Copy
19 January 2021 
#405: quantum of sollazzo – The data newsletter by @puntofisso

Read it in your browser


2021: the Simpsons' living standard is now a dream
 
A few hours ago, I came across this article on The Atlantic with a rather preposterous title: "The Life in The Simpsons Is No Longer Attainable". I found it pretty good: the ingenious storytelling of presenting the Simpson as a common working class family and showing how their living standard is now the stuff of dreams for many (from that era, I'm also thinking that the Conners in Roseanne would be another excellent example), is punctuated by great data references. "The most famous dysfunctional family of 1990s television enjoyed, by today’s standards, an almost dreamily secure existence."

The Simpsons' woes resonate, and the stats that the article presents are jaw dropping: "The purchasing power of Homer’s paycheck, moreover, has shrunk dramatically. The median house costs 2.4 times what it did in the mid-’90s. Health-care expenses for one person are three times what they were 25 years ago. The median tuition for a four-year college is 1.8 times what it was then. In today’s world, Marge would have to get a job too. But even then, they would struggle. Inflation and stagnant wages have led to a rise in two-income households, but to an erosion of economic stability for the people who occupy them."

The key lesson of data science about survivor bias was recently revisited (thanks Alex):
 

As one of the replies in that thread highlights, this is common in a variety of problems, and it's now visible in sports thanks to the growing use of analytics (I'm looking at you, Moneyball). This, however, often means that data is used with an arbitrary assumption of what data matters; for example, who decided that a defender should be assessed based on how many tackles they do? Maldini is brought up as an example of a defender who was impossible to pass, yet made very few tackles, because his positioning was incredibly good.

As this article suggests, "Some players don’t score goals, don’t give assists, don’t run much and don’t make tackles yet are very important to a team.". I'm not really into football, but the problem of using data in the wrong way, without caveats, has always bothered me.

Look below for some sponsored geotastic content – Ed Freyfogle, who's the organiser of location-based service meetup Geomob, co-host of the Geomob podcast, and co-founder of the OpenCage Geocoder, has offered to introduce a series of points around the topic of geodata. His fourth entry, on whether you should build or buy a geocoder, is below.

Till next week,
––Giuseppe @puntofisso
 

 

 
--- Sponsored content by Open Cage ---

Build or Buy? Should you try to create your own geocoder?

Welcome to part four of our series on geocoding.

Given freely available opensource software, and open data like OpenStreetMap, should you run your own geocoder? Or should you pay a service like ours to host the geocoding software for you? The whole point of open data is that you can do it yourself, right?

The short answer is that yes, you can run, or even write, your own geocoder. Unique technical requirements may mean it makes sense to craft your own custom service, but most people prefer to leave it experts and get on with their real work.

Our geocoding API aggregates many different open data sources and provides enterprise level reliability. One factor we see many people overlook: setting up the software is one thing, keeping the underlying data current is another. Put another way: building is easy, maintaining is hard. OSM alone gets 4-5 million edits per day. Still, as a developer myself and long-time OSM contributor, I understand the inclination to get your hands dirty. Hopefully it helps put you at ease to know that we’re doing our part to give back to the open data and open source geo software community.

Finally, rest easy knowing that if you ever need to the data and code is all there for you to dive into. That’s the real power of open-source and open data.

 

 

 

Tools & How To

Source Code for David Spiegelhalter's The Art of Statistics
All the R scripts that underlie the statistical examples in Spiegelhalter's book.
Apparently, he says he doesn't code and that most of this stuff is done by his collaborators, but I don't believe him ;-)

 

Choose a License
"Answer the questions or use the search to find the license you want"
(via Maurizio Napolitano)
 

Stealing Your Private YouTube Videos, One Frame at a Time
"Remember, always only test against resources/accounts you own!"
This tutorial is on the nerdy side. It shows how you can get access to a private YouTube video (one frame at a time) by exploiting a bug in how thumbnails are generated in Google Ads (yes, it's not immediately useful, but an interesting process to think about for the nerdier journos amongst you).
 
 

Data thinking

Matteo Moretti: Design and European Data Journalism
"Matteo Moretti, designer and founder of Sheldon.studio (a member of EDJNet), explains his approach to data journalism, with a special attention to visual journalism, interactive and responsive products, and informative experiences. This video is the first item of a series of interviews on data journalism from a European perspective, which are produced by OBC Transeuropa for EDJNet."
 

Complex visualizations and visualized complexity: how can we interpret the world around us?
Particularly during a global pandemic, "our world view is completely mediated by the information sources we choose: our perception and the resulting choices are influenced by the interpreters we choose."
This blog post presents some reflections on how ordinary people can interpret complexity.
 

History

Why did the Kursk sink?
"Twenty years ago, planned sea exercises in the Barents Sea ended in a horrifying tragedy. There were two explosions aboard the Kursk submarine. The vessel sank. All 118 members of the crew were killed."
I remember this terrifying story as if it were yesterday. There was little hope from the beginning, and a lot of theatre around the alleged attempts to save the sailors. I wonder what role was played by the media's thirst for an exciting story, on the side of the Russian authorities' reticence to tell us what was really going on.

 
 

Politics

The Mayors Dialogue on Growth and Solidarity
Federica Fragapane and Alex Piacentini have created this very nice and rich dataviz for the Overseas Development Institute. It shows "African and European cities taking action on human mobility".
 

The risks and opportunities of “remote democracy” 
"In recent months, Covid-19 has put European democracies to the test. In many countries, the space for democratic deliberation has been greatly reduced. In this regard, new technologies represent a major opportunity, Openpolis argues."
Credits: EDJNet.
 

Why Warnock and Ossoff Won in Georgia
"Far more people of both parties voted than usual in a runoff, but the Democratic turnout was stronger, largely because of Black voters."
Interesting article about the Georgia US Senate run-offs, especially because of the chart below, which I found somewhat surprising (although not entirely so) when compared to previous data analyses.
 
 

Maps

ONS Basemaps
"Ever wondered what your UK neighbourhood looked like 100+ years ago?"
A nice little tool developed by the amazing Ahmad Barclay that allows you to compare current and historical map with a slider.
 

Sanctions Explorer
A search engine and map to explore 28 years of sanctions data from the UN, EU, and (US Treasury Agency) OFAC.
 
 

Everything Else

Wealth shown to scale
Simple, but visually impressive.

How much money did I lose by not investing in Bitcoin?
Or... how much health did I not lose by not investing in Bitcoin and misplacing the wallet's private key? :)
This article is by the ever-talented Edurne Morillo.
 
Listen to Wikipedia
"Listen to the sound of Wikipedia's recent changes feed. Bells indicate additions and string plucks indicate subtractions. Pitch changes according to the size of the edit; the larger the edit, the deeper the note."
Etc... you get the gist.
(via Catherina)
 

 
Support this newsletter & spread the word

Become a GitHub Sponsor :) It costs about a coffee per month, and you'll get an Open Data Rottweiler sticker (and other stuff). 

If you're a supporter of this newsletter, thanks a lot for your support. Share this e-mail with a friend, or via social media


    


"In other news" is supported by ProofRed, who offer an excellent proofreading service. If you need high-quality copy editing or proofreading, head to www.proofred.co.uk. Oh, they also make really good explainer videos.
Supported by my GitHub Sponsors 
Steve Parks
Naomi Penfold
Chris Weston
Fay Simcock
Chris Noden
Jeff Wilson
& others


Copyright © 2021 Puntofisso, All rights reserved.



unsubscribe from this list    update subscription preferences 

Email Marketing Powered by Mailchimp