Thursday, April 30, 2009

Simulated Augmented Reality Windshield Display as a Cognitive Mapping Aid for Elder Driver Navigation

This paper is from the CHI 2009 Conference, by SeungJun Kim and Anind K. Dey from Carnegie Mellon University.

The aim of this paper was to study how a heads up display GPS system could be used to help elderly drivers. The authors cited sources describing how elderly drivers are less able to map themselves into their physical surroundings. It is important to be able to mentally know where you are spatially to drive safely. Elderly drivers are also less able to multitask effectively.

The goal of this study was to develop a simulation of a HUD GPS system so that elderly drivers especially would not have to divide their attention between the road and a conventional GPS system.

Drivers were asked to complete a driving simulation, one with a traditional GPS system located on the center console, and one with a GPS integrated into a HUD on the windshield. The study was conducted on 24 adults. 

In general, the participants liked the system, especially the elderly participants. The HUD display improved their response times and their safety.

I think this is a very good idea. The paper explicitly stated that this was only a simulation: they imagine the technology to implement it will be developed by others in the near future. I believe this to be reasonable assumption given that small HUDs are making their way into cars at the moment.

Even though I am not old, I could find this system useful. I have driven around unfamiliar cities with conventional GPS devices. Even though they helped me to get where I was going, I knew that I was not able to concentrate as much as I should have on the road. This system would enable you to follow the GPS directions without ever having to look down. 

I look forward to seeing if this will be implemented any time in the future.

Tuesday, March 31, 2009

ThinSight: Versatile Multi-touch Sensing for Thin Form-factor Displays

This is a paper from the UIST 2007 Conference by Steve Hodges et al.

ThinSight is a IR emitting and detecting layer to be placed behind a display such as a LCD. It has exciting possibilities of enhancing multitouch technology and interactions.

The device relies on a two dimensional grid that emits and receives infrared light. Because the light is IR, it does not interfere with what is being displayed on the LCD screen. It's multitouch properties derive from its ability to create an image of what is in front of the screen. If your hand is in front of the screen, it will reflect the IR being emitted, and the device will received it, resulting in an image of your hand on the screen. 

Actual touching not needed to interact with the device, as it can calculate the distance an object is from it, and it can also detect how much pressure a finger is placing on the screen, as the finger will cover more area and increase in size on the image as more pressure is applied.

Another exciting ability this technology provides is being able to interact with other IR devices, such as sending and receiving data to PDAs or cell phones, it can detect the IR codes emitted from the devices, and respond by emitting codes back to the device. It also can communicate with more than one device because it can "see" where the device is and send IR beams only to that place.

I think ThinSight is an amazing technology that if fully developed, could revolutionize the way people interact with computers. It may be weird to know that your computer can "see" you, but I think the possibilities are too great not to explore this technology more fully.

Friday, March 27, 2009

Emotional Design

I believe this book by Don Norman contradicts everything he has laid out in his other works. In his other books, especially The Design of Everyday Things, he argues for straightforward, intuitive design. In this book, he argues that actractive objects stimulate the mind, causing for more creative thinking, making them easier to use.

While I don't necessarily have a problem with either side of the debate, in Everyday Things, he laid out scientific, objective suggestions for making a design successful. If there is an objective method of designing, they should be followed.

I myself would prefer to have attractive appliances, but I do not think that that is a defining aspect of the object. I like objects that are easy to use, or that have great functionality (even if they require more attention to learn fully), if the design is emotionally pleasing, that is a bonus. 

I think designers should worry about how their product will be perceived by their target audience, and should attempt to make it attractive in all aspects. I think that the primary goal is that the device functions properly, and is easy to use with respective to whatever its intended purpose is.

I accept that people like good looking things, but I think the functionality of the device should be the designer's main concern.

Thursday, March 26, 2009


I read the paper "ArtLinks: Fostering Social Awareness and Reflection in Museums" by Dan Cosley et al. from the CHI 2008 Proceedings.

ArtLinks is a test system the authors designed to attempt to link visitors perceptions of an exhibit at a museum. The authors acknowledge that many people go to a museum to learn, but they also state that people may go to a museum to see what others think about certain things.

The ArtLinks system is set up for a particular exhibit, and visitors are encourage to share their thoughts about it, much like a guestbook. The main page of the system displays common words used to describe the exhibit, with words that are used frequently appearing larger than relatively infrequently used words. As the user mouses over a word, they will see lines drawn to the visitors who used those words. As they mouse over another visitor's icon, they will see lines drawn to words that that visitor used to describe the exhibit.

I think the ArtLinks idea is a thought-provoking one; however, when I go to a museum, I want to learn about the objects on display. I am not usually interested in what others have had to say about it.

I do see that there likely is value for an ArtLinks type of system in an art museum, where the exhibits are intended to be interpretted, perhaps differently, by each visitor. In such a museum, the system could be a useful way of sparking interest and different ideas for the visitor. It would give them a better vocabulary of expressing appreciation for the art by seeing what others have thought about it.

Wednesday, March 25, 2009

The Man Who Shocked The World

I found this biography of Stanley Milgram to be an enjoyable read. I had heard about some of his work, not realizing who had done it. I think social psychology greatly benefited from Milgram's work and enthusiasm for a wide range of topics.

Of course, Milgram is most famous for his shock experiments, in which he demonstrated the extent of a normal person's obedience to authority. Many people believe these experiments to be unethical in their treatment of the participants; however, I disagree. I think these particular experiments are very important to understanding human nature. They help to give an explanation of the willingness of people to follow leaders who demand them to be cruel. While the experiments were not perfect (there could have been more variations to account for other variables), they gave an unintuitive and important answer.

Milgram also performed work in other areas: using novel methods of determining public opinion through the lost letters experiment, and working on the small world phenomenon with his experiments of delivering a package from one random person to another.

I think the fact that he has been given little to no recognition from his colleagues is pitiful. Milgram approached his field with great enthusiasm and came up with clever ways of understanding humanity. I think it is unfair for his colleagues to use his work but not give him proper appreciation.

I enjoyed reading this biography, and learning a little more about the man behind some of the most important, most controversial, and most ingenious experiments in the history of his field.

Tuesday, March 3, 2009

The Design Of Future Things

In this latest book we've read by Don Norman, he discusses his thoughts on what things will look like in the future. He talks a lot about cars and home appliances. Much of his focus is centered on automating many processes, essentially taking humans out of the loop. I think it is this that he is wrong. Humans will continually demand to make the decisions and be in control, so while certain things he suggests, such as lane keeping and adaptive cruise control, will make driving safer, the technology available will not be fully implemented: humans will always have the final say.

Although he discusses it only briefly, in passing, I believe augmentation is the future of everyday objects. Augmentation, as opposed to automation, exists to help a human with a task, rather than doing it for them. He gives the example of a kitchen system the will help a cook know where they left off in preparation for dinner, provided pictures of the steps just completed, rather than making dinner for the person.

Although I believe technology will advance to the capacity to enable automation of most items, it would be foolish of companies to assume that customers want fully automated devices. It will be much wiser to develop systems that fluidly augment the activities of their users. Humans want to be in control of what they do. People enjoy certain everday activities that could be fully automated: cooking, driving, choosing their own music, reading their own books rather than listening to a voice reading to them, dressing themselves. 

I believe Don Norman missed the nature of humans when he failed to give more attention to augmentation than automation.

Thursday, February 26, 2009


OcotoPocus: A Dynamic Guide for Learning Gesture-Based Command Sets
Olivier Bau & Wendy E. Mackay

The Octopocus system is a different type of help service for computers. It helps users learn gesture based commands for programs. Although gesture based commands can be very useful, they are also often unintuitive and difficult to remember.

The Octopocus sytem provides a way to teach a set of gesture based commands. The user can select novice or expert mode. In expert mode, there is no help offered. In novice mode, however, the user can click and hold down the mouse (or whatever input method is being used). In a second, the Octopocus system will display all of the different gestures able to be given at that time. The user then selects a command and begins to follow the path layed out by the system. As the user progresses along the path, commands that had different gestures quickly fade, while commands that have a similar gesture will remain, although they will become less visible.

In the tests, the overall results slightly favored Octopocus over the traditional help menu, with errors being slightly lower, however, particpants used Octopocus more often than the help menu. They found it to be a helpful and intuitive way to learn commands.

I do not have much experience with gesture based commands, but to me, the idea behind Octopocus seems logical, intuitive, and cleverly implemented.