GraphGPT converts unstructured natural language into a knowledge graph. Pass in the synopsis of your favorite movie, a passage from a confusing Wikipedia page, or transcript from a video to generate a graph visualization of entities and their relationships.

Built with Ruby and powered by UnicodePlot

Very cool! Here’s the code for plotting a simple sin curve:

import { Mafs, CartesianCoordinates, FunctionGraph } from "mafs"

function HelloFx() {
  return (
    <Mafs>
      <CartesianCoordinates subdivisions={4} />
      <FunctionGraph.OfX y={(x) => Math.sin(x)} />
    </Mafs>
  )
}

R has many such packages. In Python, not so much. This one is designed to be easy to use while being customizable (some examples in the README).

Diagrams are faster and compact way for knowledge transfer. It’s much easier to convey system architecture with a diagram compared to writing an essay on it.

Diagrams are especially useful in providing high level overview of software design. Following are the most-widely used diagrams by engineering teams…

Minimalistic data structure sketches with links to the code examples on JavaScript

A collection of ASCII data structures, state machines, logical structures, and more.

People tend to be visual: we use pictures to understand problems. Mainstream programming languages, on the other hand, operate in an almost completely different kind of abstract space, leaving a big gap between programs and pictures. This piece is about pictures drawn using a text character set and then embedded in source code.

There’s a long history of timeline tools for the web, but most of them had you inserting the data into an XML file, a JSON object, or worse yet: the HTML itself. Markdown-style plain text to the rescue?

As files, datasets and configurations grow, it gets increasingly difficult to navigate them. There are however many tools out there, that can help you to be more productive when dealing with large JSON and YAML files, complicated regular expressions, confusing SQL database relationships, complex development environments and many others.

Simply run git log --full-history --date=short --pretty=format:"%ad,%an" > gitlog.csv and upload the resulting file. Here’s what changelog.com’s looks like (10+ commits only to keep it tight) 👇

I used scare quotes around “whole life” because those are his words and surely there’s a lot more to life than things you can quantify, but still: this is interesting

Back in 2019, I started collecting all kinds of metrics about my life. Every single day for the last 2.5 years I tracked over 100 different data types - ranging from fitness & nutrition to social life, computer usage and weather.

This data produces 44 graphs that are all shared publicly on the website.

A nice little web-based tool to help you quickly/visually grok some JSON data you’re working with. I tested it on some game data from an old JS Danger episode, and it handled it pretty well.

This is an excellent machine learning primer where you scroll the page and the author(s) walk you through the process of creating a model to distinguish homes in New York from homes in San Francisco.

Russel Goldenberg & Caitlyn Ralph from The Pudding join Amelia & Nick to talk about how they create data-driven, interactive articles, how the team works on both The Pudding’s data journalism articles and Polygraph’s client work. We also dive into how the team works with contractors and how the company manages itself using a Holocratic method.

The resulting maps are quite varied and absolutely stunning. 🤩

HN as tech trend? The results seem conclusive, but drop a comment below to share your thoughts on what James shared, his methodology, etc.

There’s a general belief in the tech circles I inhabit that Hacker News is a useful indicator for up-and-coming technologies that will hit the mainstream within the next few years. So I picked some of the major tech topics of the past fifteen years to see if that’s really true. Can I convince myself that checking the HN front page multiple times a day is a useful and productive exercise?

For the past year, my friend, Ole Kröger, and I have been developing a native animation package in Julia called, Javis.jl. Through our development process, we have been able to build a nearly 70 person developer community, sponsor Google summer of code students, and help new Julia programmers create powerful visuals! We recently presented the tool at JuliaCon, and were able to show its use for educational outreach and beyond.

Our hope is that this open source tool can be used by programmers, educators, professionals and researchers from across the globe to convey their ideas in winsome and understandable ways!

It’s fun seeing the proliferation of TODO comments over time on these bastions of open source. One not-surprising (but still unfortunate) trend: they all pretty much move up and to the right 📈, but a few have had some dramatic reversals 📉 at certain points in time. Go had a crazy month in April 2018 & TypeScript’s TODOs exploded in the Spring of 2018.

This comes from Mike Bostock (D3.js) who knows a thing or two about visualizing data in the browser.

It has a concise and (hopefully) memorable API to foster fluency — and plenty of examples to learn from and copy-paste.

April Fool’s may be over, but once we set up a system to react every time someone typed Command+C, we realized there was also an opportunity to learn about how people use our site. Here’s what we found.

TLDR; one in four users copy something within five minutes of hitting a page. But this blog post (and accompanying podcast episode) goes deep into the details and lays it all out for you with pretty charts.

Synthesized insights from Stack Overflow’s 2020 survey data:

The dataset has 33,447 salary data points which probably isn’t that many given that there are probably around 25 million software developers in the world. You have been warned.

Despite Petr’s warnings, he did go through some trouble to make the data as good as possible (short of, you know, finding or creating more data sources 😉).

Visualization help end-users understand data. Charts.css help frontend developers turn data into beautiful charts and graphs using simple CSS classes.

Utility classes are so hot right now.

Mike Bostock celebrates D3’s 10th by reflecting on what he’s learned over the years. There’s a lot to glean from Mike’s reflections. I really enjoyed this sentiment under the “Don’t go it alone” section:

To avoid entrusting your emotional wellbeing to internet randos (see #8), you must develop relationships with a small, stable group of people that you respect. In other words, find a team (or community) that can provide validation, feedback, support, and mentorship. Maybe this is obvious to everyone but me — yes, Mike, friends are good — but I feel like it’s worth repeating today when so much human interaction happens at a distance.

Superset can query data from any SQL-speaking datastore or data engine (e.g. Presto or Athena) that has a Python DB-API driver and a SQLAlchemy dialect.

This has been around long enough to be picked up by the Apache Foundation, but somehow it’s avoided my radar until today. The visualizations you can achieve with it are impressive, to say the least.

I love posts like these from startups/projects that share how they’re doing over time:

Excalidraw started as a way to procrastinate on January 1st, 2020, and ended up being a fully fledged whiteboard product only one year later! In this post, we’ll go over the most important features that made Excalidraw great at being a virtual whiteboard for sketching hand-drawn like diagrams.

They detail their open source tech stack, new features the team shipped last year, cool things people are doing with the tool, and more.

(The tool itself, btw, looks totally rad and is definitely something I’ll be toying with over the coming weeks.)