New notebooks in Quickstart

Hello all,

Since folks on the forum seem likely to have previously hidden the Quickstart section, I thought I’d let you know that some of the things we’re linking to there are new:

  1. The new D3 Gallery that @mbostock put together in the last couple weeks
  2. A Taste Of Observable, a concise demonstration for new users of what Observable can do
  3. Observable for Jupyter Users, a series of notebooks introducing Observable in a way we hope Python users familiar with Jupyter will find useful.

The Quickstart section has also been added to the unified homepage. Just as before, you can click the × close button to hide this section and use the horizontal ⋯ menu on your personal home page to bring the it back.

Please let us know what you think of these new resources! Are they helpful to you? Do they help you explain Observable to someone else?


I’ll try to convince some of my former colleagues at Karolinska (who have fully embraced SciPy and Jupyter) to give the third link a shot :wink: (don’t hold your breath though, I’ve been trying to get them into Julia for ages too and that doesn’t seem to stick either)

The second notebook in that series shows embedding an Observable visualization in Jupyter; hopefully this is both a bridging use case that brings people on board and a helpful example for people building visualizations for others to use!

1 Like

By the way, I think the “Observable is reactive” section in that notebook would benefit from warning people about a gotcha that seems to hit every new Observable user, which is writing cells that are too large and depend on too many other cells. See for example this question I answered a while ago:

It took me a while myself to figure out that refactoring cells into smaller cells is an important habit to have to avoid excessive recomputation or “resetting” of cells.


Great point. I mention that in the previous paragraph,

Because every cell has just one return value, you should break up big cells into little, focused ones.

but I don’t hit quite the reasoning you’ve put forward here which I think is on point.


Thanks for the notebooks, the Observable for Jupyter Users series is excellent! Our group is building a Jupyter Widget (Clustergrammer2) and we’re excited about seeing how much of this workflow we can port over to Observable with the underlying JavaScript/WegGL library Clustergrammer-GL library!

Cool! Let me know if you run into any roadblocks, I’m thinking about ways to improve this embedding API, for Jupyter in particular.

1 Like