I've read all the comments here and still don't understand the rationale for hexagons as opposed to squares.<p>In every sense, squares seem to be much easier to reason about and easier to hierarchically partition than hexagons are.<p>There are certain advantages to hexagons in certain contexts, like six degrees of movement instead of four in board games, but I don't see how any of those advantages translate here for geographical indexing.<p>I'd love to understand why hexagons as opposed to squares in <i>this</i> context are a superior solution rather than unnecessary complexity?
I found this presentation helpful for an intro to H3's design/motivation: "Engineering Sub-City Geos for a Hyper-Local Marketplace with Uber": <a href="https://youtu.be/wDuKeUkNLkQ?si=-9JmxZQJ2LZo6Kh4" rel="nofollow">https://youtu.be/wDuKeUkNLkQ?si=-9JmxZQJ2LZo6Kh4</a>
This online tool gives you great idea about what this means on different levels:<p><a href="https://wolf-h3-viewer.glitch.me/" rel="nofollow">https://wolf-h3-viewer.glitch.me/</a>
Is the big thing here that those hexagons have the advantages of a circles (almost even max distance in all directions from center) with the advantages of squares (no overlap)?
This is like the advantage of using 6 mile hexes in a tabletop rpg map.[0]<p>Each hex can be divided into smaller hexes until you get to the level of feet/meters as opposed to miles/kilometers.<p>[0]: <a href="https://steamtunnel.blogspot.com/2009/12/in-praise-of-6-mile-hex.html?m=1" rel="nofollow">https://steamtunnel.blogspot.com/2009/12/in-praise-of-6-mile...</a>
There is a super high quality Rust lib that implements this: <a href="https://github.com/HydroniumLabs/h3o" rel="nofollow">https://github.com/HydroniumLabs/h3o</a>
Hexagons could be a new kind of political entity of the future, where, for example, rewards for climate progress (measured by satellites) could be put into practice.