Paper Reading #6: Shimon: an interactive improvisational robotic marimba player

Comments
Adam Friedli - http://jaiachi.blogspot.com/2011/02/paper-reading-6-adaptive-mouse.html
Evin Schuchardt - http://csce436spring2011.blogspot.com/2011/01/paper-reading-6-shimon-interactive.html

Reference Information
Title: Shimon: an interactive improvisational robotic marimba player
Authors: Guy Hoffman, Gil Weinberg    
Presentation Venue: CHI 2010: 28th ACM Conference on Human Factors in Computing Systems; April 10-15, 2010; Atlanta, GA, USA

Summary
This paper describes Shimon, a robot that can play the marimba while perceiving and interacting with someone else playing an instrument and improvise according to what is heard. To allow for a wide range of motion, the robot has four arms and to better interact with people, the robot has a socially expressive head with a built in camera. The robot can match the human player’s style, tempo and harmony as the music is played.

The complete performance is made up of three different modules. In the first module, called Call-and-Response, the robot responds to a musical phrase using a sequence spotter and beat estimator. In the second module, Opportunistic Overlay Improvisation, the robot tries to play notes included in the perceived chord while also making rhythmic gestures with his head. In the third module, Rhythmic Phrase-Matching Improvisation, Shimon tries to match the style of the human while generating improvisational phrases inspired by the player.

The researchers also focused on the interactions between the robot and player by having Shimon turn away from the player when it is in the third module as if it is focusing. They also made him appear to blink and have a “slow breathing-like behavior.”

Discussion
After reading this paper, the first thing I did was look for it on Youtube. There are a lot of videos of Shimon there. To me this is a great example of Human Computer Interaction. The researchers could have stopped when the robot was able to listen to the human and play music with him, but the researchers went even further. They tried to make him seem more human by having him bob his head to the beat and turn his head either towards the player or the instrument. As a future work they even plan to enhance Shimon’s eye contact. In an interview on Youtube they discuss how they never plan to have a robot band. Their focus is the interaction between robot and human.

Taken from the paper: Shimon playing the marimba

Book Reading #13: Opening Skinner's Box

Chapter 3
Summary
This chapter discusses David Rosenhan’s test in which he and eight friends when to various mental institutions, got admitted by saying they heard a voice that said “thud” and then acted normal once in the ward. He discusses how the staff attributed all of his normal behaviors to his paranoid schizophrenic diagnosis. Slater then tells of how people reacted to Rosenhan’s article about the matter. She also discusses the still prominent questions in regards to psychiatry and how the Diagnostic and Statistical Manual on Mental Disorders evolved as a result of Rosenhan’s study. She also describes how she went and redid the study, and while she was not admitted to the institution, she was prescribed twenty-five antipsychotics and sixty antidepressants.

Discussion
This is one of my favorite chapters so far. I really feel that people are diagnosed with mental disorders and prescribed drugs way too soon. I remember in school having several classmates that had ADD, and in all honestly they acted less hyper on the days they forgot to take their pills or the days when they ran out of pills. I also think Lauren Slater was totally awesome for redoing the study, but I didn’t like reading about how she tried the Risperdal “in the spirit of experimentation.” What? Why are you taking this drug for symptoms you don’t have that is prescribed for someone that isn’t really you? One more thing to note: I thought it was very interesting that Rosenhan described the patients as seeming to know he was normal when the doctors did not.

Book Reading #12: Coming of Age in Samoa

Chapter 4
Summary
This chapter explains the inner workings of the Samoan household and how the household is actually be made up of several households throughout the village. Mead describes the way the different members of the household interact, the role and selection process of the taupo and provides more information on the matai and what they do within the household.

Discussion
One of the more interesting aspects of the household that I read about was how the girls can move from household to household. She can rate the different households open to her and then choose which one she would most like to be a part of. This can be very advantageous to her since some will give her more freedoms or have the best food. I also enjoyed the stories about the selection of the taupo and matai near the end of the chapter. It is one of first looks into the lives of specific Samoans.

Book Reading #11: Design of Everyday Things

Chapter 4
Summary
The fourth chapter focuses on how the user will know what to do with a device. Norman discusses the different kinds of constraints – physical, semantic, logical and cultural – and how they limit the possible operations for a device. He then discusses the problem with switches and the affect their arrangement has on the functionality of a device. Near the end of the chapter Norman discusses visibility when it comes to the display and immediate feedback through sounds.

Discussion
I really like the example he uses near the beginning of the chapter about the Lego motorcycle and how he applies its simple design to the four constraints he discusses. He also revisits the design of the door, which is always a good example to use because everyone has encountered a few poorly designed doors in their lifetime. I also found the last story at the end of the chapter about an institute in the Netherlands that had a lack of sounds and the negative impact that had on the users. I know I would doubt the ventilation system too if I couldn’t hear it working.

Ethnography Proposal – The Trial Run – Mall Kiosks

I plan to work with Alyssa Nabors and Paola Garza, but for the first week we plan to each try our own ideas individually to see what kind of data we can gather. We will then post a final, complete proposal Thursday February 10 outlining what we decide to do as a group.

For this trial week, I’m going to sit on benches at the mall and watch the interactions between the kiosk vendors and the shoppers. I plan to see how the vendor tries to get the customers’ attention and what works and what doesn’t. I’m also interested to see how the shoppers react. I want to see if they flat out ignore the vendor, say ‘no’ nicely, say ‘no’ rudely or just flat out avoid them. Do the crowds of shoppers part like the sea?

To study them I intend to sit on benches near the mall kiosks or stand near them.

Quantitatively, I plan to see just how many shoppers react in certain ways (ignore, feign interest, buy the products, etc) and how many vendors use different techniques (spray the people with perfume, drive a little remote control car around to catch attention, stand around, etc.).

Qualitatively, I’ll be watching to see what works for the shoppers and vendors (how well does saying “no” to the vendor vs. ignoring them work, how well does having the vendor just stand around vs. interact with the customers work). Perhaps even patterns regarding time will play a part. Maybe the vendors will try harder at the beginning of their shift rather than at the end or vice versa.

As far as how I will interact with the shoppers and vendors, I will likely just be an observer. Or if I see a vendor trying especially hard but making no sales, perhaps I’ll go right up to them and buy their product. Maybe.

I’m not sure what mall kiosks currently exist at the mall, but I plan to observe 2-3 kiosks, preferably with vendors that use different techniques to promote their product. To better assure I see the same vendors, I plan to go during the same time each week and one of the days on which I observe will definitely have to be Saturday since the mall is busier then.

Paper Reading: Special Assignment - Dr. Celine Latulipe

Reference Information
Title: Exploring the Design Space in Technology Augmented Dance
Authors: Celine Latulipe, David Wilson, Sybil Huskey, Melissa Word, Arthur Carroll, Erin Carroll, Berto Gonzalez, Vikash Singh, Mike Wirth and Danielle Lottridge
Presentation Venue: CHI 2010: 28th ACM Conference on Human Factors in Computing Systems; April 10-15, 2010; Atlanta, GA, USA

Summary
This paper describes the blending of dance performance with technology through the use of mice the dancers either held or worn, narrative and movement based choreography and targeted and adapted staging. The paper describes the evolution of their project called Dance.Draw and describes several different performances that made use of their technology.

The research began with mice the dancers held in their hands, but in later performances they make use of mice that the dancers wear and a wired sensing system the dancers wear. While they dance, images appear on a projection screen behind them to add a new artistic element to the performance. Through their early studies they were able to develop the use of the mouse, the learn the best placement of the screen and get feedback from the audience to see what was working and what was not.

The paper also describes the mixed sensing system used to know the locations of the dancers and place the images on the screen. The paper concludes by describing the “Bodies/Antibodies” performance which allowed them to fully test the mixed sensing system.

Discussion
I first decided to read this paper after Dr. Hammond mentioned this system in class. Like some of the other papers we’ve read in class, this paper presented an area of research I hadn’t known existed in HCI. Blending the arts with computers is a very unique area to study that can positively impact a lot of people, including choreographers, artists and the audiences that watch the performances. I look forward to hearing what Dr. Latulipe discusses in class on Thursday.

Image from the paper of the "Mischief of Mus musculus" dance performance

Paper Reading #5: Weight-shifting mobiles: two-dimensional gravitational displays in mobile phones

Comments
Evin Schuchardt - http://csce436spring2011.blogspot.com/2011/01/paper-reading-5-weight-shifting-mobiles.html
Stephen Morrow - http://chiblog.sjmorrow.com/2011/02/paper-reading-5-creating-salient.html

Reference Information
Title: Weight-shifting mobiles: two-dimensional gravitational displays in mobile phones
Authors: Fabian Hemmert, Susann Hamann, Matthias Löwe, Josefine Zeipelt, Gesche Joost
Presentation Venue: CHI 2010: 28th ACM Conference on Human Factors in Computing Systems; April 10-15, 2010; Atlanta, GA, USA

Summary
This paper works to close a research gap between the idea of two-dimensional weight-shifting mobiles and their applications, including navigation and augmenting digital content through the shifting of the weight.

Similar to their other study involving shape-changing mobiles (see Paper Reading #4), the researchers do a test with twelve users to see how accurately and quickly they could determine the position of the weight within the device. After a training phase, they were asked to indicate the position of the weight by selecting it on a laptop computer. While doing the test, they could not see the device and could not touch the device while the weight was moved within the device.

An image taken from the paper

After the tests, the researchers found that the users had an average error of 28.9mm for the X axis and 21.0mm for the Y axis. They also found that it took an average of 6.31 seconds to estimate the position of the weight. Based on this error, the researchers believe it would be feasible to use weight-changing as a means of informational display. The researchers also believe their results would have been better if the users had not had to put the device down in between tests. They also think the users were more accurate when identifying the Y axis position because the Y axis was longer on the device.

The researchers also compared this prototype with a shape-shifting mobile they have developed and found that the users tended to favor the shape-shifting mobile over the weight-shifting one.

Though the device is not ready for the market, the researchers point out that this may be an important area of research in human-computer interaction later on.

Discussion
Since this paper was very similar to the last (paper reading #4), it made for a quick read. A small amount of the information was the same in both the papers, but for the most part the information was new and interesting, especially after seeing a different device that utilized a different design to do similar things.

As in their other paper, I wish they could have gone into a little more detail about how they exactly plan to utilize weight-shifting in the device, but this paper focused more on proving that it would be possible to notice the weight-changes, which was still very interesting.

As noted in the paper, one area of study could be to test movement patterns and see how well users can identify them. They could also incorporate input and output into the device to demonstrate how the weight-changing could actually augment mobile devices and how the users react to it.