That ghost is just a kid in a sheet

Should Have Been Listening to Phoebe Bridgers

Spotify's Wrapped is very shareable. We're all proud of our music taste and it's interesting to see what songs soundtracked a friend's year. But insightful? That I listened to a lot of Taylor Swift was not news to anyone. What if Wrapped took a longer term view? What if it considered the full thirteen years I've been with Spotify?

Wrapped could have told me that I lost 570 days when I should have been listening to Phoebe Bridgers. While quick to listen to new music I'm slow to get hooked. The artists that I return to again and again are often those whose music I didn't initially enjoy. Phoebe Bridgers was one, but so were Radiohead and Oasis.

I'd listened to Jon Pareles discussing Phoebe Bridgers on Popcast for years. Friends had recommended her. I'd even played Stranger in the Alps several times. The potential to be a fan existed, but I failed. It was finally coming across three live performances (Fake Plastic Trees, Georgia and Motion Sickness) that changed everything. I've had Phoebe Bridgers on repeat ever since.

What is love? It's being in my top five played artists in a 30-day window. It took me 570 days from first listening to Phoebe Bridgers in May 2019 to falling in love with her. 570 days is a long time, but it took me 3,651 days to appreciate Daniel Avery and 3,567 to fall for Justin Bieber. Other artists are immediate hits. I only lost a day's potential listening for HAIM and two for London Grammar.

How We Spend Our Minutes Is How We Spend Our Lives

2022's Wrapped included “My Audio Day”, identifying the type of music I listened to during the morning, afternoon and night. Aggregating so much down to so little is somewhat nonsensical but the question of when I listened to music was intriguing.

Averaging listening times across 13 years flattened out any detail and just showed when I sleep.

Broken down by year and with weekends filtered out shows how my listening changed through jobs, roles, and lockdown. From 2010 to 2017 I worked in an IBM office where I'd start early, leave for the gym in the late afternoon and work in the evening. I was foucssed on individual projects. Most of my time in the office would be spent with headphones on.

In 2018 and 2019 I worked at Net-A-Porter as part of a development team sat around a shared desk. We were all working on the same projects and I rarely had headphones on. The only time I was listening to music was during my morning and evening commutes.

2020 brought working from home. No commute, no shared desk. Other than a break around teatime I was listening to music continuously. While still working from home my role and number of meetings changed in the latter part of 2021 through 2022. I got a few early hours of focussed work (and music) before 9.30am and a small window when everyone else was having lunch. Most of my working hours were spent in web meetings. I hadn't appreciated how stark the change from 2020 was and I wasn't expecting Spotify to identify it.

None of this is a replacement for Wrapped. It might be more detailed and more interesting to me, but unlike Wrapped it's neither shareable nor succinct. It still doesn't tell me who I should be listening to. I don't think I'm finished with this idea. Expect a 10-minute Taylor's version in the future.

Sleeve Notes

Spotify's Wrapped press site has more general trends. Everything on this page is based on my own streaming history. Spotify allow you to download the last year's history. If you want to go back further, you need to make a specific GDPR request via email.

With 13 years of streaming history, I needed to do some data manipulation but little beyond putting tracks into buckets and counting. For the missed opportunities I filtered out any artists who I'd listened to before Spotify existed. The charts were all hacked together using D3.js and the examples they provide.

There are many great album covers, but Joy Division's Unknown Pleasures is probably the best known.

Annie Dillard's original quote was “How we spend our days is, of course, how we spend our lives.”

Loving an artist isn't just about their music. Personality, politics, style, and appearance all come into it. Everything I like about Phoebe Bridges outside of music is in her Chicken Shop Date. It's also about how and where you first came across the music. It's easier to remember lyrics from songs that soundtracked your school days than those you listened to last week. It might be why there are many good personalised music recommendation systems, but no great ones.

NYT's Popcast is best podcast I listen to. It treats pop music criticism with the seriousness it deserves.

Everyone has time for Taylor Swift's 10-minute rework of All Too Well.

Algorithmic Twombly

My favourite things at the Louvre Abu Dhabi were Cy Twombly's Untitled I-IX paintings. They're structured but chaotic and beautifully imperfect.

Flow-vis for Net-A-Porter

What does a day's trading at Net-A-Porter look like? Taking inspiration from Formula 1's use of flow-vis paint, I came up with a visualisation for a YNAP hackday.

YNAP Hackday Poster

The most significant colours sampled from the product images of the top 1024 clothing products on Net-A-Porter, organised by hue.

Should Have Been Listening to Phoebe Bridgers

Spotify's Wrapped is very shareable but what if it took a longer term view? What if it considered the full thirteen years I've been with Spotify?

My Favourite Net-A-Porter Colour is Black

Net-A-Porter's Spring Summer 2022 marketing campaign, "Go for Bold", centred around a collection of colourful products. Marketing wanted to add a technology element and one idea was to use palettes as a route to finding products. Ultimately this was binned, but I built it anyway.

Has Roger Federer Perspired?

That Roger Federer does not sweat had become ingrained thinking, the sort of idea we were looking to challenge. Was it real or just a lazy cliche? We had IBM’s Wimbledon match data for all the top players and using Weather Underground we pulled in temperature data for those matches. This let us see the number of matches played by player and temperature.

Physical Web and Physical Meetings

As an experiment in using the Physical Web I wanted to create a voting system for physical meetings. A meeting would have a current question and attendees could vote with one click. There would be no entering URLs, downloading apps, or scanning QR codes.


I'm Darren Shaw. I work as a software engineer at a tiny bank in London. I previously worked at Net-A-Porter and IBM. These are my side projects. Away from work I also play at being a portrait photographer.