Ethnography: Humans and Doors

Team:
Chris Aikens
Brett Hlavinka

Idea:
As Brett presented in class, we are interested in observing how people interact with doors, and how doors alter the interactions between people. Specifically, we want to see how people use the entrance to Zachry at different times of day. We plan on developing a model of when people hold doors for others and what kinds of people are most likely to hold a door.

Example questions:

• How close must another person be for someone to hold the door for them?
• Does gender play a role in the decision to hold the door?
• Are members of the Corps more likely to hold the door for others?
• Do people using cell phones / iPods behave differently?

Augmenting Interactive Tables with Mice & Keyboards

Authors:
*Björn Hartmann, Meredith Ringel Morris, Hrvoje Benko, Andrew D. Wilson
(Microsoft Research, *Stanford University HCI Group)

Summary:
In previous research, multi-touch surfaces are seen as an alternative to keyboards and mice, and thus the two input choices seem to become mutually exclusive. This research team seeks to combine theses inputs to eliminate the limitations found in using either input type by itself. The integration of mice and keyboards offer three main functionalities - high precision/performance input, interacting with distant objects in a minimal way, and serving as proxies for the positions and identifications of users on the surface itself.

The authors give the example of three students working together on a project using a multi-touch table and individual mice and keyboards. The students are able to lock files onto their keyboards, combine inputs to work on files simultaneously, share their items with each others' work areas, and access unique files by logging in via their identified input devices.

In order to link digital files to keyboards, users can either move the files across the table and dock them, or they can move their keyboard onto the files. Both accomplish claiming files as under their ownership via collision detection. To collaborate, simply bring another input device into close proximity to link it (note that you must have the same orientation, as in keyboards approximately face the same direction).

Mice are used to manipulate distant objects that might be awkward to reach otherwise. In this way, the cursor behaves similar to how a user's finger would when selecting items on the surface. But what if the table is cluttered with cursors? In such a case, a line is projected from the mouse to its cursor across the table surface. Proximity again comes into play when linking a keyboard to a mouse. A user can touch the two devices together, or use the mouse to click on the outline of the keyboard if it is resting on the table.

The above figure shows what the researchers have so far tested their combined input design on. In the future, they plan on exploring ways to incorporate multiple users at different locations using the same surface.

Discussion:
The functionality of this idea is awesome. After reading this article, I can't imagine not being able to do this with a multi-touch surface. It seems so natural to be able to interact with touch surfaces via single input devices that I am already familiar with. The possibilities for collaboration and simultaneous multiple users are very well outlined, showing that a multi-touch table can serve some of the same functions as a regular tabletop. Pretty cool!

An example of collaboration: D) group searching H) group writing

Collabio: A Game for Annotating People within Social Networks

Authors:
Michael Bernstein, Desney Tan, Greg Smith, Mary Czerwinski, Eric Horvitz
(MIT CSAIL and Microsoft Research)

Summary:
Collaborative Biography (or Collabio) is a social tagging game that is currently available on Facebook. It is a way to generate accurate information about individuals in a motivated way. Collabio differs from other tagging applications and projects in two main ways. First, Collabio is categorized as a "tagging for you" tool. This means that the taggers themselves do not directly benefit from the tags, and instead tag in hopes that whomever they tag will do so in return. Second, Collabio differs from other Facebook tagging apps in the sense that it is more concerned with the richness of tags over the entertainment value the application itself provides. Though it is a game, it is structured in such a way that inaccurate tags are given little or no point values, thus encouraging accuracy and motivating users to get more points.







Collabio has three main interfaces - Tag!, My Tags, and Leaderboard. Tag! allows users to (not surprisingly) tag their friends. An initial set of tags is generated from information pulled off of the person's profile. As these tags are confirmed by users or new tags are added, the tag cloud grows. Points are awarded for guessing tags, with the most popular tags giving the most points. An example is shown in the figure above, which I straight up cut from the article. My Tags allows users to manage the tags that people have made for them (duh). So if you don't like the fact that everyone tagged you as alcoholic, skank, or bed-wetter, you can easily delete your little annoying facts. Finally, Leaderboard serves as a motivator for users to try and get their names to the top by tagging everyone they can.

So just how useful are the tags to the researchers? As it turns out, they beat simply performing a web search or scanning someone's Facebook profile. Through surveys and rating tests (which are not reproduced here for all our sakes) Collabio was found to produce unique and accurate tags for people that could not be generated elsewhere. Therefore, Collabio is a step forward in information extraction in a social setting via user interaction.

Discussion:
Collabio seems like a pretty fun game to play on Facebook, but otherwise it currently lacks purpose outside the realm of research. I installed it myself and the first person it brought up to rate (my friend Don) was previously only rated by the Collabio Bot. The first word the Bot tagged him with was "awesome", which I find to be true but not very helpful! After piecing it together, the initial four tags generated by the Collabio Bot where "awesome Collabio Facebook tag"... very funny, guys. Combine this with the fact that I got AJAX errors half the time I was trying to guess stuff and I'm not exactly sure how this application ever got people to use it in the first place! Maybe it would work better if other friends used it, but for now it will sit next to FarmTown as another unused Facebook App.

User Guided Audio Selection from Complex Sound

Authors:

Paris Smaragdis (Adobe Systems, Inc.)

Summary:
When someone wishes to manipulate a photo or video, they are presented with a wide variety of tools and applications. Changing colors, deleting objects, merging scenes, and many other tasks which were once impossible are now commonplace. Audio processing, however, is still a complex and complicated task. Users cannot simply point to a section of an audio waveform and isolate an instrument in an overlay. Because of this difficulty, Paris Smaragdis developed a novel interface for selecting sounds. Most audio editors concern themselves with two main points - visualization and sound separation. Audio visualization is essentially a waveform showing the air pressure over time, and is most widely used. Sound separation involves breaking down audio files into acoustic energy, which can be seen as a graph of time and frequency. Both of these points provide information, but they lack object-based interaction.

A time-frequency sound representation

Paris Smaragdis uses audio guidance to achieve sound selection from mixed audio. This task begins with the Probabilistic Latent Component Analysis (or PLCA) model. Simply put, the PLCA model estimates what pieces of an audio mixture belong to what unique instrument or sound, based on what is expected in the mixture, the presence of a given sound at a certain time, and the overall contribution that each sound makes to the mixture. The user can then sing, hum, or play an approximation of the sound they are trying to extract or edit, and use this sample as a prior. The PLCA model then tries to match the prior to parts of the audio mixture.

To test this approach, Smaragdis attempted to extract a speech after mixing it with background music. Using direct playback of the original speech, perfect extraction occurred. Having someone else say the words gave poorer results, but Smaragdis states that they still "...rival modern sound separation algorithms".

Discussion:
Obviously, the usefulness of this software is bounded by how accurately a user can reproduce a sound. Having the premixed track for playback would provide near perfect extraction, but a tone deaf person trying to edit an insane guitar solo would lead to what I can only imagine to be epic failure. Aside from the accuracy of user input, this audio selection tool sounds awesome. Anyone familiar with Audacity or other sound editing and mixing tools knows how frustrating trying to edit a unique instrument can be.

I see this work being furthered by working it in a different direction. If someone was able to extract a track by matching it to my input, could I not take my input and convert it into music? It would be revolutionary to simply sing or hum the parts you wish to include in a song and have the computer match it to pitches and note lengths. Then you could skin each input with the desired synthesized effects, or match it with recorded instrumental inputs. If I knew the first thing about making that a reality I would be hard at work on it now. But until then, I'm going to claim the idea as my own intellectual property!

A Reconfigurable Ferromagnetic Input Device

Authors:
Jonathan Hook*, Stuart Taylor, Alex Butler, Nicolas Villar, Shahram Izadi
(Microsoft Research Cambridge, *School of Computing Science)

Summary:
A reconfigurable ferromagnetic input device can be thought of as the parent class of some familiar input devices (such as a trackball mouse or a multitouch surface). Being ferrous means that the input device contains iron, and therefore the authors of this paper concern themselves with ferrofluid bladders (or liquid iron) and various iron solids. Ferrous objects can be placed on the sensing surface and used to create unique and application-specific input devices by monitoring changes in the magnetic flux above said surface.

Magnetic Field Disturbance

Although a lot of work has been done on inputs, this idea is unique in that it allows for customized input devices, and for the detection of deformations in ferrous objects. The device combines an analogue sensing boards with a digital interface board. The sensing board is made up of sets of 16 sensor coils, while the digital interface is composed of an analogue-to-digital converter with a USB output.

Two different application scenarios were also discussed by the authors. Their first application concerned virtual sculpting, wherein deformations made in a ferrofluid bladder are translated to molding clay (their example image follows). Secondly, the authors looked at using their input device as a synthesizer. Various ferromagnetic objects are used to simulate the necessary actions to obtain sounds from musical instruments (their examples included striking a piano key and playing the violin). This scenario demonstrates the breadth of application for such an input device, which is what the authors wish future users to grasp. They feel that their device is generic enough to have novel application.

Discussion:
I find this input device to be groundbreaking. Researchers tend to focus on a certain type of input and on designing tools and applications that use it, while here we have an input device that can itself be customized to meet demands and fill niches. My issue with this paper was that they didn't go into more detail on their application scenarios! I would have enjoyed seeing just how the applications worked and how user-friendly this input device really is. I see ferromagnetic input devices being used for 3D mapping and editing, and space navigation. Imagine being able to fly a plane with full 3D control by hovering a ferrous object above the sensing board. To me, the applications only seem limited by the exposure of such novel input devices.

Introducing...

I'm following Brett here, and posting my Introduction Poster online.

My name is Chris Aikens, and I can be reached at chris_aikens@tamu.edu. I am a 4th-year CPSC (keeping it old school) undergraduate. I'm taking 436 because it seems like a crazy awesome class over an interesting topic. I enjoy projects that focus on direct interaction with users, and therefore CHI is a surefire way to be exposed to the latest and greatest methods for making these interactions more complex, yet natural. In ten years, I expect to be working on some sort of software team making things that today we can only read about in science fiction novels. I see increased Connectivity and Augmented Reality as the future breakthroughs in technology. I like to compose digital music, read and write, and play video games. I also have an awesome cat.