Screen Reader Testing, but No Peeking!

My Experience with Jacob Wood's Hands-On, Sight Off: Working with NVDA.

I would like to begin by admitting there was some peeking.

Now that I have my confession out of the way, let's talk about screen reader testing! I am usually working on macOS and using VoiceOver for basic testing, but I also use Parallels to run Windows for NVDA testing. I was less comfortable with NVDA, partially because I had to work out some different key strokes using my Mac keyboard, so I jumped at the chance to take Jacob Wood's Hands-On, Sight off: Working with NVDA course.

A quick disclaimer: I received access to this course at no cost when Jacob offered it to a limited number of people during its initial rollout. I was not asked to review the course or share anything about it, and all opinions here are entirely my own.

Each module presents a different type of interaction or content, from flip cards and hotspots to tables and columns, and so much in between. There is an introduction to the content, followed by an inaccessible version with an explanation about why it's broken and references specific WCAG compliance failures. Then the accessible version can be revealed, along with an explanation of what was fixed, why it works, and how it meets WCAG Success Criteria.

WCAG Success Criteria give accessibility professionals all over the world a common language and standards to reference. But knowing which Success Criterion something passes or fails doesn't necessarily tell the whole story. This course connects those standards to what actually happens when someone encounters the content with a screen reader. It doesn't just demonstrate that something is inaccessible. It provides the opportunity to experience why it matters.

This is where the "sight off" part comes in.

I already use screen readers for accessibility testing, but I am usually also looking at the screen. This course provided me the opportunity to notice things I may miss when looking at the screen while testing with a screen reader. By focusing on the experience alone with "sight off" first, I went through testing through a bit of a different metaphorical lens. I may recognize that something is inaccessible and understand why, but that's different than trying to navigate it without using the visual information on the screen to fill in the gaps.

I did not work through this all in one sitting. If you've read my most recent post, you know I've been up to a lot over this past summer. I completed it a bit at a time and kept coming back to move on to the next module.

And I really did try to work through each module as intended: without looking.

Because this course was designed with such thoughtful care for not only the subject matter but also the overall experience, I was able to do that. If I was unable to tell where I was in the module, I could always find the "Reveal Accessible Version" button and experience the accessible version. Then I could go back through while looking to figure out exactly what had been happening in the inaccessible version.

Did I occasionally find myself peeking when I got completely turned around? Yes. (See confession at the beginning of the post). But I reminded myself that this was all a part of the experience, and I could always rely on that "Reveal Accessible Version" button to find my way again.

I initially completed the course with NVDA, then went back and tried some modules with VoiceOver. I don't have any sweeping conclusions to make about the similarities or differences between these two screen readers from this experiment. The accessible versions were accessible and easy to navigate with both. The inaccessible versions were inaccessible with both, although VoiceOver did give me some more... unpredictable outcomes than NVDA.

I recorded a video demonstrating one section of the Flip Card module, first using NVDA for the Inaccessible Example then the Accessible Example, and then did the same with Voiceover. I chose not to cut out some of my fumbling with the keys. I am not a native screen reader user, and sometimes I fumble, especially when switching quickly between NVDA and VoiceOver, and I think that's worth showing.

There is absolutely a time and place for polished demonstrations of someone working perfectly through an interaction. But accessibility testing also involves trial and error, experimenting, wrong keystrokes, figuring out where you ended up, and trying again. I didn't want to edit that reality out of the demonstration.

Watch my NVDA and VoiceOver Flip Card demonstration (Link opens YouTube in a new tab).

If you're interested in this course but new to screen readers, don't worry! You can find the information and resources you need on Jacob Wood's website, where you can also find the Hands-On, Sight Off: Working with NVDA course. (And it's super affordable!)

I've kept the course bookmarked because I know I'm going to come back to it while screen reader testing. The "sight off" experience was incredibly valuable on its own, but so are the explanations that accompany each example. Being able to revisit not only what is accessible or inaccessible, but why, makes this much more than a course I completed and moved on from. It's a resource I expect to keep using.