There are so many projects I worked on that never saw the light of day… Little ideas that got to the prototype or research phase, but… for some reason could just not be finished or released. This is one of these projects: Spotify Jukebox.
To give a little backstory on what sparked this project. It was around the time that 3FM Serious Request was starting up. If you are Dutch, you cannot miss this, but for other people it is a period before Christmas where you can donate to the radio station for a charity and they will play the song that you requested. There are a lot of other activities/auctions where people collect money for the cause and there is a glass house where the DJ’s are locked up.
When talking with my friends, I also remembered a random pub in Boston I once visited where you, through an app, could buy tokens and in turn you could spend the tokens to queue songs on the playlist. Obviously you had to buy a pack of tokens like a real mobile app, so it ended up with me queuing up Never Gonna Give You Up for the remainder of the tokens I had when we left for the night.
So why not bring this to parties and have people chose a song for charity…. Spotify Jukebox was born.
Spotify has an API where you can control the app from anywhere. Something I, for example, use with my Stream Deck on my PC to skip songs in a playlist, but you can also use it to give a webapp access to your Spotify. Even the cardgame Hitser is using it to play the 30 seconds of music when playing the guessing game.
What the guest sees
With the API in mind, I started prototyping a little app where you go to a link in the browser, eventually it could be a QR code at the party, and you could search for the song you wanted to add to the queue. After payment you would see how long it roughly would take for your song to be played based on what was currently in the queue. If you wanted to know what was playing, you could also see the current song.
In this prototype the payment was completely set up with callbacks and iDeal payments, common in the Netherlands, through Mollie payments. I was thinking about other payment methods as the transaction fees were quite high, but they would not vibe with Dutch customers in a partygoers. iDeal is just so simple to use. It did mean that to make a little on a small scale the price per song request would be 2 euros. Maybe the transaction fees in the Boston pub were the reason for it to sell tokens instead of just a track request.
What the party owner would see
After logging in with Spotify, so I could already scale it without people needing to make accounts separately, the party owner would see a dashboard where he/she could setup and see the following on 3 separate pages:
Dashboard Page
- Guest page URL so that the QR code could be generated or the link could be shared.. I blacked this out on the screenshot as the app is still running for when I am with friends. Obviously just having the payment turned off.
- How many seconds a guest would be timed out after requesting a song, to prevent having Never Gonna Give You Up 10 times in a row like I did in that one Boston pub.
- Link their own Mollie account through an API key. I never meant to make money myself with the app in a large way. It would be something for the donations initially, so why should I do all the financial administration in the testing phase? They can explain the transactions to their accountant.
- They can see what is currently in the qeue
Manage Songs
- A list with approved tracks. If you have a Metal party you don’t want Celine Dion (or maybe you do). So I gave the Spotify account owner the ability to approve songs using the API of Spotify with what songs are available.
- They could use the single track function to search for, it is that simple, a single track.
- Or they could use the import playlist function to import a full playlist that they owned.
Requests
- Data is king, so it would save, but I never implemented this function fully, the name and email of every person that would request a song. Maybe you could use it later to send a concert invite? Just an example. Having contact of guests could potentially help with promoting other things later on.
Why it failed
When you read it maybe you thought: What a great idea! And I agree, but it will not work. Not because the app won’t work as this prototype is almost fully functional working with my own Spotify account… But because of the terms Spotify sets with using their app. I built this prototype with the mindset if I could, but never asked the question if I should.
Spotify does not allow playing music from their app in a public or commercial setting. That is it. That is why this little idea failed and the code is archived…
Leave a Reply