================================================================

Name: Josh Tasman

================================================================

Project Title 1: Distributed Real-Time Rendering

Local Users: Graphics group, CAVE project

Score
	Useful to others: yes
	Has specific users: yes
	Can be scaled up to PC product: yes
	Divisible: yes
	In C++ on Suns: definately

	Overall: a useful, fun project that has applications within
the department (such as running on the CAVE) and could be scaled up to
a commercial product.  The nature of the project would draw on the
strengths of people in the group who have worked on graphics,
networking, and distributed systems, and would introduce elements of
real-time constraints, for an interesting and complex project.


Pros/Cons: We would really draw on people's expertise with graphics
(good), but everyone can't be expected to know the details.  However,
this might be a good large project for that reason: the correct design
and delegation and distribution of work becomes essential.




================================================================




Project Title 2: 3D IDE

Local Users: any developer (students, faculty)

Score
	Useful to others: yes
	Has specific users: yes- developers
	Can be scaled up to PC product: definately
	Divisible: yes
	In C++ on Suns: yes

	Overall: This project would develop an application taking
advantage of the relativly inexpensive yet advanced 3D hardware
populating many machines, including those in the Sun Lab.  The
application would be useful to anyone who programs, and allows for
major innovations in useful UI design.

Pros: A good division of design and implementation can occur between
the implementation and conceptual interface design; useful
application.

Cons: ?




================================================================




Project Title 3: Graphical Network Data-Flow Monitoring

Local Users: sysadmins

Score
	Useful to others: yes
	Has specific users: yes
	Can be scaled up to PC product: yes
	Divisible: yes
	In C++ on Suns: yes

	Overall: The field of Intrusion Detection is in its infancy.
One of the greatest difficulties encoutered so far is dealing with the
massive amounts of data generated in a typical session.  A
well-designed graphical representation of network flow and anomoly
spotting would be extremely useful.  The project could perhaps
frontend for a freeware ID system.

Pros: timely, useful.

Cons: Low-level networking details necessary for many people in the group.