| AOL Mgnet - my.aol.com | |
 Early concept principle sketch for Mgnet (pronounced “Magnet”)  Anthropological research consolidation of artifacts on recommendations  Sketched concept used for screen in my click-through prototype with recommendation image tag cloud on the right |
Problem:
A VP at AOL with distinct foresight wanted to get a head start in the web 2.0 personalization product arena. So he, along with our design VP, designated four designers to take part in AOL’s first iLab (Innovation Lab) project, called Mgnet. We participated in a 2-month-long intensive IDEO-style brainstorming, concepting, and design process that resulted in hundreds of ideas, sketches, and storyboard concepts with the idea of creating the best recommendation system on the web.
Solution: We created Mgnet, a cutting edge image-based recommendation system, where users can tell the system what articles, video, and audio content that they’re interested in by clicking images and rating items. The system gets better over time as it “learns” more about the user. Because of Mgnet’s “brain” concept (image tag cloud) it is now a patented product. My Contributions: One of my responsibilities was to research and consolidate the latest web 2.0 recommendation systems (last.fm, digg, tivo, Findory, etc), as well as finding and compiling social psychology research, and conducting anthropological research about how people make recommendations in the real world. Based on this, I worked collaboratively with the three other designers to generate hundreds of ideas, and draw sketches that summarized these ideas. Ultimately we would use these for our final storyboard concept. Our names are now listed in a patent of the “brain” (image tag cloud) idea that I helped conceive during these brainstorms. I also created a click-through prototype that we presented to our Executive team that resulted in the go-ahead to develop Mgnet. |
