Project Ideas for the CourseCapturing and Presenting Relevant InformationIn Envisionment and Discovery Collaboratory (EDC) problem solving situations, construction activities and information relevant to the current problem are tightly coupled. One kind of information that would be useful in many domains would be resources from information providers, such as news feeds. Currently, there is no direct or automatic connection between information resources and the EDCs information space. People who find interesting articles must add them to the EDC by hand. In this project, you would explore how news feed or other similar streaming information sources could be used within the EDC. You would explore issues such as the capture, organization, and delivery of information. Relevant resources may include sites such as the Boulder Daily Camera. One example design problem would be deciding how automated the capture and delivery can / should be and what tools would be necessary to facilitate the information gathering process. More Information: http://www.cs.colorado.edu/~l3d/systems/EDC/demo/demo10.html Reading: Arias, E. G., Eden, H., Fischer, G., Gorman, A., & Scharff, E. (2000) "Transcending the Individual Human Mind–Creating Shared Understanding through Collaborative Design," ACM Transactions on Computer Human-Interaction, 7(1), pp. 84-113. Contact Person: Eric Scharff Gathering information from Geographic Information SystemsEDC problems typically involve some geographic or spatial component. Both the EDC domains (transportation, flooding) and the representations (maps, diagrams) are well suited for geographic information. There is a wealth of existing data in Geographic Information Systems (GIS) that professionals often use as part of their decision making process. Capturing and using this information could help improve the authenticity of some of the situations discussed in the EDC. Information such as census information, maps, expert models, and other GIS data could be tapped for use in an EDC planning task. In this project, you would look at what information exists in GIS systems and how this information could be used within the EDC. You would explore information such as finding the information that exists and the challenges of extracting and interfacing with existing tools. Relevant resources may include US census data and Boulders GIS resources. One example design challenge would be determining tradeoffs between importing data and having multiple tools used in a query process. More Information: Reading: Contact Person: Eric Scharff Discussion Forum:Discussion: [PROJECT] EDC with GIS Capturing Feedback from Remote ParticipantsThe EDC relies on a physical construction space. Unfortunately, only a limited number of people can interact with this construction space at the same time. Is it possible to increase the number of participants in the construction process? For large group meetings, technologies such as wireless computers or PDAs may help give a voice to people who are not around the table. In this project, you would explore how new technologies (such as wireless systems) may be used to involve a larger group of face-to-face participants. You would explore different strategies for getting people to participate, and technologies that may support that participation. Relevant resources may include existing work on voting and chat technologies. One example design challenge would be determining what kind of information would be contributed by individuals, and what social situations might be necessary to support large-scale interactions. More Information: Reading: Abowd, G. D. & Mynatt, E. D. (2001) "Charting Past, Present, and Future Research in Ubiquitous Computing." In J. M. Carroll (Ed.) Human-Computer Interaction in the New Millennium, ACM Press, New York, pp. 513-535. Contact Person: Eric Scharff, Shinichi Konomi Context-Aware Computational Environments – Integrating Artifacts with the Decision Surrounding themOur past work centered on domain-oriented design environments has been based on the following simplifying assumption: all design activities happened inside the computational environment rather than some of them happening in the external world. With the EDC, we need to extend our approach by creating environments that integrate computational environments and (computationally enriched) external physical worlds with mechanisms capturing the larger (often unarticulated) context of what users are doing. For example: a fundamental shortcoming of the current prototype of the EDC is that there is no capture of the discussions in which stakeholders engage during design sessions. In this project, you would explore and investigate a variety of critical and important research problems, including the capture of design rationale, as it is articulated in the discussions and design sessions by integrating the artifact under construction with the discussions around it. This will address the failure of design rationale systems of the past that required extra efforts of scribes to document in the computational environment things that are happening in the surroundings. Some of the following research issues can be explored in this project:
More Information: Reading: Fischer, G. (2001) "Articulating the Task at Hand and Making Information Relevant to It," Human-Computer Interaction Journal, Special Issue on "Context-Aware Computing", 16, pp. 243-256. – available at: http://www.cs.colorado.edu/~gerhard/papers/hci2001.pdf Contact Person: Hal Eden Capturing Feedback Between MeetingsThe EDC emphasizes bringing people together to discuss problems face to face. However, not all problems can be solved in a single setting, and not every relevant stakeholder can be present at all meetings. Currently, the EDC only uses a simple Web annotation system to support this asynchronous discussion. How can we complement synchronous meetings with other asynchronous information sources? In this project, you would explore what features would be necessary for asynchronous interaction. More specifically, you could look at how Web tools (like discussions, annotation tools, outliners) can capture the results of meetings and how people who are not present can present their opinions. Relevant resources would include some of the sharing and annotation features provided by the Swiki. One example design challenge would be understanding what form user comments should take (such as voting, discussion, and so on) and how to summarize parts of a face-to-face meeting for people not present. More Information: Reading: Moran, T. P. & Carroll, J. M. (Eds.) (1996) Design Rationale: Concepts, Techniques, and Use, Lawrence Erlbaum Associates, Inc., Hillsdale, NJ. our own contribution: Fischer, G., Lemke, A. C., McCall, R., & Morch, A. (1996) "Making Argumentation Serve Design." In T. Moran & J. Carrol (Eds.), Design Rationale: Concepts, Techniques, and Use, Lawrence Erlbaum and Associates, Mahwah, NJ, pp. 267-293. Contact Person: Eric Scharff, Gerhard Fischer "Virtual Stakeholders" (Critics) and Making Users Feedback ActiveIn most specific domains, some generally accepted rules emerge. For example in the transportation domain, one such rule might be: "Two bus stops should not be further apart than 500 yards". These rules can be embedded in systems and "critique" (representing the design knowledge of virtual stakeholders) design activities as they take place. Critics are computational entities that can analyze a computer model of a problem and give feedback based on a certain perspective. People participating in the EDC come to the table with a specific agenda and a set of personal constraints – and these constraints may be different from the critics existing in the system. Capturing people's own specific and additional constraints, helping them make the constraints explicit, and evaluating a design based on user constraints are all very important tasks. In this project, you would explore how a user may express their perspective in an active manner, perhaps in the design of critics. Relevant resources include existing critiquing systems and other active feedback systems (such as spelling correctors.) One example design challenge would be determining the kinds of constraints a user might want to specify in an active way, and whether it is feasible to represent those constraints in an automatic or semi-automatic fashion. More Information: Reading: Nardi, B. A., Miller, J. R., & Wright, D. J. (1998) "Collaborative, Programmable Intelligent Agents," Communications of the ACM, 41(3), pp. 96-104. Fischer, G., Nakakoji, K., Ostwald, J., Stahl, G., & Sumner, T. (1998) "Embedding Critics in Design Environments." In M. T. Maybury & W. Wahlster (Eds.), Readings in Intelligent User Interfaces, Morgan Kaufmann, San Francisco, pp. 537-559. Contact Person: Eric Scharff, Gerhard Fischer Empirical Study of CodeBrokerOne of the major advantages brought by object-oriented programming languages is class libraries that contain hundreds or even thousands of classes and methods that can be reused by programmers. However, mastering such library classes and methods presents huge learning challenges for programmers. It is impossible for programmers to learn all of them before they start programming. A practical approach is for them to learn on demand, namely, to learn a new class or method when it is needed during work. To support learning on demand, we need to re-think the design of the interface of class library systems. Such library system interfaces must be able to make programmers aware of the existence of the classes and methods that they do not yet know but can be reused in their current programming tasks. CodeBroker is a library system that supports Java programmers in learning classes and methods on demand. As an autonomous software agent, CodeBroker continuously runs in the background of the programming editor–Emacs. It infers the task of programmers by analyzing the doc comments and signatures (syntactic definitions of methods) entered unto the editor, and autonomously locates and proactively delivers classes and methods that can possibly be reused in the current programming task. To ensure that only the classes and methods the targeted programmer does not yet know would be delivered, CodeBroker uses user models to represent each programmer's existing knowledge of the library. Because each programmer's knowledge changes, a user model in CodeBroker updates accordingly as the system learns from analyzing programs written by the user (programmer). The CodeBroker system has been implemented and initially evaluated. More information, including a one page description and usage scenarios, can be found at http://www.cs.colorado.edu/~yunwen. The project for this class is to empirically study the usefulness and usability of the CodeBroker system. There are at least two possible studies to be conducted.
Findings from those studies will lead to the improvement of the system and better and more efficient ways that programmers learn and reuse library classes and methods. More Information: http://www.cs.colorado.edu/~yunwen. Reading: Ye, Y. and G. Fischer: "Supporting Reuse by Delivering Task-Relevant and Personalized Information", in Proceedings of ICSE'2002: International Conference on Software Engineering, Buenos Aires, Agentinia, May, 2002, (in press) [http://www.cs.colorado.edu/~gerhard/papers/1cse2002.pdf] Contact Person: Yunwen Ye, Center for LifeLong Learning and Design –email: yunwen@cs.colorado.edu
|
