348: in other news

London Underground PM2.5
This is "an interactive map of the London Underground transportation network. The stations are coloured based upon their measured PM2.5 value."
Data available (see below). By King's College Environmental Research Group.
(via Guy Lipman)

Using Natural Language Processing to discover themes in First World War poetry

A brilliant linguistic data analysis of war poetry, with a step-by-step tutorial in R.
"The First World War was one of the first fought in world history by troops who were almost entirely literate. Quickly the war reached a frightening, dull stalemate in the trenches, with well-educated men huddled below ground a few hundred yards from men from similar backgrounds on the other side. To pass the time, to record the horrors of 20th century warfare for posterity or simply to put into words what they were thinking, many soldiers at the front line picked up a pen and paper and recorded their experiences. Likewise, those back home could write about their husbands, sons and brothers at the front.
Much of it has been collected by the Poetry Foundation on this page. In this post we will do some textual analysis (natural language processing) of the poems using R to see whether we can pick out any themes and how they changed over the course of the war."
(via Sophie Warnes' Fair Warning)

America's Quietest Routes
A data-driven answer to the question "What’s the least-traveled route in America?"
"Quietness was calculated as the annual average daily traffic (AADT, measured in # of vehicles), and routes with the lowest AADT in each state were deemed the quietest."
It includes links to Google StreetView, so it's truly interactive.

China’s radical new rules to recycle rubbish
In this brilliant, interactive and visual article, we learn, among other things, that the Xian waste processing facility will "reach its maximum capacity by the end of October this year, more than 25 years earlier than anticipated".

Democratic Primaries: Preferential Poll Results
"We present some of the results in interactive charts below. In each case, first-preference support is shown in black: this is comparable to the standard horse-race figures in public polls. But we also know the subsequent preferences of each respondent—in other words, who they would support if they can’t have their top choice. We use this information to run a simulated instant-runoff election, eliminating candidates with the least support and distributing the vote of each supporter to their
next-preference candidate."

Robots will never completely replace people
"Concerns about the widespread displacement of workers by machines are not unfounded. Robots are more productive, do not get tired and can work round the clock. However, it is also worth noting that over the last century, technology has created more jobs than it has destroyed."
This piece by Nik Faulkner for VoxEurop/the European Data Journalism Network presents the issue with a few interesting data-driven analysis and visualizations.

Aussie Smoke Plumes Crossing Oceans
Satellites have been picking up the smoke generated by Australian wildfires coming up the sky. This picture comes from NASA's Earth Observatory and all the data is available.

Mapping the Shadows of New York City: Every Building, Every Block
"You’re looking at a map of all of the shadows produced by thousands of buildings in New York City over the course of one day."
And, conversely, you're looking at the rare, narrow bits of Manhattan that receive full sunlight.
Ah, the things you can do when your city authority provides the 3D shapes of buildings as Open Data (link below)!

Land doesn't vote, but it does matter
Alasdair Rae is – deservedly – one of the most linked to folks in this newsletter. He's now written a short blog post on the intriguing topic of "how to represent the results of elections in a way that reflects the true proportions of who voted for who". The post provides a good explanation of why mere geographical mapping is not good enough, but also notes that hexmaps, on their own, aren't that good either. Reality is complex, folks.
Also, check Alasdair's geographical data resources below, as they are really useful.

RawGraphs
"The missing link between spreadsheets and data visualization."
Or, more humbly, "an open source data visualization framework built with the goal of making the visual representation of complex data easy for everyone."

Open Data Maturity – Report 2019
"This report is the fifth in a series of annual studies that assess the level of open data maturity in the EU28
Member States plus the European Free Trade Association countries (Iceland, Liechtenstein,
Norway and Switzerland, together called the EU28+ countries). The report serves as a benchmark to gain
insights into the development achieved in the field of open data in Europe. The assessment measures
maturity against four open data dimensions: policy, portal, impact, and quality. The following trends and
results have been identified in the open data maturity assessment of 2019:"
You'll have to read it... ;-)