Image Hotspot Problems - Current State, Issues, and Opportunities

Currently, Open edX has an extremely old and under-maintained image mapped input problem that is in no way accessible. It has no graphical authoring interface for the selection of the hotspots, requiring staff to input coordinates in XML to create the activity. It is in no way fit for purpose or even comparable to other tools designed for this activity (which also have massive flaws), and at best it serves as an extension point for building a new activity.

<problem>

 <p>In the following image, click inside the pentagon.</p>

 <imageresponse>

   <imageinput src="/static/imageresponse_irregularregions.jpg" width="600"   height="204" regions="[[219,86], [305,192], [305,381], [139,381], [139,192]]" alt ="A series of 10 shapes including a circle, triangle, trapezoid, pentagon, star, and octagon" />

 </imageresponse>

</problem>

Current Gaps & Potential Improvements

Studio needs a UI for defining hotspots and alternative text for hotspots
Honestly this just needs rebuilding if we are to include it in core. Too much needs to change.

Additional Opportunities

I’m unable to find another learning platform that has an actually accessible version of this activity. I know I’ve harped on about this in various documents, but this is a huge issue for any educational platform or institution that wants to offer this kind of activity because they simply cannot use this activity without knowingly or unknowingly breaking regulations on accessible learning content.

There is an opportunity to not only figure out how to make this activity at least marginally accessible, but also to be the first major platform to offer this as a model, establishing Open edX as the leading accessible learning platform.

My initial idea for how we do this would be to consult an academic accessibility expert that can describe what a suitable alternative accessible activity would be for this kind of exercise. If we can make the activity itself accessible for sighted users with accessibility needs, and ensure that the activity effectively presents itself as an appropriate alternative activity to visually impaired users (for example, if a screen reader encounters a hotspot activity and effectively interprets it as a multiple choice question, provided suitable alternative text is provided), we might be able to have the best of both worlds and serve all kinds of users.