Home Syracuse University College of Law NYSTAR - New York State Foundation for Science, Technology & Innovation

WebAnywhere Expands Web Access for the Blind

Research Summary

At University of Rochester, Assistant Professor Jeffrey Bigham has been leading an advanced research team in the development of new assistive technology to expand Web access for the blind. In 2009, they made a significant contribution towards this goal with the invention of WebAnywhere, a self-voicing, Web-browsing application that can work off of practically any computer with an Internet connection. The application was designed to address the general absence of such Web-browsing applications for blind users. Despite the fact that the U.S. blind population is more than 1 million, no applications of this sort have been made available at an affordable price. WebAnywhere is an open source project and is free. What’s even more impressive is that Dr. Bigham’s research design taps into already installed components of Web browsers to do the bulk of the work – no software needs to be installed on the computers being used. So, whether you’re in the library or at an Internet café, WebAnywhere lets you access information and communicate through the Internet in a collaborative and fully functional way.

Current Research
Professor Bigham’s WebAnywhere project has been conducted in collaboration with Professor Richard E. Ladner at the University of Washington’s Department of Computer Science and Engineering. Their system is still in beta form, but because of the positive response it is already receiving, they have made a beta form available for download on their website. The WebAnywhere project is part of a broader research framework, referred to by Professor Bigham as ‘WebInSight’. The concept behind their research was simple: to explore obstacles that impede access to the Web, and then develop new technologies to overcome them.

The WebInSight research team also wanted to provide the blind with improved access to images on the Web, a concept that has eluded Web researchers for the better part of a decade. WebInSight, and WebAnywhere in particular, have focused on providing text to images on websites that were otherwise inaccessible to the blind. Preliminarily, Bigham and a team of researchers exposed the flaws behind countless Web images that should, but did not, contain text in their coding. The WebInSight design fills that text void, however, by processing web requests associated with links (to e-mail addresses and websites, for example) and then providing alternative text to interpret through the WebAnywhere application. The result is enhanced navigability on image-rich websites for the blind, filling the gap experience by those who do not see the images on the screen.

To assign alternative text to images on the Web, the WebInSight research team developed three novel approaches: (1) enhanced Web context analysis; (2) optical character recognition (OCR); and (3) human labeling. As a result, HTML img tags now have an associated alt attribute, based on the WebInSight technology, and can then use that text-based code to convey a representation of that image to the blind. In its simplest form, a blind person can sit at a public library computer with Internet access, plug in a set of headphones to the computer, and go to the WebAnywhere home page.

From that point, audio support is triggered, and a user-friendly URL bar tailored to the blind opens, allowing the user to surf the Web with the WebAnywhere technology acting as a voice-guide. The voice companion describes everything, whether email messages or up-to-date schedules for public transportation. The self-voicing, Web-browsing application makes virtually all Web content available from any computer, running on any operating system, and due to its simple design, the load time is less than five seconds.

Funding and Papers
Professor Bigham received support during his doctoral dissertation on WebInSight research from the National Science Foundation and a subsequent Boeing Professorship.

Papers on WebInSight research include the following:

Bigham, J.P. et al., WebInSight: Making Web Images Accessible, Proceedings of the International ACM SIGACCESS Conference on Computers and Accessibility (Assets 2006), pp. 181-188. Portland, Oregon, 2006.

Bigham, J.P. Increasing Web Accessibility by Automatically Judging Alternative Text Quality, Proceedings of the 12th International Conference on Intelligent User Interfaces: Short Papers Track (IUI 2007), pp. 349-352. Waikiki, Hawaii, 2007.

Jeffrey P. Bigham
University of Rochester
Department of Computer Science
Rochester, NY