ISSUE 328   ·   March 23, 2021

In the News

What Data Can't Do

"Whenever you try to force the real world to do something that can be counted, unintended consequences abound..."
New Yorker | Hannah Fry

Why the Pandemic Experts Failed

Every day for the past year, the COVID Tracking Project collected COVID-19 data for the United States. In this retrospective, two of the project's founders share what they discovered about U.S. healthcare data, along with key insights about data usage in general. This is a great article that shows how the system "largely worked as designed." And that's a problem.
The Atlantic

Your Face Is Not Your Own

When the U.S. Ninth Circuit Court of Appeals ruled in 2019 that it's legal to scrape data from public websites, it was hard to know what dangers might be enabled for emerging technologies. It's early days still but Clearview AI is an interesting — and scary — player to watch.
New York Times

Sponsored Link

A Modern Data Architecture, co-written with Sisu

In this guide, we’ll cover why modern data stacks scale better and when you should update your stack.

Tutorials, Projects & Opinions

Toward Better Data Culture From First Principles

Uber collects and processes petabytes of data everyday for things like route optimization, fraud detection, pricing, and experimentation. This post explores issues they encountered while scaling and how they successfully approached them using First Principles.
Uber Engineering

Effective Data Monitoring

If you're using alerts to monitor data quality issues, this post is for you. Here are ten tactics for reducing noise and optimizing your data alerts.
Anomalo | Jeremy Stanley

Experimentation at Tubi

Tubi's experimentation culture has been instrumental to the company's success. Over the past 3 years, the rate of experimentation at Tubi has increased 18x and one in three machine learning experiments in the last quarter showed KPI impact. This post explores how they do it.
Tubi Engineering Blog

How To Make Smarter Data-driven Decisions at Scale

Join this webinar to learn how fellow data and analytics leaders leverage centralized data through modernization and democratization initiatives. You’ll hear first-hand experiences using data as a strategic asset to drive smarter decisions – like when to launch new product offerings, where to automate manual processes, and how to earn customer trust.
// sponsored

Resources

New edition of Data Journalism Handbook

This just-released edition of the Data Journalism Handbook offers a diverse collection of perspectives on how data journalism is done around the world. In its 54 chapters, leading researchers and practitioners offer both practical insight and critical reflection. Follow the links to download an open-access version or to purchase it in print.
Liliana Bounegru

Data Visualization

Which color scale to use when visualizing data

When visualizing data, you’re almost always working with color. This first post in a new series offers a great overview of color scales, how they're important, and why they're not as clear-cut as you might think. This is an awesome series with lots of examples and references.
Chartable | Lisa Charlotte Rost

ggplot Wizardry: Tricks and Secrets for Beautiful Plots in R

This popular talk from the recent Outlier Conference includes a hands-on tutorial, slides, code, and a recording of the 20 minute presentation.
GitHub | OutlierConf2021

Sign up to get Data Elixir's  data science newsletter in your Inbox >>

Data Elixir is curated and maintained by Lon Riesberg. If you have questions or suggestions for the newsletter, just reply back to this email.