 |
Saul Greenberg | GroupLab | Dept Computer Science | University of Calgary |
This page collects several research and project ideas that may be suitable for undergraduate and graduate topics. They are just brainstormed ideas, so there is no guarantee that all are worth doing! In no particular order...
- Video analysis tool
- A video analysis tool allows researchers to analyse sections of (preferably digital) video tape. There are a variety of papers that describe the attributes of these systems, as well as a few commercial systems. This project (best seen as an undergraduate thesis) is to design and construct such a tool for use by an HCI usability lab. The student will have to read the background literature, derive the key requirements of such at tool, and prototype/test/evolve its design. Some technical challenges are expected in terms of controlling the video via software.
- Task-Centered Walkthrough Tool
- In task centered walkthroughs, a person or team takes the position of a user doing a task from begining to end. At every step of the task, they ask the questions "Is it believable that the person would do this" and "Does the person have the knowledge to do this". In this project, the student would design a tool that supports this walkthrough process. One approach is to combine walkthrough analysis with a storyboard tool, where the person can capture and annotate interface scenes and link the walkthrough description to it. This is a good design project, as it would have a fair balance of design, implementation and evaluation within it.
- A Groupware Toolkit that works on the Web
- My group has developed a groupware toolkit called GroupKit. It predates the web, so it works outside of it. In this project, the researcher would reconsider the architecture of GroupKit and see how it should be redone so it works within standard browsers. The researcher would have to do background reading on GroupKit as well as the current genre of Java-based groupware toolkits, and would have to produce new requirements. The project would likely require strong technical knowledge for doing the implementation (which would likely be Java-based).
- Session Managers
- GroupKit, our groupware toolkit, allows programmers to create session managers---widgets that let people see what groupware sessions are around and join them. In this project, you will design a variety of session managers. The goals are: to determine and analyze different metaphors for session managers; to package some of these within the GroupKit distribution; to build (if necessary) higher level programming constructs that simplify the creation of these session managers.
- Extending GroupKit to support Casual Awareness Facilites
- For people to contact one another, they should be aware of who is around and whether they are available for conversation. There are a variety of schemes for doing this. In this project, the student would survey these schemes, and include (by implementing) some of them as GroupKit widgets.
- Creating a Business Meeting Suite in TeamWave Workplace
- TeamWave Workplace is a groupware system based upon a rooms metaphor. Individual rooms can be stocked with a variety of applets. In this project, the student would design a suite of applets conducive for formal business meetings. The student would have to survey some of the Decision Support Systems literature to see what they recommend, and reconsider how these would work together within a rooms environment. This could be a fun project, as it would require implementation of a variety of groupware applets. Evaluation would be a key component to see how things actually work in practice.
- Connecting PDAs to Teamwave Workplace
- Over the summer of 1998, Michael Boyle wrote several programs that allow Palm Pilots to communicate with GroupKit. This project extends the idea, where the pilot would work together with TeamWave. This is fairly open ended... the point is to look at the design opportunities that such a fusion would represent. Its not clear that there is enough here for a full project.
- Evaluate the PDA/GroupKit Meeting Tool
- We still have to evaluate the tool described above. In this project, the student would design several experiments that analyze how people use the above tool. The result should be: a good experimental methodology; an analysise of the way the current system is used; suggestions for redesigns.
- Designing an Integrated History/Bookmark System for Web Browsing
- Linda Tauscher analyzed the way people access web pages, and described several schemes for making it easier for people to retrieve pages they had just accessed. In this project, the student would review Tauscher's work on web access, as well as Abrams work on bookmark use, and design a new utility that give people an effective way to bookmark and revisit their web pages. Note: Netscape source is now available.
- Searches through Personal Spaces on the Web
- As an extension (or together with) the above project, this project leverages the fact that people often visit pages they had already visited to assist with web searching. The idea is that a personal cache is mantained, and searches particular to the personal cache can be done. The best way to do this project is to combine it with the New Zealand Digital Library project, which has terrific facilities for compressing, indexing and retrieving information.
- How People Reuse Files
- In previous work, myself and a variety of my students have analyzed the way people access things. This includes command lines in Unix, Man pages on Unix, phone dialling patterns, and Web Pages. I would like to see this analysis done on file use. The idea is to determine patterns on file access i.e., how people visit files they had previously visit. To do this project, the student would have to instrument a standard GUI (e.g., Windows or Mac) to collect data, and then analyse the data for these patterns (we already know what patterns to look for in our previous work).
- The Use of Transparent Layers for Overviews and Detail Views
- Donald Cox and Jas Chugh produced a see through system, where one could have a viewport of the space on a transparent front layer, and an overview of the entire space on a back layer. They did a small usability study of the system, which was published as a short paper. In this project, the system would be re-implemented on an SGI to use true transparency (alpha blending), and the tests redone more formally.
- A Canvas that Integrates Various Visualization Techniques (can be extended to include Groupware)
- Carl Gutwin, Andy Cockburn and I produced a variety of visualization techniques that work within a groupware setting: radar overviews, see through displays, fisheye lenses, and magnifying lenses. All these were implemented as separate systems. In this project, the student would merge all these approaches into a single device that allows particular effects to occur under program/user control. In essence, this is a cool information visualization project that could be extended to a groupware setting. However, it would require good coding skills. Some graphics knowledge is probably helpful.
Note: Some electric circuit design knowledge would be an assett for the projects described below.
- Ambient Displays
- Hiroshi Ishii and his students at MIT described ambient displays, which are displays (not necessarily screens!) situated in our physical environment that present information on the background of conciousness. Examples include shifting light patterns on a ceiling to represent some computer activity. This open-ended project is simply the exploration of these devices. I expect that background reading would collect what has been done on it, some taxonomy created, and example systems produced. Could be lots of fun, but would require the student to actually build physical devices. Some electric circuit design knowledge would be an assett.
- Computers for Intimate and Playful Interaction
- Computers can be much more than a display, a keyboard, and a mouse. In this project, the student would explore the idea of computers to support intimate and playful interaction. Examples include computerized jewelry and furniture, where the use of one would evoke a response in its companion piece (e.g., when one person strokes a pendant, a sound or motion is produced on their partner's pendant). This is a great area for creative people. You would have to look into background in tangible media, look for examples in the arts. You would also have to develop designs, prototypes, and proof of concepts. Some interesting technical problems exist as well in terms of wireless networking and miniturization.
- Wireless Networking for Small Devices
- (Good for someone with a telecommunications course). In this project, you would create the software and hardware that would allow two small devices to communicate. In essence, you would have to explore what is available (e.g., infra red) and write a communciation layer so that these small devices could communice to each other via computers.
- Digital but Physical Surrogates: Second Pass
- Hideaki Kuzuoka and I designed a variety of physical devices under digital control. Each device reacts to how a person moves about their instrumented office: thus a person can get a sense of what the other is doing and whether they are available for conversation. In this project, you would redesign the system to take into account problems, aesthetics, and other nuances that were not considered in our first version. For example, we have only worked with diads (two people) but not with several people.
- Video-Mediated Communication and its Relation to Proximity
- As part of the above project, we also created video units that reacts to a person's proximity to it. The idea is to control the quality of service of the video and audio as a relation to people's closeness to the paired devices. The goal is to minimize distraction and privacy intrusions, while still providing a good channel for casual interaction. Approaches include altering the video image as a function of proximity (e.g., blurring, resolution reduction, color reduction, timing of video snapshots). An interesting extension is to consider: how it would work with more than two people; how the video screen itself could move as a function of proximity (e.g., to face the person); how the video screen could tranmit head motion by mechanical movement.