Update 2020-10-05: My version (the upper one) was slightly modified. The previous state (referred to by the first three comments), is now frozen in screenshots (embedded, fullscreen). There is also a little “making-of” video here.
Inspired by CogDog & JRDingwall’s “kitchen“, I played with H5P interactive textbook elements, and tried to “cook” my own content type.
Contrast
The above version violates the doctrine of the “Split-Attention Effect” (because the explanations are not close to the items described; see here for more). So I also created a version using the original H5P content type of “Image Hotspot”, see below.
Which one do you like better?
This is definitely much better! I find a number of things on their layouts could use improving, but this makes a clear case for it. Not only the cognitive load to process, what about the repetition of feedback being in the same place? I would suspect learners might scan it more quickly because it looks the same.
What did it take to change? You should submit/share it with the H5P developers.
Many thanks Alan for your encouragement. I documented the changes on https://github.com/x28de/Hotspots/commits/master/H5P.GlossaryHotspots-1.0 — see the two commits of today Sep 28, 2020.
I also like the new version, but mostly because everything is bigger, so it’s much easier to see and read.
There’s a pull request that’s over a year old and that would allow to achieve the same (+some other things).