“My Dating Site Thinks I’m a Loser”: Effects of Personal Photos and Presentation Intervals on Perceptions of Recommender Systems

The article focuses on how people changing their answers if they perceive that the system is being affected with hopes of manipulating it as they see fit. For example, Netflix suggests movies based on what you have previously watched. If the movies it suggests are movies you don't like or wouldn't want people to know about, Are you are likely to change your queue to change what the system thinks about you? If the system has a picture of you, are you as worried about what it will say? For the paper, the authors set up a user study with a dating system to see if people changed their answers midway through and also how the picture affected it. The system displayed the same match set for everybody, regardless of how they answered, so it gave the testers a clear picture of how people try to manipulate systems. They found that people will start to change their answers if given intervals to see their progress/matches. They also found when people post a picture, they are less worried about what the text says.

I can agree with the basis of this article. Many times you will either consciously or subconsciously alter your decisions to get the system to say what you want it to. I've noticed this on netflix, for example when I rented Night at the Museum, it suggested all kinds of family movies including Disney kid's things. I'm not sure what prompted it, but I think the next one I got was Rambo.


My Dating Site Thinks I'm a Loser: effects of personal photos and presentation intervals on perceptions of recommender systems

Ethnography

My original idea for the ethnography was to compare line efficiency between multiple lines and a single line with then was helped by individual processors. For example, a grocery store where each checker has a line, or McDonald's, where there is just one line, which then whoever is at the front goes to the first available checker. This I decided would be hard to test and compare.

After talking with a few friends, we decided to get together on a group project. We eventually came to the conclusion to do a comparison between computer labs on observing how many people are actually doing work vs just surfing or wasting time. We plan to compare several computer labs at different times, and also potentially the gender of the user.

The Design of Everyday Things

The book, Design of Everyday Things, tells how a designer thinks about common place things to make them easier to use or more efficient. For example, how do you know if a door is a push or pull, or even where to push or pull? The author covers simple things, such as placement of the handle or bar to indicate which. The author later goes on to discuss how many times a bad design is really to blame in place of user error (which is officially the cause). The author argues that a good designer needs to get design the interface so that it will be easy and simple for the user.

This book is interesting in that it explores in great detail something that almost everybody takes for granted. I think this is useful for CHI because it makes you think how to make your user interfaces simpler. For example, put pictures by your buttons to make it more intuitive or change the layout so that there is less stuff per page to not intimidate the user as much. However, I think this book was much more verbose then necessary in describing the various examples and points.

Mouse 2.0: Multi-touch Meets the Mouse

Mouse 2.0: Multi-touch Meets the Mouse
Nicolas Villar, Shahram Izadi, Dan Rosenfeld, Hrvoje Benko, John Helmes, Jonathan Westhues, Steve Hodges, Eyal Ofek, Alex Butler, Xiang Cao1, Billy Chen

This paper discussed various new ideas to implement into a mouse to add additional adjustable controls instead of the class 2 dimensions with buttons. The designers tried several different methods, such as movable arms, capacitance, and light as methods of detecting unconventional input. They built sensors to experiment with other movements, such as hand position on the mouse and rotating the mouse.

I find that more than 2 dimensional controls on mice (not counting buttons and scrolling) is probably unnecessary for most day to day activities. However with more advanced things like 3D design or gaming additional controls could prove very useful. For example, I recall playing flight simulator ages ago and was thrilled when I realized that the joy stick also had a third axis which could control power. There is no reason that a mouse should be restricted to only 2D sensing.

A Practical Pressure Sensitive Computer Keyboard

A Practical Pressure Sensitive Computer Keyboard
Paul H. Dietz, Benjamin Eidelson, Jonathan Westhues and Steven Bathiche

The designers built a keyboard with can detect how hard you push the keys. They used several different levels circuits laid out in a grid matrix. By pushing the key down, you completed different circuits, which is how they measured how hard the keys were pressed. This is notable because they keyboard has really not changed much since it was first widely used in the typewriter. In the paper, they describe how similar technology has been used in electronic music keyboards and other methods, however it has not been applied to the standard qwerty keyboard. From the user studies, they reported that many people would be interested in a keyboard which can expand user input.

I think this could be very useful for input other than conventional text from keyboards. The designers mentioned that in gaming that it could control how fast you move. That would be very useful as with conventional keyboards only send a discrete true/false type response. However, I don't see that this would be very useful in just typing/drafting papers. I don't mind holding the shift key to do a capital letter, and in some ways that would be easier in my opinion than pushing the key harder.