Constructing charts and graphs

Jeffrey Heer, a computer science professor at the University of Washington, provides an overview of building charts for analysis and exploration. It’s an iterative process between acquisition, cleaning, integration, visualization, modeling, presentation, and dissemination. [via @albertocairo]

Tags: , ,

Making useless things

Simone Giertz, bringer of joy and self-described expert in shitty robots, makes machines that succeed in failing. In her TED talk, Giertz talks about her path from “useless” things to expert. It’s all the more relevant after she found out she has a brain tumor.

Giertz’ talk resonates a lot.

During the early years of FlowingData, when there was a comment section on every post, graphics I made would occasionally gain traction over the interwebs. In my own version of Godwin’s law, if a comment thread grew long enough, someone eventually would chime in: “Cool. Someone must have a lot of time on his hands.”

I was in graduate school at the time, with a dissertation staring me in the face, so I didn’t actually have much time. But I made time, because I didn’t know what I was doing, and that was fun for me.

I grasped on to the “cool” part of the comment and discarded the rest in my head. Someone liked something I made enough to tell me so! That turned out to be a great decision.

Tags: , , , ,

Data with the feels

Data can be fact and analytical. It can help you make objective decisions. Data can also evoke the feels, helping you understand and relate to something that used to be foreign. Lisa Charlotte Rost, currently a Knight-Mozilla OpenNews fellow at NPR, makes the case in this talk.

Tags: ,

R in the data journalism workflow at FiveThirtyEight

R has found its way into a good number of news groups who do data journalism. Andrew Flowers for FiveThirtyEight talks about how they use the statistical computing language throughout their workflow.

R is used in every step of the data journalism process: for cleaning and processing data, for exploratory graphing and statistical analysis, for models deploying in real time as and to create publishable data visualizations. We write R code to underpin several of our popular interactives, as well, like the Facebook Primary and our historical Elo ratings of NBA and NFL teams. Heck, we’ve even styled a custom ggplot2 theme. We even use R code on long-term investigative projects.

Tags: , ,

OpenVis Conf talks all online

OpenVisConf talks

If you’re looking for a knowledge bomb during your lunch breaks, the OpenVis Conf talks from this year are all online. Naturally, you can sift through the talks with a visual interface that gives you a good idea of what each talk is about before you get into it. Nice.

Tags: ,

Working with R at the New York Times

Amanda Cox

Amanda Cox from the New York Times was on the Data Stories podcast. You should listen. She talks about how she uses R, workflow at the New York Times, and some of her favorite projects.

I listened while picking up my son from daycare. I hope some of it seeps into his consciousness through osmosis.

One note. In the beginning Amanda talks a little bit about how she got started. She was a statistics graduate student getting tired of the theory side of things. Her program didn't look at a ton of data in the first year, which led her to the New York Times, a placed aimed at practicality.

However, no surprise, it varies a lot by program. For example, the UCLA and Berkeley statistics departments get you looking at data early on. I haven't taken a course in years and am far removed from academics, but I only imagine it's more true with the whole data science field evolving into a real thing.

Tags: , , ,

R is quirky, flawed, and an enormous success

Most people who use R on the regular learned the language in the context of a subject outside of programming. They learned R as they learned statistical methods, or they picked up bits of R as they learned about visualization. However, if you learn R purely as just a language — without the domain-specificity — or you already program in a different language, R might seem strange at times.

In this talk, John D. Cook explains some of the "quirks" in R and why, maybe, they're not so strange.

I picked up R after three semesters of computer science in college, and that little bit of background actually seemed to make learning R a lot easier for me. So maybe the key is to be a horrible programmer at first (I was and still am in some ways), and R won't seem so quirky. [via Revolutions]

Tags: ,

Power of the reveal

Hannah Fairfield, who does graphics at the New York Times, talks about using visualization to show specific narratives. Something more than just "here's some data."

Tags: ,

Why exploring big data is hard

The talks from OpenVisConf 2015 went up, so I'm slowly making my way through. In this one Danyel Fisher from Microsoft Research talks about the challenges of working with data that doesn't quite fit into your standard CSV data model. The visualization has to account for the mess.

Tags: ,

Future of visualization

Jeffrey Heer, computer science professor and co-founder of Trifacta, describes the future of visualization in a short 10-minute talk. It's one where people aren't taken out of the analysis loop, but computers can provide a bit more help than they do now.

Tags: ,