Tom Larkworthy

Jumpgate

Visualizer: Own Cell Renderer

Writable FileAttachments

Debugger: Notebook Dataflow Debugger (ndd)

Change History

How the Observable Runtime works

Tests

Module map

cellMap

Bidirectional Observable JS <=> Runtime Toolchain

lopecode-old

Atlas: Runtime Overview

lopecode

Instead of the binary blinking you may want to add a fade-out which imo reflects better that grey is just a state without recent changes.

Fabian Iwand

Its tricky coz the color flashing is via a function that has to draw from the color scale which is (correctly) a categorical. The Plot code is in a cell called "viz" if you want to look. I don't want to do surgery over the top of Plot, but I don't think there is a good way to do it because the tickX shares the color scale with the BarX that does the background rects.

I would like this multiple times, but I can't. well, I guess I could use my other test accounts…

Visnu Pitiyanuvath

I like this tool a lot too. An unexpected side effect of implementation was that it only displays rows if the cell emits an event, which means the chart is only focusing only on what is really changing (unlike the sidepanel or notebook visualizer). So it has this nice emergent property of fitting the problem at hand.
But yeah, developer tooling is a niche interest here, I hope it will be discovered by other means rather than raw like count. It got more action on Twitter and I have various plans for it in other notebooks. It can be used to illustrate some dataflow patterns I have been struggling to communicate so far (like the flowQueue https://observablehq.com/@tomlarkworthy/flow-queue). It would not have been possible for me to make this without Plot, which saved an incredible amount of work. I am quite excited about dataviz aided development.