Understanding where systems fail, and redesigning them through behavioral science and ethical innovation.
The Incident That Should Change Everything Last month, a routine Cloudflare configuration update took down a large portion of the internet. ChatGPT, Uber, government websites, and countless downstream services went dark. The cause was mundane: a Bot Management configuration file grew too large and triggered a latent failure mode. A month earlier, AWS us-east-1 suffered a cascading outage when…
I’m thrilled to announce that {bidux} v0.3.3 is now available on CRAN! This release is about meeting you where you are: Whether that’s knee-deep in a database connection, racing against a deadline with Quarto, or just needing a quick UX fix without the ceremony. Because behavioral insight shouldn’t require a behavioral scientist in the room. Your Database, Your Telemetry, Your Way Let’s be honest:…
A Crowd, A Crisis, and the Wrong Problem The crowds gathered fast. Flags, chants, phone cameras, all aimed squarely at the latest political flashpoint. The air pulsed with conviction. Something needed to be done now . On the surface, it looked like decisive action. Underneath, something quieter was happening: a narrowing of attention, a collapse of deliberation, and the slow suffocation of the…
I’m excited to announce that {bidux} v0.3.2 is now available on CRAN! This update is a polishing release that focuses on API ergonomics and package optimization . We’ve flattened out old complexities (goodbye nested lists!), added knobs to fine-tune telemetry with ease, and even given you a “mute button” for verbose output. All these changes maintain 100% backward compatibility , so you get…
A Cold Chicago Winter, a Warm San Diego Welcome My first rstudio::conf was 2018 in San Diego, late January. Coming from Chicago, that felt like a small miracle. I had about six or seven months of R and the {tidyverse} behind me, was still studying social psychology in grad school, and had only ever done academic conferences. Walking into this one felt different; bigger stage, crisper production,…
I’m excited to announce that bidux v0.3.1 is now available on CRAN. This release brings telemetry deeper into the Behavioral Insight Design (BID) framework, strengthens stage consistency, and smooths the migration path for users moving from earlier versions. A Modern Telemetry Workflow Telemetry data is increasingly central to BID workflows. Until now, telemetry ingestion produced list-like…
🚀 {bidux} v0.1.0 is live on CRAN! Users don’t just see your dashboard: They interpret it, navigate it, and act (or fail to act) based on it. That’s not just design; it’s cognition. Learn more below or explore the repo: github.com/jrwinget/bidux Why UX Is Too Often an Afterthought In data analytics and dashboard development, we often get the logic right: The data is clean, the calculations are…
Wouldn’t it be great to know the chances of being injured or having your car totaled in a car accident before ever being involved in the accident? If you live in a less densely populated area, this question might not be very interesting. But if you live in a big city, like Chicago, you might be a little more concerned about this. By identifying factors that predict these bad accidents, we might be…
Note: Twitter’s (now X) API policies have changed significantly since this post was written. Leaving this post up for historical purposes and conceptual understanding, but be aware that the code and methods described here may no longer work with the current API. Last week, I decided to kill an afternoon by creating a Twitter bot . Why? Mostly, I was procrastinating on revisions for a manuscript…
Chicago red light camera violations In this post, I walk through a simple exploratory data analysis of red light camera violations in Chicago. Data import Data downloaded from the Chicago Data Portal . Code library (tidyverse) library (modelr) (red_light_raw <- read_csv ( "2019-04-28_chi-red-light.csv" )) ## # A tibble: 466,107 × 10 ## INTERSECTION `CAMERA ID` ADDRESS `VIOLATION DATE` VIOLATIONS…
It’s been quite some time since I’ve written here, so I thought I would use one my of 2019 #rstats goals as an excuse to brush off the dust. In this post, I write about my first #tidytuesday submission of the Economist’s “TV’s golden age is real” data set (original #tidytuesday code here ). I also make a few improvements to some of the graphs and add tables with the gt package. Special thanks to…
Hello, world! Thanks for taking the time to read my first blog post! While I am trained as a social psychologist, I’m a big proponent of leveraging data science tools to understand the world around us. Thus, I plan to discuss a mixture of data-driven topics and insights here. I hope to use this blog as a place to write about things that interest me, but focus will likely be on statistics, research…