Clinic Projects

Please click on a link below to view the Harvey Mudd College Computer Science Clinic projects for the corresponding time period.

Clinic Projects for 1994-1995

Voice Mail Internationalization

Client
Octel Communications Corporation

Faculty Advisor
Professor Josh Hodas

Student Team
Juile Lowery, Craig Persiko (Team Leader), Jim Winstead, and Peter Yoo
This project was concerned with the internationalization of voice mail systems, and involved work on models and abstractions for the complex system of prompts that need to be stored in order to make a modern voice mail system work. One of the results of this project was an interactive tool which assists translators in translating voice mail prompts between languages, by showing the possible contexts of phrases and by making it easy to change the phrase structure.

Radiology Treatment Scheduling System

Client
Optivus Technology, Inc.

Faculty Advisor
Professor Ran Libeskind-Hadas

Student Team
Darin Grant, Gregg Snodgrass, and Dan Snyder (Team Leader)
This project involved design, implementation, and testing of a patient and personnel scheduling system in a department of a modern hospital, Loma Linda University Medical Center. This project entailed extensive systems analysis to determine methods for scheduling a large number of concurrent tasks to minimize time lost due to confusion in the face of schedule changes. The result is a client/server implementation suitable for near-term deployment.

Network Management by Delegation

Client
The Aerospace Corporation

Faculty Advisor
Professor Mike Erlinger

Student Team
Elecia Engelman, Nathan Tuck, Adam Wells, Chris White, and Phil Winston (Team Leader)
This was a research project in distributed network management. An advanced, heretofore unimplemented, form of management called Management by Delegation was studied for feasibility and a prototype was implemented by the student team. Management by delegation allows network management tasks to be distributed among multiple hosts within the network, resulting in improved network performance (through parallelism), scalability (since all management tasks don't fall to a single host), and reliability (since there is no dependence on any one host). needs.