475: quantum of sollazzo

#475: quantum of sollazzo – 5 July 2022

The data newsletter by @puntofisso.

Hello, regular readers and welcome new ones :) This is Quantum of Sollazzo, the newsletter about all things data. I am Giuseppe Sollazzo, or @puntofisso. I’ve been sending this newsletter since 2012 to be a summary of all the articles with or about data that captured my attention over the previous week. The newsletter is and will always (well, for as long as I can keep going!) be free, but you’re welcome to become a friend via the links below.

I read this pretty great blog post titled “Wasting Time in Tech Interviews – ranting about crappy tech interview experiences”, which totally resonated. Back in the days, the crazy interview process is one of the things that got me out of big tech: I don’t think I’d have ever been able to pass those 6-7 whiteboarding interviews. Beyond the fact that they are so frustratingly disconnected from the reality of the roles, I’m just not good at fiddly exercises on the spot. But as I often say, they do select for one specific skill – the ability to commit. Committing to preparing algorithm knowledge on-the-fly, to the process, to the company, etc. They select amazing committers, not great on-the-job performers.

Every week I include a six-question interview with an inspiring data person. This week, I speak with Christine Jeavans of the BBC.

The most clicked link last week was the brilliant MarkWhen, the tool to create timelines from Markdown.

‘till next week,
Giuseppe @puntofisso

Six questions to...

Christine Jeavans

Christine is a Senior Data Journalist at BBC News.

What is your daily data work like and what tools do you use?

I work in the BBC News Data Journalism team which is made up of journalists, data scientists, statisticians and data designers. We work on a mixture of daily news and longer-term projects, on a wide range of topics and often collaborating with colleagues across the newsroom. Datasets we use may be open source, academic or FOI / self-gathered.
On a typical day I would be using R for analysis and charts, sometimes Flourish too. Colleagues in the team have Python skills and I’m learning from them. As a team we use Dropbox, Jira and Github to keep track of our projects and to share the workload.

Tell me about a data project that you're proud of...

Like many data journalists, I spent much of the past two years covering the coronavirus pandemic. Early on, a link was established between Covid deaths and deprivation but I was intrigued by a later comment from the Chief Medical Officer, Professor Chris Whitty that places worst-hit by Covid had also had high Victorian child mortality. I talked to Dr Alice Reid at the University of Cambridge who has been mapping 19th Century population data, and wrote a piece using her maps side by side with Covid mortality maps. It was shocking to see how poverty-linked ill health still reaches down the centuries, despite so many advances in living standards and treatment.

...and a data project that someone else did and you're jealous of.

There’s so much great work out there - and one of the things I love about the data journalism community is that there is always something to take inspiration from. It’s quite old now but I still admire this piece from the Tampa Bay Times - Why Pinellas County is the worst place in Florida to be black and go to public school. It shows how the state’s policy of re-segregration failed black children and builds a compelling case with the data - simply and very effectively. The piece was used as the introduction to the paper’s year-long  investigation which was a Pulitzer prize finalist. When you see that 154 children out of 160 at the worst school failed maths or reading it sends a chill down your spine for their futures.

If I say "dataset", you think of...

A box of information to puzzle out. It’s exciting to discover what it will reveal and satisfying to find the right way of analysing it. However, as with the examples above, most of the datasets I work with represent people and it’s important to me to keep the human element in the writing, analysis and visualisation.

Give someone new to data a tip or lesson you wish you'd learned earlier.

Never assume. Always look at the data definitions, read the notes and warnings and ask a subject-matter expert for advice if you need to. It’s easy to get quite far down the road of analysing a dataset only to later find that it doesn’t mean exactly what you thought it did. Also, by really thoroughly understanding the data and its limitations, you get a better and more solid story.

Data is or data are...

Data is (sorry, data scientist friends!)

Topical

Mapping Glastonbury

The V&A Museum has a multimedia Glastonbury archive, which is now made available through this interactive website.

What Can FBI Data Say About Crime in 2021? It’s Too Unreliable to Tell

“The transition to a new data system creates huge gaps in national crime stats sure to be exploited by politicians in this election year.“
Another incredible data-driven investigation by the Marshall Project.

The migration from the Former Soviet Union to the EU in six charts

A provocative data analysis by Six Questions graduate Francesco Piccinelli, who looks at migration patterns from the former USSR countries

Has Your State Made It Harder To Vote?

FiveThirtyEight looks at voting law variation in the US, noting some huge differences between different states. “In some, nearly every step of the voting process has changed since 2020.“

Car ownership in London

CARTO’s Helen McKenzie has made this striking interactive visualization that displays car ownership in London.
(via Massimo Conte)

Tools & Tutorials

Find areas that match a set of demographic variables and produce an infographic

“In this month’s tutorial, I’m sharing how I used Esri Demographics + smart map search to find US counties with a vulnerable elderly population and produced infographics with the Business Analyst Web App.“

Felt

Felt is a web service to make maps on the internet, which has just gone into public beta.

Pen and Paper Exercises in Machine Learning

211 pages of ML exercises in this PDF paper by Michael U. Gutmann at the University of Edinburgh.

What background color should your data vis have?

Datawrapper’s Lisa Charlotte Muth focusses her latest blog post on varied approaches to chart backgrounds.

Data thinking

From Source to Table: How Journalists Are Investigating Food Stories Worldwide

“Other FERN stories have been more driven by data — or a lack thereof, according to Cunningham. As part of an investigation with the Midwest Center for Investigative Reporting and the Guardian, FERN found that North Carolina had records of only 33 public complaints against livestock operations across 10 years — a sharp contrast to other hog-farming states that had thousands over a similar time period.“

Fair Game

“Commonly used by researchers and journalists, data scraping is an underacknowledged privacy concern.“

Hard Data: the erotics of infographics

Definitely the sort of headline that will raise your eyebrow. “It is the outcome of a visualization-first approach that results in stories pursued not based on their editorial merit but simply because the numbers and spreadsheets are available to create an arousing image.“
I feel seen.

Do you really understand the influential warming stripes?

“Information designers will notice what is “missing”: no title, no axes, no legend to explain what the colours mean. And yet, we intuitively get this gut feeling of a trend towards a world that is dangerously warmer.“

Dataviz, Data Analysis, & Interactive

Gender Data Lab

“Genderdatalab.org is the collaborative place where community members can build gender and sex disaggregated datasets to tackle the Gender Data Gap. We share knowledge and open datasets around gender data. If you want to visibilize inequality by using open data, join the community. It’s open and free!“

The Wood Database

“After cataloging the wood’s common and scientific names, weight, approximate cost, and any other notes or observations that I thought were unusual to that species, I printed the file out and used it as a reference guide in my shop.“
And then he made it into a website. The author, Steve Earis, is a woodturner based in the UK.

Talkative men

A quirky example of data sonification based on how much men and women spoke in one Zurich Cantonal Council meeting.

AI

Local Newsrooms Should Adopt AI Ethics as They Adopt AI: 5 Recommendations

“If we want to use AI to support local news, how can we make sure this technology’s business advantages don’t come at the expense of ethics or the societal benefits newsrooms provide? And, at a fundamental level, what should AI ethics for local news look like?“

quantum of sollazzo is supported by ProofRed’s excellent proofreading. If you need high-quality copy editing or proofreading, head to http://proofred.co.uk. Oh, they also make really good explainer videos.

Sponsors* casperdcl and iterative.ai Jeff Wilson Fay Simcock Naomi Penfold

[*] this is for all $5+/months Github sponsors. If you are one of those and don’t appear here, please e-mail me