Selected Projects

<< Back to home page
Delphi Museum Browser

Delphi Museum Browser

Spring 2007
My thesis project at UC Berkeley was the Delphi Museum Collection Browser. I worked with a team of Berkeley students and the Phoebe A. Hearst Museum of Anthropology to expose the Hearst's collection on the web. I worked primarily on the design and evaluation of the user interface. I contributed heavily to personae and scenario development, task analysis, lo-fi paper prototyping, and user testing with interactive prototypes. I developed the visual design in Photoshop and then implemented the design using CSS, XHTML, and the Smarty templating engine for PHP. I also designed and developed many of the site's interface interactions and AJAX functions with the jQuery JavaScript library.
Laser Scanning our Cultural Heritage

Laser Scanning our Cultural Heritage

Spring 2007
The CyArk organization creates accurate 3d models of cultural heritage sites using laser-scanning technology and distributes them for free over the Internet. As an intern, I was taught how to operate scanning equipment and process scan data. I participated in field work and post-processing for two projects: Fort Winfield Scott in San Francisco and the Cheney House in Berkeley.
VoteGuide

VoteGuide

Fall 2006
VoteGuide was a Center for Citizen Media project lead by Dan Gilmor, author of We the Media, and Bill Gannon, Sr. Editorial Director at Yahoo! News. I worked with students in the J-School and Ryan Shaw to build a community portal focused on California's 11th congressional district race between Richard Pombo and Jerry McNerney. The site aggregated relevant news stories, campaign materials, debate video, photos, and a blog maintained by the journalism students. I developed the site's visual design and built custom Drupal templates. I also helped code custom Drupal modules with PHP.
Internal Metadata and Online Video

Internal Metadata and Online Video

Fall 2006
Working with Owen Otto and Vlad Kaplun, we produced a needs assessment study and design recommendations for a hypothetical multimedia system combining UC Berkeley's course webcasts with internal metadata. Internal metadata is a term we coined to refer to information related to a frame or segment of a video, rather than the video as a whole. In our study we looked at several types of internal metadata including, deep tagging (keywords) and text entries (longer, free-text notes or comments). We used three primary methods: interviews (4 subjects), a survey (158 responses), and a comparative analysis (8 systems).
Labsys

Labsys

2004-2005
In my senior year of college I built Labsys to manage the School of Print Media Labs. The labs employ around thirty students and serves hundreds of others with computers, scanners, proofing devices, and miscellaneous services. I developed Labsys with PHP and MySQL on the back end and XHTML and CSS on the front end. I incorporated open-source blogging and wiki software with custom built components for scheduling and problem tracking. Labsys was very well received and the lab staff continues to use and improve the system.
JazzTimes

JazzTimes

Summer 2004
JazzTimesis a monthly magazine with circulation around 100,000 worldwide. Their website is written in ColdFusion on top of an Access database. Working with the Chris Lewis, the online publisher, I did away with the preexisting HTML frames and added headers and footers to every page to simplify management. I converted many flat pages into a few database driven pages. Finally, reorganized the site's navigation, created a new visual design, and implemented the new design in CSS.
Classifing Wikipedia Articles

Classifying Wikipedia Articles

Fall 2006
I worked with Ryan Shaw and John Chuang on an automated method for accessing quality in Wikipedia articles. We identified twenty features of a Wikipedia articles such as the number of revisions, inbound and outbound links, reading level, references, and markup structure. Then, using articles marked by the Wikipedia community as "good" or "featured" for training material, we evaluated several algorithms including K-Nearest Neighbors, Random Forests, and Support Vector Machines. In the end we were able to correctly classify about 98% of our test set.
Zeu$ Power Meter

Zeu$ Power Meter

Fall 2006
The Zeu$ Power Meter measures the energy consumption of anything that can be plugged into the wall and transmits data wirelessly to a computer. As part of a product design course, I worked with Allan Gu, Sameer Kelkar, and Jacques Burrus to develop a working prototype meter. We built the device with sensor motes and a custom-made plastic enclosure. I built a prototype web application that aggregated data from the meters (assuming you had more than one) and charted energy usage. The idea was to allow people to compare the usage of, for example, their refrigerator to that of a new refrigerator. I acted as the team's project manager and together we produced the prototype device, a business plan, and several posters.