This is a famous picture by the artist Imperial Boy (帝国少年), who works in the anime industry. I sometimes claim that the entire genre of solarpunk is simply a riff on this picture.
If it’s not just “trees on buildings”, where does the Imperial Boy picture get its magic? Looking at it carefully and trying to analyze what I like about it, I think that much of it is about architecture, and even more of it is about the use of urban space — about how the structures in the picture shape the kinds of things you’d do if you were there. For example, here are five things I like:
The bipartisan deal contains a pot of money to repair America’s roads and bridges, and build a few more besides. This is the way we usually do infrastructure in America. First we build a ton of roads and bridges that are highly expensive to maintain, especially with our ruinously high construction costs (see this recent article by Jerusalem Demsas). Then, because costs are so high, we wait for a long time to repair the roads and bridges, until civil engineers start screeching, roads get potholed, and there’s a bridge collapse or two. Then we muster up the political will to throw the requisite shit-ton of money at the problem, the potholes and weak bridges get repaired for twice the amount it would have cost had we done it on a regular schedule and three times the amount it would cost if we were a normal rich country. And the whole cycle begins again.
How would you change this structure so that you could put a masonry brick on top of it without crushing the figurine, bearing in mind that each block added costs 10 cents? If you are like most participants in a study reported by Adams et al. in Nature, you would add pillars to better support the roof. But a simpler (and cheaper) solution would be to remove the existing pillar, and let the roof simply rest on the base.
A series of problem-solving experiments reveal that people are more likely to consider solutions that add features than solutions that remove them, even when removing features is more efficient.