CS263 Final Projects

Spring 1996

The SR Domain, Stephen Anthony Edwards.

Abstract: I propose to implement a new scheme for describing reactive embedded controllers. Based on the synchronous model of time used in the Esterel language, it will support building heterogeneous systems composed of blocks described in other models of computation. The semantics of this domain is a succession of least fixed-points of continuous functions defined on a pointed CPO that ensures determinism. I will implement this domain in the Ptolemy prototyping environment.

The Use of Formal Analyses in Compiling for Better Locality, Joe Darcy and Karl Czajkowski.

Types and Set Constraints, Masood Mortazavi

Extending Matlab for Symbolic Computation, Jianbo Shi and Trey Jackson

Abstract: Matlab, a numerical computation language, has become an increasingly popular prototyping tool for scientist and engineers. One of the main reason for its popularity is its simplicity. Matlab has only one data type, matrix, and all functions are build on top of matrix or matrix operations. As a result, one no long need to think or express program at a level of each array element or array index. Instead, the programmer is allowed to focus on the mathematical meaning of the algorithms, and write codes that are not only short but also very readable, almost pseudo-code like. However, as a language designed mainly for numerical computation, it is very hard to do any symbolic computation that uses data structures. In this project, we are looking at ways of extending Matlab to handle symbolic computations, while still maintaining its simplicity. Our main idea is to add one more data type, graph, into Matlab. We feel graph, instead of pointer, is the right level of abstraction for expressing symbolic computations, since most of data structures or symbolic algorithms are expressed in turn of graph or special form of graph such as trees, or lists. The problem now is to define the right set of build-in functions, and constructs that allow one to express large set of symbolic computation easily in this new language.

Language Support for Mixed Parallelism, Tzu-Yi Chen

VRML 2.0 (Moving Worlds) Semantics, Daishi Harada and Randy Keller

Split-Java, David Gay

A Functional Language for Real-Time Multimedia Applications, Amar Chaudhary

Brian Strand

An Algorithm and Correctness Proof for Evaluating PTAs, Zhendong Su

LRENE: A Specification Language for GUI Semantics, Steve Fink

Abstract: My project is the design of a specification language which describes the semantics of a graphical user interface. The language, LRENE, gives the programmer a syntax with which to describe common GUI tasks (e.g., displaying editable views of objects), control (e.g., disabling menu items as a result of events), and structure (e.g., similarity between views). LRENE promotes UI consistency by making consistent interfaces easier to build, rather than harder. Because it is a high-level description language, LRENE allows more effort to be spent on GUI design by supporting rapid prototyping and consistent semantics.

Languages for Specifying the Presentation of Tree-Structured Documents, Vance Maverick