MUSE - Let's see what's on the table

For the project central to this course I'm going to be working with Kat and Ruth to create an interface to help different groups brainstorm effectively. We chose this option because the other options seemed to appeal less to our specific interests. Ironically, I think what we had the most trouble with was brainstorming ideas about how we should approach this problem.

After meeting last night for over an hour the only thing we really knew for certain was that we wanted an interface that represented ideas generated during brainstorming session with physical tokens and that had some sort central area for ideas to collect in order to be reviewed by a group at large.

We got a lot more out of in-class brainstorming session, which I'm attributing both to being more awake and to a lot of guidance from Orit. What seemed to help us make the most progress was using the whiteboard to map out the process we envision being most productive for people as they try to work collaboratively to come up with an idea.

What we came up with is MUSE. We're designing the interface as a way for groups to plan events or procedures starting practically from scratch. The whole idea of the interface is that people can come up with their own ideas in a social, low-pressure way, and easily share and work with their teammates to refine and combine their ideas. That's why it's called Muse - not only is it about finding inspiration, it's also about finding the collaboration in in an individual's idea.

This whole process would take place around a large touch surface. People would have their own place around the table where, for the beginning of the brainstorming session, they are given time to come up with ideas by themselves. For every idea they have they will be able to grab a token from a pile set up near their workplace, and write, record or draw their idea on the surface. This piece of digital information will be attached to the physical token they have, then set aside. The token will be marked in some way to signify the kind of idea that it represents. People can ultimately follow whatever brainstorming style works best for them, though we envision people first getting their ideas out, sorting them into categories, then refining their ideas before they get shared with the group.

What would happen next is that everyone's ideas are made public to the group. Since the ideas are represented with physical objects, all the "ideas" can be physically placed in the middle of the pile. People within the group are then free to look at ideas that seem interesting. People are given a certain amount of time to familiarize themselves with their peers ideas before moving onto the next stage of brainstorming.

This next stage of brainstorming would be a lot more free-form. Ideally, it would occur at the small-group or individual level. People would take the ideas generated before and have a chance to elaborate or improve on them. For example, if your neighbor had an idea you liked, you could get another token, and copy your neighbors idea into it, then make the changes or additions to it that you think might improve the original idea. Alternatively, maybe two other members of the group are working on the other side of the table and you think your ideas goes along really well with some of their ideas. You can physically take your ideas over to where they are, then share all of your ideas, working together.

After this brainstorming period, the group would then come back together to review the ideas that we still on the table. As a group, they would sort their collective ideas into categories, represented by real-life receptacles such as bowls, then see what was actually feasible for the kind of event or topic that was being brainstormed. Different restriction that applied could be represented, as well. For instance, if there was only time at an event for three speakers, you could get a physical restraint that would only contain three ideas.

When the ideas had been decided on, they could then be captured and given to the relevant people. Ideas would have been generated and the process would be complete.

I think we still have soem details to work out, such as the best token to represent ideas and what kind of restriction we're envisioning, but I think we have a solid starting point to work off of now.

Homework 1B - Analyzing TUIs through different frameworks

For this assignment we were asked to look at a TUI (Tangible User Interface) presented in a paper, then analyze it through two frameworks, also presented in papers.

I chose to look at Tangible Message Bubbles, presented by Kimiko Ryokai, Kayes Raffle and Andy Brooks. Tangible Message Bubbles is an interface which consists of a number of tools designed to let children electronically communicate with other people in manner much more intuitive (especially for children) than using a keyboard and mouse. The tools consist of containers (an accordion-shaped one and a balloon-shaped one), which are set to capture audio and video recordings of the child when the child places their face at a whole in the container. After they have recorded something, the child can play the recording back by squeezing the container. The speed at which they squeeze the container controls the playback speed, while shaking the container modulates the child's voice in the recording.

When the child is happy with the result, they can transfer the recording to a touchscreen board (called a message board) simply by having the container near the touchscreen. The recording is represented as a bubble on the surface of the message board. Several bubbles can then be sorted and dragged over an icon, symbolically chosen to represent a person in that child's life. The message would then be sent.


Token and Container Paradigm

The first framework I looked at was Token and Container (TAC) paradigm, presented by Orit Shaer, Nancy Leland, Eduardo H. Calvillo-Gamez and Robert J.K. Jacob. The idea behind this paradigm is that tokens (real-world objects of some kind) represent digital information, or a way to manipulate digital information, and their use is dictated and demonstrated through constraints. The set of information being manipulated is the variable, and it's manipulation is achieved through some sort of action.

In the case of Tangible Message Bubbles, the tokens would be different parts of the containers used to record and manipulate the recordings. Specifically, the opening for recording, the bellows of the accordion, and the walls of the balloon. The objects then hold the recording and allow the child to record and manipulate new recordings. While more manipulation can be done on the message board, this seems like more of a GUI, and less within the realm of TUIs. Insomuch as it's part of the system, however, the message bubble and icons representing people would be tokens.

Interestingly, I think that the containers act as their own constraints, as well. The use of both the accordion and the balloon is dictated by their specific shapes. The openings in both object records to the object itself, and bellows and walls manipulate information stored in the object itself. The massage board, however, also acts as a constraint for the containers, though, since information is transfered to it through proximity. Then, finally, the surface of the message board and placement of the bubbles and icons is another constraint.

The only variable manipulated in this is interface is the recording, but it can be manipulated through several actions:

1. Putting one's face to the opening used to record - records new recording
2. Squeezing the object - plays back the recording in various ways
3. Putting the container in proximity to the message board - transfers recording to message board
4. Moving the message bubbles over an icon - sends the recording

Reality-Based Interaction

This paradigm, presented by Robert J.K. Jacob, Orit Shaer, Audrey Girouard, Leanne M. Hirshfield, Michael S. Horn, Erin Treacy Solovey and Jamie Zigelbaum looks at four different aspect important to "the real world" and emphasizes the importance of including them in TUIs. These four aspects are naive physics, or a persons perception of how objects interact in the real world, body awareness and skills, or an awareness of how one's body moves and the ability to do so, environment awarveness and skills, or awareness of the world around you and the ability to move around in it, and social awareness and skills, or the ability to interact with others in different ways.

Based on this paradigm, Tangible Message Bubbles incorporates naive physics by connecting the amount of pressure and force applied to the containers, to the speed of playback, as well as shaking the container to vibrations in the recording. This is a lot more intuitive then pressing a button with an abstract symbol, or going through a menu to find an affect. It also incorporates body awareness, by using a child's physical ability to squeeze object to allow them to play their recording, as well as record by simply talking into the object. This, again, is more intuitive then finding a play or record button. The environmental awareness comes from the ability to transfer data between objects through proximity, and the intuition of recording by simply talking into the object. Lastly, the social awareness is inherent to the whole idea. The goal is to ease a child ability to digitally communicate with family and friends, so the idea of making a recording is quite social. Kids are just excited about talking to their friends and loved ones.

One of the tradeoffs is that having one object do several functions isn't entirely intuitive for children, and the precision of manipulation is limited. The developers intend on limiting the functions of one object and it should be kept in mind, however, that it's for children, who don't need a highly precise tool to send a message to their grandmother.

First Post

As an Architecture major and CS minor, I enrolled in this class primarily because it seemed like an interesting way to work towards my intended minor. Now I think it might be pretty relevant to my major, too. Our first homework assignment was to read and then present a research paper about tangible user interfaces. I (obviously) chose to present a paper about an interface for urban design called URP.

URP is an "urban planning workbench" designed to help urban planners recognize and solve problems in building placement. It's basically a table which can project information onto it's surface. When you place a model of a building (which you've already given the computer information about) on the table, the table will display shadows, reflections and changes in wind patterns created by the building, just to name a few of URP's abilities. URP was introduced in 1999 (which is much earlier than I had previously thought this kind of research was being done) by John Underkoffler and Hiroshi Ishii and was received quite well by the architects and urban planners who had a chance to use it.

The article got me thinking about how obvious the connection between architecture and TUIs is. Architecture is an art form which is inherently 3-dimensional. While it's technically possible to imagine 3-dimensioanl forms on the 2-dimensianl suface of a computer screen, I would argue that you lose a lot of what makes a building art when you do so. That being said, architecture is also highly mathematical and benefits greatly from using computers. CAD software is notoriously counter-intuitive and difficult to learn (I'm taking my first course on design software this fall, so I'll get to experience this this firsthand!). TUIs allow you to keep the tangible, real-world part of architecture that is at it's core, while gaining the mathematical precision of a computer.

I'm really excited to see what else we're going to cover in this course. I think there are a lot of other ways TUIs could be used to help architects.