The advent of PDAs, digital cameras, audio recorders, and wearables enables the collection of useful information in a variety of media. But how should it be organized? Information Pipette aims to use software agents to organize automatically the information to "make sense" for a particular task or user's context. For example, we are working on a scenario of assembling directions by taking pictures along a route or by recording audio advice at each step. Audio is run through a speech recognizer and pictures through an Optical Character Recognition (OCR) program (to read the street signs). While each of these methods alone is ineffectual, their information taken together and compiled with knowledge contextualized for the task can produce conherent descriptions.