Every ten days or so, I share a quick digest of what I've been working on and reading. Here's the latest. More in the series here.
I've figured how to track how many links get clicked in these messages. So, stealing a trick from Quantum of Solazzo, I can exclusively reveal that the most-clicked link in my last roundup was the article from Tim Harford on how one of the biggest ever financial bubbles was not in fact, the Dutch tulip mania (which scarcely caused a ripple) but the British Railway Mania, which devastated the middle classes.
My data sonification podcast Loud Numbers was featured in this new scientific paper analysing 32 recent sonifications and visualizations of climate data.
Here's what they wrote:
Geere and Quick have specialized in making “sonification podcasts” for their series LoudNumbers (https://www.loudnumbers.net/). Each program typically starts with a pedagogic explanation of the design strategy, which effortlessly translates into listening tips (i.e., providing a legend for how the listener can extract meaning from the sounds). In “The End of the Road”, they built on Pape Møller's laboriously collected time series data on insect population density on a rural road in Denmark, and more recently, they turned data from traditional ice measurements in a village in Alaska into techno music.
If you're new around here and haven't heard my Loud Numbers project then I'd definitely recommend you give it a listen.
As promised last time, I've completed my portfolio rework. This was long overdue, but as I'm enjoying a little lull in client work it was a good time to get it done.
To bring it up to date, I've put together new case studies on the work that I've done over the past 18 months for Possible on the Car Free Cities campaign, for Conservation International on the Exponential Roadmap for Natural Climate Solutions, and for Christine Arena on countering climate misinformation on social media.
You'll also find older case studies in there on my work for Information is Beautiful, on my Carbon in Context project, and on Loud Numbers - my data sonification podcast and studio.
If you know anyone who you think might have interesting data stories they'd like to surface to a broader audience, then send them my way! Best place to direct them to is duncangeere.com.
If you've been looking skyward in the early evenings over the last week or so, you'll have noticed two bright stars near the western horizon. They're Venus and Jupiter, which have been slowly drawing together and will now move apart again.
This ten-day composite by Soumyadeep Mukherjee beautifully captures their celestial dance.
Via Corey S Powell.
I love this new European night train map by Back On Track. It's a schematic subway-style map that shows route frequency, operators, and interchange locations for all of Europe's night trains - perfect for planning rail trips across the continent.
Trains are the most sustainable way to travel long distances, and taking a night train can save you money on hotels while giving you more time at your destination. With transportation rapidly becoming the EU's biggest emissions problem, sustainable travel is more important than ever.
Want to travel more sustainably but struggling to fit it into your work holiday allowance? Persuade your company to sign up to the Climate Perks scheme, where employers offer additional “travel days” for low-carbon holidays.
With a few of my remote worker colleagues at Possible, I've been trying out the Presence system developed by Mu-An Chiou. The link is kind of a long read, so feel free to skip to the para starting "Fast forward to the present day".
It's basically a system for representing presence in a virtual setting, getting a bit more of that "we're all in an office together" vibe, without actually needing to be.
We use a Google Jamboard rather than Figma (because most of my colleagues don't use Figma). We create a new "slide" each day, keeping the archive for posterity. But otherwise our system is pretty similar to Chiou's - we share a pic from our workspace (defined quite broadly), something we're working on, music we're listening to, and then just chat kinda asynchronously throughout the day with sticky notes.
It's super lovely, very wholesome, and if you work remotely and miss a bit of that office culture, I'd recommend you give it a try too.
If you're not tired of reading about ChatGPT, then I recommend my former boss Matt Locke's excellent piece in his How To Measure Ghosts newsletter about how these technologies aren't the oracles that they appear to be, but that they're actually maps.
Hm, that was all one long sentence and kinda messy. Let's ask ChatGPT to refine it.
If you're still interested in reading about ChatGPT, I highly recommend my former boss Matt Locke's insightful article in his newsletter How To Measure Ghosts. In his piece, he explores how technologies such as ChatGPT aren't just oracles but are more like maps. He challenges the conventional notion of these technologies and provides a fresh perspective on their true nature. You can read the article here.
Thanks ChatGPT, but that's now a bit long. Can you make it shorter again?
Check out my former boss Matt Locke's excellent piece in his How To Measure Ghosts newsletter. He challenges the idea that technologies like ChatGPT are oracles and instead argues they're more like maps. You can read it here.
Better, but now it kinda reads a little weird - the first . Rewrite it entirely but keep the length.
Read Matt Locke's thought-provoking article in How To Measure Ghosts newsletter about the true nature of technologies like ChatGPT. He challenges the conventional notion that they are oracles and suggests they are more like maps. Check it out here.
Er no, now that reads kinda SEO-y and promotional. Write it again but less promotional, and in the style of William Shakespeare.
Pray thee, peruse the excellent treatise by my erstwhile master, Matt Locke, in his How To Measure Ghosts newsletter. He doth challenge the prevailing belief that technologies such as ChatGPT be oracles, and instead posits they resemble maps. Read his musings hither.
Ethan Mollick has another good post on the same subject, looking at the outcomes of the first controlled experiments on the use of generative AI technology.
The key findings:
- A study of programmers found a increase of 55.8% in productivity when using AI
- A second study asked professionals to write realistic memos, strategy documents and policies. The ones who were given ChatGPT completed tasks 37% faster, and their average writing quality increased as well.
- A third paper released this week establishes which occupations and industries faced the most exposure to AI. The lists consist both of very large industries and many highly-skilled jobs, from academics to teachers to psychologists to lawyers.
Finally, he reports that the results of a much-less-scientific Twitter poll, which found that more than half of users reported that they've used the technology without telling anyone at least some of the time. "There are secret cyborgs among us," he writes.
Read his musings hither.
OK that's enough for this week. See you again in ten days.