For most computer users, improvements in the performance of the hardware and software, and the introduction of new user interfaces generally lead to increased productivity. For many disabled users, however, these improvements often introduce new barriers. The most pronounced example of this is the affect the Graphical User Interface (GUI) has had on blind users. For more than a decade, blind people have relied on screen reading programs which use synthesized speech to read text from the screens of computers. Unfortunately, existing screenreaders cannot cope with the GUI. As a consequence, blind computer users are severely disadvantaged through the growing popularity of GUIs.

There is another growing problem which affects all disabled computer users. The usual approach for providing access is adapt a particular computer to meet the specific of a particular user. Each type of disability requires a different access technique. Providing adequate access was relatively straight forward while each person used a single computer, and each computer was used by a single person. Now, however, we all use many different computers for home, study, work, and leisure activities. Furthermore, many of these computers are shared by different users. While it is impossible to make all computers accessible to all disabilities, it is possible to achieve the same result by handling access functions outside of the computer and providing a standardized interface. The Archimedes Project at Stanford is developing a "Total Access System" in which each disabled user has a personal "accessor" which performs all access functions. Each type of computer is equipped with a bidirectional infrared interface called a "Total Access Port" (TAP) which can communicate with any accessor. With this combination, any disabled person can communicate with any host computer, appliance, or electronic device which has a TAP.

This talk will describe the Total Access System and show how it solves problems such as those outlined above.