If you produce or employ printed materials that are distributed to or across the general population, you'd like to support processing of your printed materials by the visually impaired. Understandably, you don't want doing so to cost a lot.
With the 2D barcode approach to encoding information that we advocate, the cost of adding the visually impaired to the population who can read your printed materials is near zero. Chances are you'd never be involved with the process; we imagine printers will master the necessary skills and produce the enhanced materials for you using low cost available tools.
Two dimensional pdf417 barcodes, such as those shown at the top of the menu to the right, support encoding roughly 1000 characters into a space about one inch long and 3/4 of an inch high. Certain handheld scanners can scan these barcodes and convert them to a machine (e.g. PDA) processable form.
The Computer Science challenge in our research is to choose a method for encoding information so that when it is processed by a machine (e.g. PDA) it can be delivered to a user in a manner that maximizes the user's needs and goals. Except under rare circumstances a user doesn't want to hear about all the details on a menu, or all of the nutritional information on a soup can.. Thus, information such as a menu's contents must be described in accordance with certain guidelines before it is encoded in a barcode. And a user's preferences must be captured and remembered, so that when the encoded information is processed, their preferences determine the nature of what is delivered.
Our goal: Deliver the information a user wants to hear as early as possible, and deliver in a manner that allows them to determine they've heard what they wanted to hear.
We've used web technologies such as XML and preference-driven delivery methods we've defined to meet our goal. If you're familiar with the challenges associated with delivering web content to wireless PDAs and cell phones, you understand the challenges we've had to address.
