Richard C. Waters

Audio interactive tutor

An all-audio, hands-free and eyes-free interactive tutor provides time-efficient and reduced-boredom instruction by varying a course of instruction based on the correct and incorrect responses of the user so as to frequently repeat and provide positive feedback for poorly-learned items of knowledge, to periodically refresh well-learned items and to suggest that rest be taken or to switch to easier material whenever too many human errors have occurred. The audio interactive tutor includes an audio input module, in the preferred embodiment a voice recognition unit based on a finite state grammar, an audio output module, a course of study and a user model. In a preferred embodiment, the course of study provides for human error control as well as for voice recognition unit error control, which synergistically cooperate to render the tutor substantially fault-tolerant and therewith enables the employment of commercially-available but not excessively accurate voice recognition units. The tutor may be embodied to operate on a PC or workstation, or may be embodied to operate as a stand-alone tutor. Spanish language instruction is disclosed in an exemplary embodiment.

Diamond park and spline: Social virtual reality with 3d animation, spoken interaction, and runtime extendability

Diamond Park is a social virtual reality system in which multiple geographically separated users can speak to each other and participate in joint activities. The central theme of the park is cycling. Human visitors to the park are represented by 3D animated avatars and can explore a square mile of 3D terrain. In addition to human visitors, the park hosts a number of computer simulations, including tour buses and autonomous animated figures. Diamond Park is implemented using a software platform called Spline, which makes it easy to build virtual worlds where multiple people interact with each other and with computer simulations in a 3D visual and audio environment. Spline performs all the processing necessary to maintain a distributed, modifiable, and extendable model of a virtual world that is shared between the participants. For more information visit http://www.merl.com.

QOTA: a fast, multi-purpose algorithm for terrain following in virtual environments

To keep avatars and other moving objects on the ground in virtual environments, it is necessary to find the points where these objects should contact the terrain. This is often done using collision detection; however, this is inefficient, because general collision detection solves a problem that is inherently more complex than merely determining terrain contact points. Because the Quick Oriented Terrain Algorithm (QOTA) focuses solely on the problem of intersecting lines of a predetermined orientation with a terrain model, it provides very rapid support for terrain following. For example, given a 13,000 polygon terrain, QOTA running on a 250MHz R4400 MIPS processor can calculate an intersection point in less than 19 microseconds (1.9 x 10 -5 seconds). Given a preferred orientation, such as the direction of the gravity vector, for the lines to be intersected with a terrain, QOTA uses a pre-processing step that sorts the terrain polygons into a quadtree and adds bounding boxes and polygon edge equation parameters to speed up polygon containment checking. In the example above, this preprocessing takes approximately 2 seconds. In addition to terrain following, QOTA is useful for detecting certain limited kinds of collision detection and determining containment. This work may not be copied or reproduced in whole or in part for any commercial purpose. Permission to copy in whole or in part without payment of fee is granted for nonprofit educational and research purposes provided that all such whole or partial copies include the following: a notice that such copying is by permission of Mitsubishi Electric Research Laboratories of Cambridge, Massachusetts; an acknowledgment of the authors and individual contributions to the work; and all applicable portions o f the copyright notice. Copying, reproduction, or republishing for any other purpose shall require a license with payment of fee to Mitsubishi Electric Research Laboratories. All rights reserved. Copyright

Cliché-based program editors

A cliche-based program editor supports the construction and modification of programs in terms of standard algorithmic fragments (cliches), as opposed to merely supporting the construction and modification of programs in terms of syntactic or textual units. Three experimental cliche-based program editors have been implemented, exploring the tradeoff between power and speed. The Knowledge-Based Editor in Emacs (KBEmacs) is the most powerful of the three editors. It supports a wide range of editing operations and can represent a wide range of cliches, because it uses an internal representation called plan diagrams, which combines features of flowcharts and data flow schemas

Macroexpand-All: an example of a simple lisp code walker

If you like to write Lisp macros, or even just use the macros other people write, you have no doubt felt the desire to see what particular macro calls expand into. The standard Common Lisp function macroexpand is very useful in this regard; however, since it only expands the topmost form in an expression, it does not necessarily show you the full result of a macro expansion.

To nreverse when consing a list or by pointer manipulation, to avoid it; that is the question

A situation that arises all the time in Lisp is the need to create a list of elements where the order of the elements in the list is the same as the order that they are created in time---i.e., the first element computed is the first element in the list, the second element computed is the second element in the list, etc. There are two basic ways of doing this: the nreverse approach and the rplacd approach. In the nreverse approach, you push the elements onto the list as they are computed and then use nreverse to put the list into the correct order after all of the elements have been computed. In the rplacd approach, you maintain a pointer to the end of the list and use rplacd to put each element directly into its proper place in the list.