Braille Monitor               January 2024

(back) (contents) (next)

The Evolution of Overlays

by Curtis Chong

From the Editor: At our 2023 National Convention in July, one of the topics discussed was website accessibility and the role of what are commonly called overlays to make that technology more user-friendly for blind people and others with disabilities. After a presentation by accessiBe, Curtis Chong was asked to make this technical gobbledygook understandable to those of us who simply want to go to the World Wide Web and get the products and services we need. Because Curtis’s time was cut, what we are running here are the remarks he intended to give had time allowed.

Why are we talking about overlays and why, two years ago, did we adopt Resolution 2021-04, regarding the use of overlays to make websites accessible to the blind? What is an overlay anyway?

To begin with, since the pundits are telling me that the term “overlay” is the wrong way to describe the automation that is supposed to help websites to be more accessible, let me suggest the term, “accessibility widget.” So, from here on out, at least during my remarks, I will refer to what we used to call overlays as “accessibility widgets.”

The use of automated tools to address some of the more obvious accessibility issues on the web is not really new. To fix some web accessibility issues, website accessibility companies sometimes deployed a limited form of automation along with the accessibility fixes they installed on behalf of their paying customers. None of us knew or cared that this type of limited automation was being used since, after all, we were not prompted to enter a special screen reader mode or do anything else to use a particular website. It was not until about three years ago that people began to be told that any kind of screen reader mode was available for certain websites.

So what is an accessibility widget? Simply put, an accessibility widget is automation that is supposed to be smart enough to find and fix the coding on a website so that it meets the accessibility requirements in the law. Among other things, this technology is said to be able to:

Accessibility widgets can do a lot of other things for people with disabilities other than blindness. Since this is a convention of blind people, focusing on accessibility from the nonvisual access perspective would seem to be the appropriate approach for me to take here.

Accessibility widgets cannot:

Remember, this is only a partial list. Let’s just say that as things now stand today, artificial intelligence, as implemented in accessibility widgets, can fix some, but not all, of the nonvisual access issues which, unfortunately, we encounter too often on the Web. (Let me digress by pointing out that accessibility widgets do not seem to help us to deal with that pesky visual CAPTCHA.)

So, if the accessibility widgets of today cannot fix all of our nonvisual access issues on the Web, why, then, do companies that offer accessibility widgets boast so many paying customers? AudioEye says on its website that it is trusted by more than 95,000+ leading brands and happy customers. accessiBe says that it is trusted by 198,210 small businesses and industry leaders alike. UserWay, a company which styles itself to be “The Leading AI-Powered Web Accessibility Solution” urges us to join “over 1 million websites and millions of users with disabilities who trust UserWay for their digital accessibility needs.” I can only assume that a lot of small businesses and organizations imagine accessibility widgets to be a less expensive and more effective method to make their websites accessible. Since I have already established that accessibility widgets fix some, but not all, of our nonvisual access problems and since we know that there are literally millions of websites that we will never touch, there will be businesses and organizations out there who may never know that they have an accessibility problem—that is, unless or until blind people come along.

To those companies who develop and sell accessibility widgets, we, the blind, may not be the customers who actually pay you, but we are the ultimate beneficiaries if you get it right and the victims of what you do if you get it wrong. You have chosen to use automation which announces the availability of a screen reader mode on the websites you support. For the uninitiated screen reader user, this is a powerful encouragement to turn on this screen reader mode with the expectation that this will make the website more usable for someone running a program like JAWS or NVDA. If turning on the screen reader mode actually works, no harm, no foul. But if it doesn’t, you really need to give us a way to turn off your screen reading mode (if this is even possible) and to reach out to you for help. You have grown your business on the premise that accessibility widgets alone will help your paying customers to deploy more accessible websites. You really need to dispel this myth and do more to encourage your customers to do a lot more than simply implement a solution based on automation. You, the providers of accessibility widgets, need to engage with the blind community so that you can learn from real blind people how your automation works and where it breaks down. Hold listening sessions with real, living blind people. Find out what excites us about the Web and what causes us no end of frustration and annoyance.

What should we, the National Federation of the Blind, do? For one thing, we, the largest organization of the blind in this country, must use our power as a vehicle for collective action by the blind to work with the major players in the accessibility widget marketplace. Resolution 2021-04, passed two years ago at our convention, gives us a framework regarding accessibility widgets and the companies that deploy them. We can use those concepts and principles to engage with companies like accessiBe, UserWay, and others, and we can stand firm in our understanding that automation alone cannot fix all of our nonvisual access issues on the Web. We can learn how to recognize when a particular website is using an accessibility widget, and we can demand to have methods and procedures that we can share with others to reinforce the positive aspects of accessibility widgets while diminishing or eliminating those aspects of the accessibility widgets which stand in our way.

We can and must challenge claims that artificial intelligence and automation are the most effective way to address our nonvisual access issues. We need to continue encouraging organizations and companies to develop the knowledge and expertise they need so that accessibility is considered and built from the ground up in everything they do. We must insist that companies like AudioEye, UserWay, and accessiBe stop leading their paying customers to believe that their accessibility widgets alone are the only way that true accessibility can be achieved, and we should continue encouraging them to partner with us on a basis of mutual respect and understanding.

For those small businesses who cannot or will not establish accessibility teams, if your website is truly not usable by someone who is blind, then come up with a way that doesn’t involve your website where a person can do business with you.

Accessibility widgets are not going away. The companies selling this technology are simply too large, and their technology is simply too pervasive. I know that most people will groan whenever they hear “overlay” or “accessibility widget.” Groan if you will. But I know that if we put our minds to it, we can use our energy, experience, and understanding to move this in a direction which will work for us. Thank you.

(back) (contents) (next)

Media Share