Opening Skinner's Box

This book detailed ten of the more famous psychological experiments conducted in the 1900s. Each chapter is devoted to one such experiment, and each one had a profound effect on future research and our lives today (even if we don't realize it). The opening chapter is about B.F. Skinner and his mouse boxes, wherein he used positive reinforcement to train mice to push levers (that's Skinner in the first picture). The author, Lauren Slater, did her best to get background information on each of the experimenters, sometimes going after family members and old research partners. It is this extra information that makes the book so enjoyable. I actually think that Slater is a bit crazy, but her personal insight brings character and feeling to what I would normally view as dispassionate scientific research. In some cases, such as with Bruce Alexander's Rat Park, Slater even attempted to carry out her own related research. In this case, she did a bunch of drugs to try and get herself addicted... which didn't work out for her (luckily).

Lauren Slater doing something with her hands on stage

The one thing that I found strange about this book was the fact that each chapter moves through a cyclical loop of introspection and explanation. She starts with musings and opinions, moves to facts, and then returns to musings. It's almost as if Slater trails off with her own thoughts about the experiments and forgets that she is relating them to the reader. Each of the ten experiments even related to CHI in some way! It's important to consider the psychological effects of computing instead of just the technical ones.

Anyway, it's a great book! It reads like a fictional story instead of a look at experiments.

The Inmates Part 2

After ripping on designers and programmers for the first half of his book, Alan Cooper offers a shred of hope. Well... basically he promotes his own company... but that's beside the point. Cooper has three basic points to make in the last few chapters: create personas, set goals, and use scenarios. These three parts are closely related, as scenarios are run where personas try and accomplish their goals.

• Personas - detailed potential customers created by the designers. Each persona should accurately reflect a certain demographic of customers. The designers should focus on meeting the needs of set personas instead of those of everyone. It is better for 10% of people to love your product than for 100% of hate it.
• Goals - set things that customers want to accomplish. Goals are not tasks. Tasks are steps that must be undertaken to meet end goals. Creating realistic goals for your personas to carry out can help you define what your product should do.
• Scenarios - situations where goals are needed to be met by personas. Again, creating reasonable scenarios can help you design a product that is both functional and helpful. If in the course of running a scenario your persona cannot accomplish their goals, then you should modify your product.

That's pretty much it. In the last chapters, he goes over these key points and gives examples of how his company has used them. So if you want to be a good programmer, you should either hire them or follow suit. It seems that today we as computer scientists have these design tactics ingrained into us. When designing projects, we normally sit around and make up scenarios to describe our thoughts and ideas. It's hard to image that ten years ago, programmers didn't follow these concepts.

Image Recognition for Intelligent Interfaces

Summary:
Professor Trevor Darrell from UC Berkeley was the invited speaker at the 2008 IUI conference. As such, his paper is just an abstract. Seriously, as you read this sentence you might be reading the same amount as what is in his abstract anyway. He mentions that new advances in image recognition have made image-based interfaces a viable alternative to current interfaces that try to analyze physical objects. He then goes into the various parts of the problem he will discuss.

Discussion:
That's about it. I'm assuming that he then started talking about his abstract. Thank you, Manoj, for the shortest paper ever.