Mike Bostock

Also AM4 described here in the DashQL paper: https://arxiv.org/pdf/2306.03714.pdf

I expect it will be difficult to do this in a backwards-compatible way, and it will probably cause a lot of churn with the CSS. I recommend we consider more...

Okay. I can take a look at #1669 when I have time but it’s not yet done and I see some hints of difficulties (do the container media queries still...

Since I haven’t had a chance to test yet, perhaps someone would be kind enough to explain how the container queries still work given that `container-type: inline-size;` is removed? The...

It disables margin collapse.

> I don't think we even need a WeakMap, as we could just do a quick check through the DOM with querySelector Why do you think that (going through the...

> Using the DOM feels simpler [information-wise], in the sense that we wouldn't need to track the association of type of clip and clipPath in parallel, as we could just...

The way I think about the WeakMap is that there’s a `context.frameClip` and `context.projectionClip` property that are lazily set in `getFrameClip` and `getProjectionClip`. The only thing the WeakMap provides is...

Yes, we’re aware it works like this. Not sure it’s fixable given how the underlying coordinate system is defined (band scale), but we might be able to do something.

Not sure how we would include parameterized pages, unless we used the **dynamicPaths** option or link crawling #1627 to find the parameter values? But we definitely could include non-parameterized page...