Welcome to Datagonist

Juan

Juan

1/30/20263 min
0
datagonist

Writing and reading data stories is probably one of my favorite use cases around data. Transforming a collection of numbers and rows into a series of words that gives voice to that information is something I’ve found, honestly, fun and insightful.

One of the first times I did this was with a dataset from a game I’d been gathering myself. After each match, I’d write down the score of both teams and their composition, tracking by how many points the winning team won. Simple stuff. Once I had enough data, I loaded the dataset into RStudio and wrote a small script that produced the numbers I was looking for, along with a few charts.

"Done," I said. "Let me tell my friends about this."

Then I had a moment. Sharing a bunch of screenshots and charts wasn’t the best way to do it. It was messy, and it was missing context. So instead, I moved my script into what I believe was R Markdown—a document format that combines R code, its output, and user-written text—exported it, wrote a few sentences explaining my findings and shared that with them. I'm not clear if that was my first time sharing, what I'd call a data story, but for sure it was one of the first.

Datagonist is dedicated to data stories. It’s a space I’ve created to write and discover narratives where data is the protagonist. As such, the platform aims to provide both writers and readers with the tools they need to make their point. Features like inline charts and dataset previews embedded in stories are the very first steps I’ve taken toward that goal.

What’s missing are the stories. And that’s where you come in.

The stories Datagonist aims to host aren’t limited to a particular style, type, or topic. The only requirement is that there’s data, and a story that accompanies it. To keep this first post on theme, I want to share a few numbers describing the development process of the application itself.

This first version of Datagonist took me 163 code commits. For the non-tech-savvy reader, that means 163 times I added or changed code in the codebase. Of those, 26 were bugs I introduced during the process. My most common time to commit was in the evening, with 7 PM being the peak.

 

My Evening Commits

During testing, I wrote 17 different stories, left myself 37 comments, uploaded the same picture more than once, changed the font sizes four times, the tagline three times, and got frustrated twice. Once while deciding on a color palette—which, as you can see, I ended up abandoning—and once when I realized, almost right before launch, that the text editor had a very ugly bug.

Small facts like these, longer stories, and anything you’d like to share are the kinds of content I’m hoping to see on Datagonist.

Welcome. I hope you enjoy it.

— Juan

Dataset Preview

commits.csv5 rows2 columns0.0KBcsv
hourcommits
1934
2013
2124
229
232

Discussion

Start the conversation

Sign in to join the discussion

Loading comments...