I haven't read the post yet but right away I noticed how well the information is presented. I think it's because the post has a tree-like structure - I can take a quick look at the titles and the images underneath them and have an idea about both approaches, then I can read about the details. This way it's much more interesting than if there were no images. It's also aesthetically pleasing. I also love the combination of text, code snippets, diagrams, tables and charts - this way I can look at the issue from multiple perspectives. So cool.
I wonder if framing the problem "correctly" and applying domain-specific transformations (as you mentioned are out of scope for this post) would present a synergistic response, that would otherwise be lost in poorly framed solutions?
Thank you!
I would guess that certain transformations can be efficiently exploited by certain framings, but I have no intuition as to whether some transformations are dominant regardless of the framing. That's a very interesting question!
