Matplotlib

Produced by the study pipeline itself.

Matplotlib.

One command from Rodata to charts.

Reproducible from the repository.

A diagram instantly tells an engineer if a bending moment looks physically correct based on the supports.

Matplotlib gives you that immediate visual feedback.

Okay, so we have our data organized, scaled, and visualized.

But abstract math and data arrays only take us so far.

Yeah, absolutely.

And I don't think just the fact that AI can help you design an experiment, of course, like write your Matplotlib code.

I mean, I haven't memorized 100% of Matplotlib.

It's probably like rare.

There's probably like five people in this world that... can do it from scratch.

That TypeScript part is the one that gets me.

I've been there with Matplotlib.

I googled the same five function signatures every single time, and if Claude disappeared tomorrow, I'd be in trouble for about a week, and then I'd be fine.

But Daniel's describing something deeper.

And there was a lot of confusion about how to, to do things like that.

In Python, there was, I think, a lot of confusion about, um, w- even, like, what Matplotlib was in relation to ggplot.

Uh, you know-

Yeah

When it has surprised me.

Um, you know what, when it has surprised me, so I don't remember, it was, like, uh, six or seven months ago, uh, an AI attacked a developer, a maintainer of a library called, uh, Matplotlib.

And what it did was it tried to get a repo, which it was trying to fix a piece of code that was simple, and the maintainer, he said, "No, you're a bot.

I'm not gonna ...

But initially when I saw-- This is, like, think, like, 2013 when data science was, like, the hot new thing.

It felt to me like just a weird way to do data exploration for people that were in love with Python libraries like Matplotlib and all that kind of stuff, and it made sense as a way to add some documentation and, and comments around code, but it didn't make a lot of sense to me at the time.

But now, um, it real- I see how it supports an iterative kind of development that's really, really nice because...

So for those of you, we've had a couple episodes on Notebooks, but if you haven't done much with them, especially with Python, it keeps track of the variable stuff.