CS190 2005 Project Ideas
The Parking Lot Monstrosity
You may know that at Brown one is on the (waiting) list for several parking lots at a time, and one never knows one's status. Imagine a web interface to the database: with a Brown Auth-ID, I can look at the size of the waiting lists for all lots, the number of spaces per lot, the number of cars allocated to a lot (I expect this exceeds the number of spaces by some small amount), and at my position on the waiting list. I can even decide to switch waiting lists if I so choose.
Things to consider: designing a user interface to display information effectively; maintaining security (dealing with Brown auth-IDs, for instance); and gathering requirements from the parking folks, who will surely be reluctant, while fulfilling the desires of the community. This seems to be a smaller project, but there may be more to it than has as yet been considered.
Submitted by Prof. John Hughes (Computer Science Department)
The Music Suggestion
The following is a quick idea for a project from a professor who teaches a music class on networked improvisation.
The basic idea for our setup is that several performers at computers are doing real-time audio DSP and sharing control data via the network, so that each computer can interpret and reassign the control data generated by other players. One thing we were lacking was a comprehensive server app that would serve as a message-passing application, allowing individuals to sign on, have their IP address registered, poll other individuals, control message routing configurations (between other players), etc. The server app could also keep statistics on control message activity and generate decisions on routings, mappings, based on what had passed through the server.
Submitted by Prof. Butch Rovan (Music Department)
BikeQuest
One of the most frustrating aspects of using a bicycle as your main mode of transportation is the time required to find safe and efficient bike routes. There are all sorts of computer and online mapping programs dedicated to route planning for cars, but they really aren't adequate for bike route planning. Are you interested in maps? Do you have an interest in route planning? Are you interested in UI design? Would you like to help the biking community?
BikeQuest, a blend of mapquest style online maps and routing merged with ebay style route feedback. This system would allow users to query a map database for routes which are known to be bike friendly. How would it accomplish this? An initial set of route ratings would be input into adatabase, based on their perception of how bike friendly various roads would be. Overtime, the database would be trained by feedback from a user community who would rate certain route segments with various bike related criteria; for instance, the existance of heavy traffic or horrible road conditions. The predefined data would be merged with the ever changing user feedback, to provide the best known, bike friendly, route.
Unlike people driving cars, bikers often want to just go out for a spin of X miles, where the starting and ending point are the same. So, in addition to regular map routing, BikeQuest could offer loop routing. Just imagine a world where, the next time you want to go for a bicycle ride, you just decide how far you want to go, you bring up a web browser, input the data and are presented with a route.
Submitted by Mark Dietrich (Computer Science Department, Technical Staff)
A Food Services Dilemma
Atomic is a database program currently in place in the SMO for tracking hours for payroll and for maintaining payrate information. It is based in Microsoft Access 1997 and was implemented around 1999. It is used to keep track of almost all employee information during the semester and is essential to the major efforts of the SMO every semester, including recruitment and signing up new workers, reapplications, pay raises, and shift changes. The problems include that that: it is outdated and large portions are no longer useful or functional; it crashes from time to time and generally requires days to get up and running again; it is not extensible and is preventing the office from upgrading from MS Office 1997. See a formal proposal for more information.
Contact Jacquelynn Henry (BuDS)
The Engineering Request
My IBM cluster is finally coming up to life, and our new conf room was built to eventually work as a smart room. It already has 4 projectors, and a wall covered with screen material. But there is no software yet. I am interested in building a system for remote collaboration here, where documents and applications could be shared, and high-end systems as well as workstation participants could interact. Access Grid could be a foundation framework for this, but how data is captured, managed, and presented could be improved. The emphasis could be on ways to improve the product design and review cycle in industry.
Submitted by Prof. Gabriel Taubin (Engineering Department)
The SAO/CIS Proposal
As you can easily imagine, there are large numbers of homegrown and student-built software around campus. Occasionally, CIS gets called in to maintain these things, which can be very difficult because CIS was not involved in the design and the app may be constructed from unfamiliar tools and technology. See a formal proposal for more information.
Submitted by Alan Usas (CIS)
Main contact Ricky Gresh (Student Activities Office, Director)
A Few Public Service Ideas
Increasingly, immersive games are seen not just as toys, but as effective training tools for skilled specialists and leaders. A growing number of universities are establishing programs to train undergraduates to build games for serious and educational purposes (see wistechnology article and seriousgames ). Significant opportunities exist in Providence to build collaborations between undergraduate/graduate software design teams and business and community organizations. Below are some ideas that could be carried out in conjunction with a couple of community based organizations and high school teachers.
CAD for KIDS
An architectural and engineering authoring environment for middle- and high-school students. This program would offer specific design palettes for rigourous, Providence-specific design tasks, such as building your own school, building your house, designing your dream school, or designing an addition to the mall. Resulting designs could be integrated into "multiplayer persistent online Providence."
MULTIPLAYER PERSISTENT ONLINE PROVIDENCE (MPOP)
A virtual Providence that records changes to environment that come aboutthrough user modifications. This application could draw on satellite imagery to add to the realism of navigation. The resulting environment would operate both as an online Sims-type environment, as well as a platform for designing subsequent, user-specific applications. A "city-planning" application might model traffic flow changes that would result from a highway change, while an "alternative energy education" application might show students what a wind-farm would look like on top of the Fleet Building or Neutaconkanut Hill. CAD for KIDS is another such example. This would be the first graphical design environment for building Providence-specific games and planning simulations, and could allow students and community members to participate in the design of specific applications.
'Augmented Reality' Wiki Authoring Environment (ARWAE)
Wikipedia has established itself as one of the most comprehensive collaborative web-projects ever. The site features over 300,000 entries. Looking ahead, subsequent generations of Wikipedia-type data repositories will likely be accessible via mobile computing devices, such as PDAs or eyeglasses-based viewers. Context-specific data can be fed to the device, such as a listing of bus lines that stop at a specific bus stop, or the architectural information for a famous building. The ARWAE project would open opportunities to incorporate GIS and visualization technologies into their final product. Building a platform to author and distribute "augmented reality" information to users would also challenge students to understand user/editor psychology and information networks.
Submitted by Mackay Miller (Swearer Center, Assistant Director for Youth Programs)
Borealis
I am a software engineer working on the Borealis project for Stan Zdonik. We have a need for a monitor tool to visualize activity in a Borealis application. Borealis applications are organized as a dataflow diagram. The diagram has both XML and C++ class representations. Throughput statistics are collected by Borealis that can be used as dynamic input to the monitor. The project can scale to the capacity of the class as features can easily be dropped or added. Some of the things we might want to monitor include: box movement; load shedding; node load; node failure; and queue size.
Submitted by Bradley Berg (Computer Science Department)
The Football Idea
In the football office, we mail out numerous recruiting questionnaires to high school students--well over 10,000--and manually enter the returned information into a filemaker database, which of course is extremely time consuming. A program which allows high school football athletes to fill out our questionnaire online from our web page and have the information directly inputted into our recruiting database would be incredibly useful. The questionnaire is already available on our web site.
Submitted by Donna Fanion (Brown Football)
The "FBIdea" (or some better name)
Implement an indoors wireless monitoring/surveillance system that tracks certain objects of interest (laptops, projectors, or even people, etc) using a combination of COTS passive and active sensors/actuators (e.g., RFIDs and wireless motorized cameras). Users should be able to register continous queries that express what objects need to be monitored and what events are deemed "interesting" (so that alerts can be sent to appropriate parties).
Submitted by Prof. Ugur Centintemel (Computer Science Department)
Grants Management System Proposal
Background:
The Center for Statistical Sciences (CSS) is one of seven research centers in the Department of Community Health. CSS supports a number of different faculty and each faculty has multiple grants. Most grants are multi-year and have multiple people with different percent effort e.g. a staff member may be paid from 4 different grants in any fiscal year and this may change monthly. Currently, each grant is managed with an excel spreadsheet with only a single person having access at any given time.
Problem:
CSS does not have an efficient management system for the grants. Retrieval of simple information is an arduous and time-consuming task. Time and effort, budgets, expenditures, etc. are all processed and managed separate from each other. It would be useful to have a way to integrate the processes and be able to query and report on various aspects of the grant throughout its life.
In addition, multi-year grants are the norm and existing information from spreadsheets would need integration into the new system. A new grants management system would save time, money, and the information would have more integrity if it was entered and managed through a central interface. Ideally, the system will allow multiple people to work in the system simultaneously. Finally, items such as fringe benefit rates, indirect cost rates, salary, graduate student tuition and health fees change often. We often have one rate from January through June and another from July to December. Grants can start in any month. So keeping up with the grant year and the fiscal year is also challenging and a consideration for programming.
Submitted by Holly Graham (Center for Statistical Sciences)
Please email the CS190 TAs to propose a new project idea and have it added to the list.
Last modified: Wed Jan 26 16:56:01 EST 2005