I think the idea can work, however the execution is still far from perfect.<p>Just on the binary search for example, the "balance" representation is re-used to signify the concept of "comparison", however the tokens put in balance are both of the same size, with only an image to differentiate them. So the link is not implicitly made from their size, weight, and the intrinsic feature which makes the one searched token the right one. This would be confusing for anyone trying to understand the algorithm.<p>Then, you got a first depiction of the problem, with each object being ordered by size. So the initial assumption is that the choice element would be on their size. Afterward, the target token has a star on it, and the generic, incorrect one, a question mark.<p>That could work, however, they both have the exact same size.<p>This is only an example of how the concepts used to explain and signify things visually are only partially followed-through. This greatly impairs the impact of the explanation.<p>And this is only on the binary search, arguably the simplest algorithm depicted here.