the preferences.txt file, so if a single login is used on a machine, that’s counting multiple people. Conversely, if you have multiple machines, you’ll be counted more than once.
• Showing the data by day, week, or year all show the same overall trend.

This is a pretty lame visualization of the numbers, and I’m not even showing other interesting tidbits like what OS, version, and so on are in use. Maybe we can release the data if we can figure out an appropriate way to do so.