Sorry Shivangi, I still have the “Automation for Jira” email about the old PR for this sitting in my inbox as a reminder that I needed to go back and retest this. I’m just now seeing your last note that you are still working on this. I will hold off on any further testing until you tell me your fixes are ready to be retested/verified. Thanks!

Was playing around with this again tonight, and the “fill-in-the-blank” questions are exhibiting the same bahavior as before.

Hello Chris, I am still testing the things that you mentioned and will let you know once I have made and pushed the new changes. I am still working on the resolution part as well, thank you so much for the example - it made things simpler.

Thought enough time had passed since the PR went in last week to retest this issue on 25.X (Maria DB branch) using JAWS+Chrome, but it doesn’t look like the changes have gone into effect yet??

As the Student is taking the Assessment, and is using their screen reader to read the text of the question, it is still saying “fill in the blank, read surrounding text” after each blank, so the sample question I created with 3 blanks is being read by JAWS as:

Roses are (blank 1, fill in the blank, read surrounding text), violets are (blank 2, fill in the blank, read surrounding text), and pumpkins are (blank 3, fill in the blank, read surrounding text) which is the very best of the autumn colors.

Note, this is the experience while using tabbed navigation, using the arrow keys is actually worse, because JAWS will read the text that proceeds the first blank, say “blank 1, fill in the blank, read surrounding text” multiple times, before moving on to read the next block of text and repeat the process for the other blanks, so you end up hearing the “fill in the blank, read surrounding text” part numerous times while navigating through the text of the question.   

There’s a good chance that Shivangi’s fixes haven’t gone in yet, so I will give it a few more days and retest later this week.

Hi Shivangi,, thanks for your work on this!

I watched your demo video, and while it definitely looks cleaner, its hard to fully assess the interoperability of the screen reader with the reworked fill-in-the-blank question without knowing what specific keystrokes you are pressing each time, if that makes sense?

For example, its not clear to me whether you are using the tab key and/or arrow keys to move from the first part of the question, to the first blank/editable text field, to the next part of the question, to the 2nd blank/editable text field, and so on.

Also, I’d be curious to see how the screen reader would read the entire question when using the JAWS (insert+down arrow) “read all” command, as well as what the interaction looks (sounds) like when you start filling in the blanks with the screen reader.

I believe you submitted a PR already, so once these changes make their way into master, I can test this out for you and let you know if I have any other suggestions.

Thanks again!