Between 1802 and 1817, James Sowerby cataloged and illustrated 718 minerals across seven volumes. Nicholas Rougeux restored all of the illustrations over several months, carefully arranged them by color, and made them browsable on a page. The result: British & Exotic Mineralogy.

Read about the slow process here. Also in poster form.

Saturday Morning Breakfast Cereal can see the future.

Here’s the good stuff for July.

Dana Goldstein, with illustrations by Yuliya Parshina-Kottas, imagines what school might look like if students go back. Face shields, distancing, masks, and pods.

I’m having trouble imagining any of this working in practice, especially with the young ones.

Watch enough YouTube, and you end up in a bubble of videos catered to everything you like and believe in. TheirTube, by Tomo Kihara and Polina Alexeenko and funded by the Mozilla Foundation, imagines the point of view of six personas:

Each of these TheirTube personas is informed by interviews with real YouTube users who experienced similar recommendation bubbles. Six YouTube accounts were created in order to simulate the interviewees’ experiences. These accounts subscribe to the channels that the interviewees followed, and watches videos from these channels to reproduce a similar viewing history and a recommendation bubble. Everyday, TheirTube retrieves the recommendations that shows up on their Youtube home page.

In case you’re wondering what my YouTube homepage looks like — and I know you are — just watch every J. Kenji López-Alt upload, and you’ll be just about there.

You’ve probably seen the videos. ProPublica is tracking to see what happens after:

ProPublica wanted to find out what happens after these moments are caught on tape. We culled hundreds of videos to find those with the clearest examples of officers apparently using a disproportionate level of force against protesters and reached out to 40 law enforcement agencies about the 68 incidents below. For each incident, we inquired about any disciplinary action, investigations and whether the department would disclose the officer or officers involved. While some departments provided details or relevant public records, others leaned on state laws to withhold information.

See also ProPublica’s recent release of NYPD civilian complaints against police officers.

From Google’s People + AI Research team, David Weinberger and Yannick Assogba build a machine learning system that plays Tic-Tac-Toe. They discuss the choices, not just the technical ones, along the way in the ten-part podcast series:

A writer and a software engineer engage in an extended conversation as they take a hands-on approach to exploring how machine learning systems get made and the human choices that shape them. Along the way they build competing tic-tac-toe agents and pit them against each other in a dramatic showdown!

This is a podcast for anyone, from curious non-techies to developers dabbling in machine learning, interested in peeking under the hood at how people make and shape ML systems.

I’m a few episodes in. It’s entertaining.

This is an especially good listen if you’re interested in machine learning, but aren’t quite sure about how it works beyond a bunch of data going into a black box.

For The New York Times, Jonathan Corum illustrated the dozens of spacecraft orbiting planets and objects in the Solar System. The piece starts at the sun and then makes it way towards interstellar space. Showing active and inactive spacecraft, it’s part history lesson and part cute animation.

A $600 per week benefit expires for the unemployed at the end of July. Congress is still deciding what to do after. For The New York Times, Ella Koeze highlights the percentage of usual income the unemployed will receive as a function of annual earnings and weekly benefit amount if the benefit goes away.

Each dot represents a state. The percentage ranges in the background provide a point of reference for where each income group and state falls. The scrollytelling format starts with individual points and then tours through the shifts.

An aside: I might just be imagining things, but I feel like there’s been more scrollytelling lately. Is this a function of doomscrolling on our phones? Also, whenever a chart anchors in the background and a text frame scrolls over, I think of Snow Fall from 2012.

In search of songs that define music in the 1990s, Matt Daniels and Ilia Blinderman for The Pudding look for songs that that Gen Z still recognizes. Also, the songs that are mostly foreign to the younger generation:

In 1999, “Wild Wild West” was the song of the summer. Yet it is fading far faster than any other ’90s hit with comparable starting popularity. Twenty years ago, it was inescapable. Maybe Millennials are still too sick of it, even for nostalgia rotation. Perhaps it wasn’t even that great of a song to begin with, artificially inflated by Smith’s celebrity and cross-promotion with the film Wild Wild West.

Sorry, Will Smith.

The results are based on data gathered by The Pudding in an interactive survey.

Studies suggest that wide adoption of masks can reduce the spread of the coronavirus. A meta-analysis by Ali Mokdad and his research group at the Institute for Health Metrics and Evaluation estimates at least a 30% reduction and up to 50%, which can lead to a big difference, as illustrated by Connie Jin for NPR:

Wear the mask.

The numbers are high here in the United States, and at this point, they look bad on their own. But compare it to other countries that are currently hit hard, the U.S. looks even worse. For The New York Times, Lauren Leatherby makes the comparisons.

Pierre Ripoll provides several ways to visualize periodicity using animation. Moving dots, rotating spheres, concentric circles, oh my. He uses D3.js and it’s an Observable notebook, so you can see what’s going on under the hood.

This week, people were taking a closer look at the Georgia Department of Public Health’s Covid-19 status page (again), which led to an unnecessary pile-on.

With lockdown orders arounds the world, places that we’re allowed to go sound different. The MIT Senseable City Lab looked at this shift in audio footprint through the lens of public parks:

Using machine learning techniques, we analyze the audio from walks taken in key parks around the world to recognize changes in sounds like human voices, emergency sirens, street music, sounds of nature (i.e., bird song, insects), dogs barking, and ambient city noise. We extracted audio files from YouTube videos of park walks from previous years, and compared them with walks recorded by volunteers along the same path during the COVID-19 pandemic. The analysis suggests an overall increase in birdsong and a decrease in city sounds, such as cars driving by, or construction work. The interactive visualization proposed in Sonic Cities allows users to explore and experience the changing soundscapes of urban parks.

The 3-D view shown above is visually interesting, but the top-down view is the easiest to read, looking like a stacked area chart over a map.

At distinct points on the mapped paths, a gradient line represents the distribution of quieter and louder sounds. Louder sounds appear to take up more space during the pandemic.

It’s hard to say how accurate the sound classification is through this view, but as I poked around, it seemed a bit rough. For example, the chart for Central Park in New York shows bird sounds making about 0% of the footprint, but you can hear birds pretty easily in the audio clips. I’d also be interested in how they normalized between YouTube clips and their own recorded audio to get a fair comparison.

Nevertheless, it’s an interesting experiment both statistically and visually. Worth a look.

Some industries are more compatible with remote work than others. Jonathan I. Dingel and Brent Neiman at the University of Chicago estimated the scale of the differences. For Reuters, Sarah Slobin reports using a variable width bar chart to show likelihood of close contact with others against likelihood of in person work:

Professional, management and technology jobs run the gamut from accountants and architects to lawyers, insurance underwriters and web developers. This group is much more likely to retain the privilege of collecting a paycheck while working remotely, and is based in major metropolitan areas, like New York and Los Angeles.

See additional breakdowns by geography and job loss.

For ProPublica, Caroline Chen, with graphics by Ash Ngu, provides a guide on how to understand Covid-19 statistics. The guide offers advice on interpreting daily changes, spotting patterns over longer time frames, and finding trusted sources.

Most importantly:

Even if the data is imperfect, when you zoom out enough, you can see the following trends pretty clearly. Since the middle of June, daily cases and hospitalizations have been rising in tandem. Since the beginning of July, daily deaths have also stopped falling (remember, they lag cases) and reversed course.

I fear that our eyes have glazed over with so many numbers being thrown around, that we’ve forgotten this: Every day, hundreds of Americans are dying from COVID-19. Some days, the number of recorded deaths has reached more than 1,000. Yes, the number recorded every day is not absolutely precise — that’s impossible — but the order of magnitude can’t be lost on us. It’s hundreds a day.

Cherrypicking statistics is at an all-time high. Don’t fall for it.

Unemployment has hit the United States hard over the past several months, for some harder than others. Lena Groeger reporting for ProPublica:

Part of the reason for this disparity is that many workers of color, especially Black workers, didn’t come into the crisis on equal footing. At the beginning of 2020, when the U.S. was at what most would have considered peak economic prosperity, the unemployment rate for Black workers was more than double that of their white counterparts. “The classic fact about Black unemployment,” said William Darity Jr., an economist at Duke University who studies racial inequality, “is that it’s been two times the white rate since we started measuring it.”

Each line represents a different subpopulation, so you can scroll over specific lines or select specific groups with the buttons. It’s similar to this New York Times interactive from 2009.

But this is 2020, so Groeger uses the overview as the initial view and then it shifts into scrollytelling. Groups highlight and the time frame expands as you read. This eventually takes you back to the initial view, where you’re invited to explore the data.

The overview first provides an opportunity for the reader to set a baseline as it relates to their own demographic. Then you see how specific groups are different or similar to that baseline. At the end, with a different baseline set, you can compare one more time.