• As the previous week I wrote about the Advent Calendar of White-box Testing and other Advent traditions, the beginning of December has countless virtual traditions that suggest reflecting on the year passing in one way or the other. One of such is Spotify Wrapped which sums up your listening habits into several colorful infographics. Such campaigns that appear at a very specific time require thorough load testing, so here’s the article on how Spotify load tested for 2022 Wrapped.

  • Spotify is famous for its modern software engineering practices, innovation, experimentation, work culture… (just read “The Product Model by Spotify” by Joakim Sunden and you will get it) Therefore, it’s worth exploring Spotify’s engineering blog. Or if you want to focus more on quality engineering, there is a nice article focused on quality and productivity topics by Antoine Craske about “Quality Engineering Productivity at Spotify”.

    “Quickly turn ideas into products and experiment to improve the user experience, growing into new markets, and remaining competitive as a content streaming provider.” - somehow reminds me of Effort, Impact and Experimentation in Testing.

  • Another interesting piece from Spotify I found some time ago when I was exploring ways how to perform quality assessments in the project space - “Squad Health Check model”. (You can easily see the date it was published - so you can see Spotify was experimenting with gamification long before I started my career as a tester…) Some time later I found better and more tailored models for that same purpose, which I actively refer to this day:
  • And at the same time one idealizes their software development practices, work culture, speed, and other fancy things, they announced layoffs of 17% of their staff (~1500 out of 9000+ people). Many people on social media are blaming Spotify for layoffs with a 5-month severance package, but I am not sure what is worse - layoffs here and now or quiet firing. Oh, Bittersweet Symphony…

  • But let’s not end with a sad note. Let’s end forte! Something to try out! Let it be about music and AI, and a mix of both - let’s try Facebook’s MusicGen model demo available in HuggingFace which generates a 15-second melody out of your text query and (optional) music sample. It takes some time to generate that short melody, but it is surprisingly good. For advanced users, I suggest using MusicGen Colab space which is faster and has more options. Try it out!