Let’s move this section to https://observablehq.com/@observablehq/plot-projection

Let’s have an example above the fold; see e.g. https://observablehq.com/@observablehq/plot-line which only has a sentence or two before you see a functional example. And the first few examples should visualize some abstract data (e.g., a choropleth of county unemployment or a dot map of earthquakes or covid cases) in addition to being a map.

This appears to be closer to Reading rather than London. 😅 https://www.google.com/maps/place/51%C2%B030'00.0%22N+1%C2%B000'00.0%22W Changed to 51.5° N, 0.13° W, Westminster. https://www.google.com/maps/place/51%C2%B030'00.0%22N+0%C2%B007'48.0%22W

I’m wondering whether we should be discussing Plot.dot + projection here, in the Plot.geo notebook, or over in the Plot: Projection notebook? My intent with that notebook is to cover how the projection option replaces the x and y scales, so it feels more suitable to discuss Plot.dot (and Plot.line, and whatever other marks end up supporting projections) over there.

(Although specifically in the case of Plot.dot, it’s nice to show it as an alternative to Point GeoJSON geometry.)

Is it worth mentioning that Plot expects spherical GeoJSON (like D3) rather than (now-standardized) planar GeoJSON when using one of our built-in spherical projections? Maybe that’s something that could be discussed in the appendix below or in the Plot: Projection notebook.

The Plot.geo(d3.range(10, 171, 10), might be worth adding a comment here or quick explainer sentence on what that does (at least it would help me 😅.

Would it help if I pulled radii out into a variable so you can see what it is?

Yeah I think that makes it really clear what's happening. Thanks!

funny we have a log scale, but not an exp scale

Could you explain the z:null option?

z:null explains to Plot.line that this is one line with a variable color—as opposed to a different line for each color.

Can I control the color of undefined values, say I have null values for unemployment...?

It looks like you can use color: { ... unknown: "#ccc" }

Yes, see the **unknown** option for the color scale. https://github.com/observablehq/plot#scale-options