The coronavirus can possibly infect a lot more people than there are those who can provide medical care. But if we slow the spread, and there are fewer people in need of care at the same time, the difference might be less overbearing. This version of the “flatten the curve” graphic by Alexander Radtke, first made by Rosamund Pearce for The Economist, illustrates the difference in animated form.

Vicki Boykis riffing off the George Box quote, “All models are wrong, some are useful.”:

The point is that, whatever data you dig into, at any given point in time, that looks solid on the surface, will be a complete mess underneath, plagued by undefined values, faulty studies, small sample problems, plagiarism, and all of the rest of the beautiful mess that is human life.

Just as all deep learning NLP models are really grad students reading phone books, if you dig deep enough, you’ll get to a place where your number is wrong or calculated differently than you’ve assumed.

I think of statistics as uncertainty management. It’s about estimates and figuring out how much you can trust them. Working with data is rarely about getting an exact truth.

Hannah Davis works with machine learning, which relies on an input dataset to build a model of the world. Davis was working with a model for a while before realizing the underlying data was flawed:

This led to a perspective that has informed all of my work since: a dataset is a worldview. It encompasses the worldview of the people who scrape and collect the data, whether they’re researchers, artists, or companies. It encompasses the worldview of the labelers, whether they labeled the data manually, unknowingly, or through a third party service like Mechanical Turk, which comes with its own demographic biases. It encompasses the worldview of the inherent taxonomies created by the organizers, which in many cases are corporations whose motives are directly incompatible with a high quality of life.

Speaking of responsible visualization, Datawrapper provides 17 charts and maps you can use in your stories, without causing unnecessary panic.

Below is an embedded example:

You can easily edit the charts, download the data, or export as an image file. Datawrapper also describes the sources behind the charts and tips on using them in your stories, making for a solid resource.

We’re seeing a lot of maps now about coronavirus. There are a lot counts, rates, and a little bit of panic involved. Kenneth Field provides guidance on mapping this data responsibly:

We’ll focus just on data for China in the following series of maps. They are designed to look like small maps you might find on a news media web site. Relatively simple, and with just the basic facts. Many of the issues I’ll note are equally applicable whether you’re making a small static map or an interactive web map.

These past few weeks, and especially this one, has been full of uncertainty. Probabilities, odds, and rates. In this issue of The Process we talk about ways to visualize this uncertainty.

The CDC recommends that you do not touch your face to minimize the spread of the coronavirus. We do this quite a bit without even thinking about it, so Do Not Touch Your Face uses machine learning to help you adjust. Train the algorithm, and then the algorithm trains you.

The New York Times has a rundown of what happened on Super Tuesday, with five takeaways. One of the sections uses a ternary plot to show support for Sanders, Biden, and other candidates:

In case you’re unfamiliar, a ternary plot shows the ratios of three variables as a single position. In this case, the variables are support for Sanders, Biden, and Other. More support moves a point closer to the respective corner of the triangle.

Davis Vilums set a goal to cycle every street in London on his way to work (without being late). After four years, he accomplished his goal:

I am a passionate cyclist, and I love the streets of London. Most of my travels are daily 25-minute rides to work. Over time my route became boring. I decided to make it a little bit more interesting by taking the parallel streets on my way there. I bought a map of central London and started to colour in the streets to mark the routes that I have taken. And then I got obsessed with it.

R turned 20 last week. The first public release of R was on February 29, 2000. Jozef Hajnala with the look back:

The power of R comes by no small part from the fact that it is easily extensible and the extensions are easily accessible using The Comprehensive R Archive Network, known to most simply as CRAN.

My initial reaction was, “Wow, 20 years. R is old.” And then I realized, “So R was only… three when I started using it… wait a minute.”

Andrew Wang-Hoyer built over 200 animations on and off over two years. They are satisfyingly hypnotic. They also only use SVG, CSS, and HTML, and you can get the code on GitHub.

I feel like sans-JavaScript is becoming a thing. Is this becoming a thing?

With Super Tuesday on the way, there’s still a lot of uncertainty for what’s going to happen. FiveThirtyEight has their forecast, but even with results expressed as odds and probabilities, the outcome almost seems static and concrete. So FiveThirtyEight has a different way of poking at their forecast. Pick the winners in each state, note how the conditional probabilities change as you go, and see what might happen in the rest of the primary given your picks.

For each month in her child’s first year, Amanda Makulec took a picture of her baby and a pizza. Each slice represents a month. Hence, pizza baby.

For many, Gmail automatically categorizes incoming emails to the primary inbox, promotions, and spam. The Markup and The Guardian tested the categorization on presidential candidate emails:

Their results:

I don’t use Gmail, and I don’t get any of these emails, but I’m curious how these candidate emails differ. Does Buttigieg write more personal messages whereas Sanders’ is more like an advertisement?

The Centers for Disease Control and Prevention made this graphic to show what beard styles work and do not work with a respirator. If there is hair in the way, the seal breaks. The CDC made it a couple of years ago for No-shave November, hence the playful tone, but with coronavirus concerns, it’s once again made relevant.

It reminds me of the trustworthiness of beards.

Also, wash your fingers.

Every month I collect useful tools and resources for visualization. Here’s the good stuff for February 2020.

From reddit user shoru_lannister, here is a pie chart of their living room corner.

Do we have another contender for best pie chart? I think the pyramid pie chart still has an edge.

While working on maps inspired by USGS maps from the 1800s, Sarah Bell made a typeface to match:

While making my own USGS-inspired maps, my search never returned the exact type of font I was looking for. The fruitless search was serendipitous however, because it provided the push to make my own. It was designed for map labels that are no larger than 80-100pt, but usually much smaller. I decided to name it BellTopo Sans with the plan to create a serif version.

She made it available here.

[arve url=”https://www.youtube.com/watch?v=bO9-dyYK3qQ” loop=”no” muted=”no” /]

We can download data as a single snapshot in a single file, but oftentimes that data is generated piece-by-piece. In the map above, NASA shows how they piece together rain data with a network of satellites:

The ten currently-flying satellites in the Global Precipitation Measurement Constellation provide unprecedented information about the rain and snow across the entire Earth. This visualization shows the constellation in action, taking precipitation measurements underneath the satellite orbits. As time progresses and the Earth’s surface is covered with measurements, the structure of the Earth’s preciptation becomes clearer, from the constant rainfall patterns along the Equator to the storm fronts in the mid-latitudes.

[via kottke]