Few years back when one of the developers asked me if screen reader detection on the browser a good idea, I counter questioned What can you do if you can make a screen reader detection? The answer is a straight a lot. Enhanced experience can be provided for screen reader users where ever necessary. After closely monitoring responsive web design, I could relate it far better how the content authors are detecting the device used and providing device specific experience to them. I have been convinced that screen reader detection can provide a better user experience.
But what the risk with screen reader detection?
There are many reasons why screen reader detection is a bad practice. Below are few of them I could think right now.
I am a visually challenged person. However I dont want to expose my disability to the world as much as I could. The reasons could be several hundreds. The approach is the same on digital content. I dont want the content authors to detect my disability through the use of screen reader. I am sure many other similar users are of same opinion as mine.
The term disability is broad so as accessibility. Many of the content authors who interpret disability differently may create barriers though they want to break them. I have observed many strangers talking in a loud voice identifying me as a visually challenged. They for a while may forget that I am visually challenged not hard of hearing.
The other problem I see is with updating the secondary content. Eg: In modern web, a search field in many cases does not contain a search button to submit search. In the process of providing enhanced experience, a search button can be provided only for screen reader users. If screen reader detection is a Possibility, developers may forget to update the screen reader only content.
The other concern I see is with various assistive technology/ browser / operating system combinations. Content authors might not be able to provide similar experience with all the combinations or at least highly reliable combinations. The content that is purposefully hidden to one screen reader may not behave similar to other screen reader. Eg: Funds transfer facility in my bank account is not available if I use with JAWS. I am able to transfer funds using NVDA. Being my banking provider one of the largest banking service providers in India, I am not sure if they purposefully made funds transfer not available for JAWS users, my experience is different between the two highly used screen readers in the market.
Certain content authors may think a step ahead and may provide text only experience for users if screen reader detection can be made. This brings out a clear discrimination between general users and users with disabilities. Screen reader users are not second grade citizens, do not provide secondary way of browsing. Also maintenance of two different websites for the same purpose is a bad idea.
Conclusion
Approximately 80% of the participants responded that they are comfortable with screen reader detection in the 5th screen reader user survey
by WebAIM. Approximately 86% participants responded that they are comfortable detecting the screen reader for better accessibility. I agree with Marco as mentioned in his article why screen reader detection on the web is a bad thing, It is due to the lack of knowledge and very little context provided by web aim respondents might agreed that they are comfortable detecting screen readers. I love to see more views by experienced accessibility consultants on the topic so that the respondents will have better understanding and context answering such questions in the future.
Being a pragmatic accessibility consultant for several years with the modern web technologies providing a good experience for every visitor irrespective of disability is not an impossible task. All that required is a careful thought right from wireframe, through-out the design and development process. Testing the website with end users especially those with disabilities will help in detecting real problems. It’s the content author’s responsibility to fix them and provide a comfortable seamless user experience to every user irrespective of disability. In the worst case if the content authors want to detect the screen reader usage provide a mechanism that collects the user consent before detection. As mentioned by Dylan Barrell in Assistive Technology Detection: It can be done right the mechanism should be as simple as website requests and user decides allowing the user to make a choice.