460: quantum of sollazzo

#460: quantum of sollazzo – 22 March 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’ve been away for three weeks, in part to do an 8-day long road trip of Andalusia which was amazing: beautiful scenery, way too much food, hikes in between the terrific and the terrifying (check out the Caminito del Rey), and catching up with friends who moved there and who I hadn’t seen in ages. So this week’s newsletter is a bit longer because so much has happened in data journalism over the past few weeks, sadly mostly related to the war in Ukraine.

This week marks the beginning of series 3 of my “Six Questions”, a collection of short standard interviews in which I hear from someone who works with data. In the 34 interviews published so far we’ve learned from data journalists, designers, analysts, policy analysts, and scientists. This week, we meet Lesley-Anne Kelly, the Data Content Lead of the recently formed data team at DC Thomson, the media company that runs a number of popular publications in Scotland.

Before I close – you might remember that some time back I created some road-colouring maps of London that show roads coloured according to their suffix, i.e. whether they are roads, streets, avenues, ways, and so on. They sold really well, and I have only 5 left. I need to empty my loft so… here’s a 70% discount which basically will get you the map for a fiver.

‘till next week,
Giuseppe @puntofisso

Six questions to...

Lesley-Anne Kelley

Lesley-Anne is the Data Content Lead of the recently formed data team at DC Thomson.

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

My team works on a mix of our own content whilst also supporting the wider newsroom. This gives us a wide variety of subjects to work on every day ranging from the ever present coronavirus to sport.

I prefer to work in R whenever I can but I also spend a decent amount of time arguing with Excel and Google sheets and our visualisation is done in Flourish. I’ve recently started to learn more about Github Actions which opened up a massive door to automation in newsrooms and I’ll be continuing to tinker with that in the near future.

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

We have a saying in the data team which is “NYT dreams on a DCT budget”. Based on that thing we all do of seeing a fantastic NYT data project and being down on yourself that you couldn’t ever do the same… and then you get to the bottom of the project and the byline runs into double digits. We are a three woman team and don’t have the resources of the big guys but we try to bring some of their ideas to our local audience. An example of that was our project that took 11,000 local WW1 roll of honour records and geocoded them, mapped them out, and created a scrollytelling article that we hope told some of the stories behind the data.

We are also due to publish a series of stories based on my absolute favourite data of all time – Scottish baby names – very soon. I will die on the hill that you can learn more about trends in a country by looking at what people name their children than you can from just about any other dataset. It also really signifies one of the things that I feel really passionately about which is that data journalism can be fun and shouldn’t just be reserved for the serious subjects. This seems to me to be a hangover from data teams often being lumped in with investigations teams – generally because newsrooms weren’t really sure where to put them…

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

As a regular reader of this newsletter I know that everyone and their mother talks about the Pudding article about women’s pockets – and for good reason as it is incredible. I’ll expand on that and say essentially everything the Pudding has ever done (I want to be them when I grow up) but if I had to choose one (other than the aforementioned pockets article) I’d have to go for their story on the history of rickrolling.

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

Oooh gimme. No really though, a dataset is just a source and data journalism is just interviewing that source. The fact that the sources tend to not talk back is a bonus though.

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

If you have any repetitive data task at all it’s in your interest to learn to code. R, python, SQL – whichever you feel most comfortable with. I would not have made it through 2020 if I hadn’t got serious about learning R when that whole pandemic thing kicked off.

Data is or data are...

I know the technically correct answer is ‘data are’ but if I’m in casual conversation my brain is going to say ‘is’. I need to be re-programmed.

Topical

I’m not sure people appreciate quite how bad the Covid situation is in Hong Kong

“....nor what might be around the corner”, says John Burn-Murdoch in this thread which summarises his analysis for a Financial Times article. It’s rather scary.

Kyiv to fall to Russian forces by April 2022?

A forecast from Metaculus, a website dedicated to creating community crowdsourced predictions. Also take a look at the connected questions.

Live monitoring of all sanctions against Russia

“Sanctions are the main weapon of the West in Ukraine conflict. But their effectiveness remains an open question. We’re providing daily updates on which embargoes are imposed on whom and answering your most important questions.“
By the excellent team at Correctiv.

War in Ukraine: the daily map to follow the situation city by city

Automatically translated into English from the original French, this article is another live tracker of the war in Ukraine and it is complemented by a thread on how they use Wikipedia and why they publish their source code by data journo Victor Alexandre.

Kyiv Convoy Tracker

Although it is no longer updating after the convoy apparently departed, this web page shows the incredible opportunities offered by the Sentinel satellite imagery.

Airwars – Ukraine population density

“This interactive map depicts the population per square kilometre of Ukraine” before the invasion. The rest of the website is pretty interesting, too.

Ukraine (humanitarian) Data Explorer

A tool to explore the humanitarian impact of the conflict, developed by the UN OCHA – the United Nations Office for the Coordination of Humanitarian Affairs.

In which EU regions do people work less and get paid more?

“The average working week in Europe has shortened since 2000, but the differences between EU regions remain significant. People in some Greek regions work up to 12 hours longer than people in some regions of the Netherlands.“
This article shows some clear patterns, some of which you might find surprising.

Tools & Tutorials

How we tracked flights over Ukraine

The Economist’s Off The Chart newsletter never ceases to amaze me for the transparent, clear discussion of technical approaches to their data analysis and visualizations. In this issue we learn how they tracked the emptying out of Ukrainian skies upon the breaking out of the war, which resulted in this article.

Flight routes and map projections

Following on from the previous link, take a look at short Datawrapper article on the use of the orthographic projection (currently not available in Datawrapper).

Investigating Russia Around the World: A GIJN Instant Toolkit

A useful set of resources by the GIJN.

Data diffs: Algorithms for explaining what changed in a dataset

“tl;dr: part 1 explains what an explanation algorithm is, and part 2 describes an open source SQL data differ.“

Goldilocks Zone Finder

Find your Goldilocks zone (US-only).

The Atlas of Economic Complexity

“Harvard Growth Lab’s research and data visualization tool used to understand the economic dynamics and new growth opportunities for every country worldwide.“
(via Massimo Conte)

Best of NICAR

J++’s‘ newsletter has a good round-up of the best hands-on sessions at this year’s NICAR, which is probably the most important annual data journalism conference.
They also share this interesting table by The New York Times’ Matthew Ericson, showing that in 18 years of NICAR, Excel is still a force to be reckoned with, while QGIS is definitely acquiring more and more importance in the newsroom.

Resources from NICAR 2020, 2021 & 2022

A fully searchable list.

Wordle, 15 Million Tweets Later

“So, to really understand Wordle, I had to look at the tweets – all 15 million of them.“
This is a pretty cool Observable notebook.

Data thinking

Lancet – England does not have the worst death rate in Europe

Five assumptions about COVID get fundamentally disproven by this paper, which is summarised in this Twitter thread.
Colin Angus also makes some good points about the methodology.

A Royal Mail adventure…

“…or why it’s hard to use structured data to tell the whole picture…”.
A fun thread by Keelan Fadden-Hopper, but it will make you realise how difficult data modelling is.

Dataviz, Data Analysis, & Interactive

The Toll of War Visualized

A sobering article that includes Valentina D’Efilippo’s famous poppy field.
(via Massimo Conte)

Dear Mr. Putin, what do you want?

“Data about Eastern European countries show the real aim of Russian foreign policy – and what is wrong with that”, says Francesco Piccinelli in one of his recent newsletters.

AI

Neural nets are not “slightly conscious,” and AI PR can do with less hype

“Prominent AI researchers should take accurate science communication more seriously.“

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 Steve Parks

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