The Moby Problem: open letter to Matt Ogle’s successor at Spotify

Almost every week, Spotify adds a Moby track to my Discover Weekly or Release Radar playlists – probably the playlists I listen to the most. The problem is: I don’t like Moby, and he’s not going away.

I’ve figured out exactly why Spotify keeps recommending me Moby. I’ve also figured out what types of user behaviour can discourage a recommendation system from continuing to recommend certain music. On Spotify, skips are weighted heavily. That is to say, if you skip a track, Spotify interprets it as you not liking a song or artist. I quite consistently skip the Moby tracks in my recommended playlists, but a week goes by and there he is again.

The Moby problem is not actually about Moby. It’s about the way recommendation algorithms work, and about the way we feed music data to them. The reason why Spotify keeps recommending me Moby is because I have a few Moby works in some of my playlists. I actually like his early rave stuff from the 90s, but I don’t care much for his chill out and trip hoppy stuff. Moby is perhaps also one of the most remixed electronic artists. Occasionally (and rarely), a really great remix sneaks into my playlists.

Hypothesis: playlists are weighted more heavily than skips

Three factors around playlists seem to be playing a role in Spotify’s assumption that I love Moby:

  • Moby’s inclusion in my playlists (passive)
  • Moby being played from my own playlist (active)
  • Moby being added to my playlists (active)

The weight in the algorithm should probably get heavier towards the bottom of this list, since it signals stronger intention and commitment. There may be many other factors at play too.

The fact that I like a couple of songs from an artist, some of which from over 20 years ago, does not mean I’d like to be kept up to date on his newest music though. Most of the Moby tracks that appear in my Release Radar are actually inter-genre remixes, so that really doesn’t make much sense either (e.g. if I like drum & bass, why would I like a techno remix of a drum & bass song?).

The remix problem

Then there’s another issue with remixes. One of my most-played playlists, called If Red Bull was Music, includes an EDM remix of a Moby track. It’s the only Moby track I listen to regularly, besides perhaps the Moby stuff in my Discover Weekly and Release Radar, when I forget to skip.

The problem is: it’s not a Moby track anymore. Sure, Moby is the original artist, but it doesn’t sound like a Moby track at all. It’s almost like categorizing a hiphop beat that samples Mozart as a piece of classical music.

It seems like Spotify is barely taking this into account when two artists can be lumped into the same category (electronic), even when that category is too broad to mean anything.

The solution

Let me banish artists! Give me a big fat ban button.

But hey, I’m a product person: I know the Moby problem is a symptom and you shouldn’t develop features to address symptoms — that’s how you kill a good product.

Spotify has a great product and Discover Weekly & Release Radar are a strong part of my music habits nowadays. So what it needs to do, is get better at understanding users’ actions and intentions, and how they weight them.

Personally, I think it’s important for them to look at how users interact with the music in their recommended mix playlists, and then weigh that much heavier. No engagement with a certain artist (or actually: skips), then that artist slowly becomes invisible, like in the Facebook news feed.

So to whomever is succeeding Matt Ogle, one of the creators of Discover Weekly, who just departed Spotify for Instagram, please solve my Moby problem. Let me escape this filter bubble.

(Just in case: hey Moby, I love your music, but most of it just doesn’t fit my taste so well. Keep doing what you’re doing!)

5 Bots You’ll Love

Since launching its chatbot API last April, Facebook’s Messenger platform has already spawned 11,000 bots. Bots are popular, because they allow brands to offer more personalized service to existing and potential customers. Instead of getting people to install an app or visit your website, they can do so from the comfort of their preferred platform, whether that’s WhatsApp, Messenger, Twitter or something else.

Bots, basically automated scripts with varying levels of complexity, are ushering a new wave of user experience design. Here are some of my favourite bots.

AutoTLDR – Reddit

AutoTLDR bot

AutoTLDR is a bot on Reddit that automatically posts summaries of news articles in comment threads. tl;dr is internet slang for “too long, didn’t read” and is often used at the top or bottom of posts to give a one-line summary or conclusion of a longer text. It uses SMMRY‘s API for shortening long texts.

The key to its success is Reddit’s digital darwinism of upvotes and downvotes. Good summaries by AutoTLDR can usually be found within the top 5 comments. If it summarizes poorly, you’re unlikely to come across its contribution.

Explaining the theory behind AutoTLDR bot.

Subreddit Simulator – Reddit

Subreddits on Reddit center around certain topics or types of content. Subreddit Simulator is a collection of bots that source material from other Reddits and, often quite randomly, create new posts and material based on that. Its most popular post is sourced from the “aww” Subreddit and most likely sourced two different posts to create this:

Rescued a stray cat

Check out other top posts here. Again, the reason why it works well is because of human curation. People closely follow Subreddit Simulator and upvote remarkable outcomes, like the above.

wayback_exe – Twitter

Remember the internet when it had an intro tune? wayback_exe takes you back to the days of dial up and provides your Twitter feed with regular screenshots of retro websites. By now, it’s basically art.

It uses the Internet Archive’s Wayback Machine, which has saved historic snapshots of websites.

old site 1

old site 2

pixelsorter – Twitter

If you’re into glitch art, you’ll love pixelsorter. It’s a bot that re-encodes images. You can tweet it an image and get a glitched out version back. Sometimes it talks to other image bots like badpng, cga.graphics, BMPbug, Lowpoly Bot, or Arty Bots. With amazing algorithmic results.

 

Generative bot – Twitter

Generative bot

Generative Bot is one of those bots that makes you realize: algorithms are able to produce art that trumps 90% of all other art. It uses some quite advanced mathematics to create a new piece every 2 hours. Seeding your Twitter feed with occasional computer-generated bits of inspiration.

Want more inspiration? We previously wrote about DJ Hardwell’s bot.

What are your favourite bots? Ping me on Twitter.