Browsing by Author "Maulsby, David"

Now showing 1 - 8 of 8
  • No Thumbnail Available
    ItemMetadata only
    COMPUTER ANIMATION AT THE UNIVERSITY OF CALGARY
    (1984-04-01) Wyvill, Brian; Maulsby, David; McPheeters, Craig
    The computer has been used as a tool to aid animation artists in a number of ways; from controlling rostrum cameras to producing complete animation sequences from three dimensional descriptions of real objects. This paper describes Graphicsland, a suite of programs developed to explore the medium of 3D animation. Graphicsland recognizes that there is no single "correct" approach to solving the problems of design, storage, movement, rendering and production of animated film and video. Instead a variety of techniques are offered and the design of the system is such that it may easily be extended to incorporate new algorithms and graphical databases in the future.
  • Thumbnail Image
    ItemOpen Access
    I, METAMOUSE
    (1988-06-01) Witten, Ian H.; Maulsby, David
    "Metamouse" is a Flatland device that learns procedures in interactive graphics taught to it by untrained users. It lives in a simple graphics world and induces procedures in real time, taking full advantage of the interactive situation to suppress, or at least control, variability in the teaching sequence. The teacher must demonstrate what is to be done, employing constructive methods to make constraints readily apparent. Metamouse predicts actions, asks for constructions, solicits input parameters when required, and induces a program (including conditionals and loops). By eagerly carrying out predicted actions (which the teacher can undo if they are incorrect) it helps the teacher to satisfy appropriate felicity conditions. In this paper Metamouse explains its world, its body and mind, and what it can do.
  • Thumbnail Image
    ItemOpen Access
    INDUCING PROCEDURES INTERACTIVELY ADVENTURES WITH METAMOUSE
    (1988-12-01) Maulsby, David
    Direct manipulation interfaces have greatly extended the class of casual computer users and encouraged them to conceptualize the system through metaphors. They have not, however, successfully incorporated facilities for end-user programming without breaking out of the direct manipulation paridigm. This thesis supports the contention that "teaching" provides an appropriate metaphor for programming in such an environment. It presents a system for inducing procedures that enables users of a graphics editor to teach it routine tasks by working through example traces. A central problem in the design is to meet the requirements for instructibility without imposing excessive demands on the teacher. A key component of the system is its teaching metaphor, a graphical apprentice called Metamouse. Metamouse is the target of the teacher's demonstrations. It is an eager learner designed to encourage constructive methods, clarify ambiguous situations, reduce errors and extraneous activity, and discourage free variation in teaching. Its behaviour is expected to be understood by users at a metaphorical, intentional level rather than from a precise specification. Metamouse has been fully designed but not yet fully implemented. However, a pilot system has induced procedures with variables, generalized actions, conditional branches and loops. Its ability to reduce errors and extraneous activity by prediction, and to identify underspecification, has been demonstrated. Tests showed that the metaphor is easily understood. Consequently the thesis argues that it is feasible for a system to induce procedures interactively from casual users. This significantly broadens the scope of application of machine learning techniques and opens new areas of research in knowledge acquisition. It facilitates the investigation of intelligent user interfaces and, last but not least, benefits the many users of interactive graphics systems.
  • Thumbnail Image
    ItemOpen Access
    Inducing procedures interactively: adventures with metamouse
    (1988) Maulsby, David; Witten, Ian H.
  • Thumbnail Image
    ItemOpen Access
    Instructible agents
    (1994) Maulsby, David; Witten, Ian H.
  • Thumbnail Image
    ItemOpen Access
    PREDICTIVE INTERFACES: WHAT WILL THEY THINK OF NEXT?
    (1991-11-01) Greenberg, Saul; Darragh, John; Maulsby, David; Witten, Ian H.
    This chapter explores the application of predictive techniques to the human-computer interface. If a machine could predict what actions a computer user was about to take, it could perform them automatically, thereby reducing the amount of manual input required. This has proven to be particularly helpful for people with physical disabilities. The present chapter introduces predictive interfaces, discusses their general attributes, and provides examples of several quite different systems. The systems differ substantially from each other in their task domain, intended users, algorithm for generating predictions, and user interface.
  • No Thumbnail Available
    ItemOpen Access
    PROTOTYPING AN INTELLIGENT AGENT THROUGH WIZARD OF OZ
    (1992-09-01) Mander, Richard; Greenberg, Saul; Maulsby, David
    Turvy is a simulated prototype of an instructible agent. The user teaches it by demonstrating actions and pointing at or talking about relevant data. We formalized our assumptions about what could be implemented, then used the Wizard of Oz to flesh out a design and observe users' reactions as they taught several editing tasks. We found: a) all users invent a similar set of commands to teach the agent; b) users learn the agent's language by copying its speech; c) users teach simple tasks with ease and comple x ones with reasonable effort; and d) agents cannot expect users to point to or identify critical features without prompting. In conducting this rather complex simulation, we learned some lessons about using the Wizard of Oz to prototype intelligent agents: a) design of the simulation benefits greatly from prior implementation experience; b) the agent's behavior and dialog capabilities must be based on formal models; c) studies of verbal discourse lead directly to an implementable system; d) the designer benefits greatly by becoming the Wizard; e) qualitative data is more valuable for answering global concerns, while quantitative data validates accounts and answers fine-grained questions.
  • No Thumbnail Available
    ItemOpen Access
    PROTOTYPING AN INTELLIGENT AGENT THROUGH WIZARD OF OZ
    (1992-09-01) Mander, Richard; Greenberg, Saul; Maulsby, David
    Turvy is a simulated prototype of an instructible agent. The user teaches it by demonstrating actions and pointing at or talking about relevant data. We formalized our assumptions about what could be implemented, then used the Wizard of Oz to flesh out a design and observe users' reactions as they taught several editing tasks. We found: a) all users invent a similar set of commands to teach the agent; b) users learn the agent's language by copying its speech; c) users teach simple tasks with ease and comple x ones with reasonable effort; and d) agents cannot expect users to point to or identify critical features without prompting. In conducting this rather complex simulation, we learned some lessons about using the Wizard of Oz to prototype intelligent agents: a) design of the simulation benefits greatly from prior implementation experience; b) the agent's behavior and dialog capabilities must be based on formal models; c) studies of verbal discourse lead directly to an implementable system; d) the designer benefits greatly by becoming the Wizard; e) qualitative data is more valuable for answering global concerns, while quantitative data validates accounts and answers fine-grained questions.