By Vinton G. Cerf, VP & Chief Internet Evangelist
The computer is one of the most versatile instruments ever invented. Its functionality is
limited only by our capacity to program and our ability to imagine possibilities. Despite this, it
has proven to be notoriously hard to reliably deliver applications that adapt well to the broad range of accommodations needed to enable access for people with disabilities.
The Variety in Disabilities
Disabilities come in seemingly infinite variation. We should not be surprised at this, since the human body is complex. Some disabilities are temporary (e.g. a broken wrist) and some are chronic and persistent (e.g. deafness, blindness, motor and cognitive impairments). No matter the case, people often need various types of tools to make computer-based applications usable and accessible.
I have progressive neural hearing loss, so I am very dependent on binaural hearing aids to function on a daily basis. My wife has two cochlear implants – a technology that still leaves me astonished that it works. I am reliant on captioning for television, recorded videos, and video conference calls. More generally, my disability has made me conscious of the value of technical responses to assist people with disabilities.
Accessibility Issues Persist
We know a lot about various kinds of interventions to improve accessibility. Magnification of images and text, maximization of contrast and font style, “screen readers” to describe orally what is on a computer screen, speech transcription for people who are deaf or hard of hearing, head-mounts for pointing when arms and hands are not reliable, the list goes on.
Some of these features are built into computer, pad, and mobile operating systems and can be activated to provide the needed accommodation. It is not always easy to know how to do this easily. Perhaps, more importantly, the designers of computer-based applications may not have experience or intuition in the use of accommodation tools to make design decisions that achieve the desired usability level.
Unless the programmer (or advisor) is an experienced user of screen readers, he or she may
not know how to make a web page maximally accessible with available screen readers.
Users may not know about configuration options that would make a device more readily
useful.
Even when we write detailed specifications for color combinations, contrast, font
sizes, and accommodation for magnification (e.g. does the web page still render in a useful way when magnified?), it may not be the case that a programmer will have the experience and skill to transform recommendations and standards into accessibly-crafted applications.
It is here that tools for the crafting of accessible web pages with templates and examples of
use might make a difference. Just as the creation of web pages was once done by manually writing HTML-encoded web pages and is now often done with convenient tools for composition and layout, one wishes that there were composition tools that naturally produce accessible web pages.
There exist testing tools, guidelines, and web-page composition editors but they often provide only limited guidance or assistance. Programmers seeking to produce accessible applications need to have a lot of experience using mechanisms, standards, and accommodation tools to achieve the desired result.
What Can Be Done?
Without a doubt, there are some tools aimed at solving this problem but more are needed.
We need more training and examples of good solutions that would improve tool
makers’ ability to help programmers produce accessible digital objects and services.
Investing in understanding what makes an application accessible is another area for serious
research.
Computer professionals need to spend time with people who rely on assistive
software and hardware to gain deeper intuition about accessible/usable design. Serious
research is needed to develop a deeper understanding of perception and how steps can be
taken to design for accessibility.
Standards are needed to allow users to express their configuration requirements so that various operating systems can easily support accessible applications. Application programming interfaces and libraries are needed to support configuration for accessibility. Training for accessible design should be a requirement for programmers intending to create applications for users with disabilities.
I believe that many of the ingredients we need are already present, including a few people
with extraordinary backgrounds and experience in designing for accessibility. We need to
distill this knowledge and incorporate it into reality available applications intended to help
programmers produce accessibility by design.
While by no means trivial, there is considerable experience in the computer science and engineering communities, but the tools needed to achieve reliable accessibility continue to be elusive. None of these ideas are new. They just need to be applied with increased determination and with codified practices.
These are exciting times for some accessibility capabilities. Speech-to-text, text-to-speech,
speech recognition and understanding, language translation, speech re-rendering for
improved acuity, optical character recognition (e.g. to translate menus), machine learning
tools to improve interaction with voice, video, and text for people with disabilities are all examples of technology developed to aid accessibility. We can expect more to follow.
