While the name isn't fully decided yet, I thought I'd write a short post announcing the dapp idea I've been working on the past few days with my friend Claude.
First up, it should be noted that this is my first time diving into something like this, this type of code, etc, so there's been a lot of back and forth, questions and other token-wasting activities, but I'm generally happy with the progress and outcome so far and I think with feedback and a few testers later it can only get better.
I'm gonna go through some of the main things we've built and then maybe end it with a recording of the website in its current stages along with a "why this".
(AI generated logo)
Hive Scrobble (anyone still remembers scrobbles?)
is a fork of the open-source Web Scrobbler project (MIT license). Web Scrobbler handles the hard part, detecting what's playing on your browser across Youtube, Spotify web, Soundcloud and a bunch of other platforms. What we've added, along with some other changes, is a Hive scrobbler backend.
This means that when you listen to a song past a certain threshold, Hive Scrobbler broadcasts a custom_json op to Hive with Hive Keychain. That's pretty much it, your listening history goes on-chain signed with your posting key through keychain so its relatively safe. The extension also exists on your browser to let you know when songs get stored/remind you you have Hive Scrobble on.
This is what it looks like when it generates an operation:
"app": "hivescrobblesai/1.0",
"artist": "NF",
"title": "The Search",
"album": "The Search",
"timestamp": "2026-04-05T21:14:00Z",
"duration": "4:17",
"percent_played": 94,
"platform": "youtube"
You can see this in the wild by checking the block explorer on this testing account.
Some things to note
- you just need a hive account and the extension for it to work, we may add a way for others to store their data through our accounts until they're ready to upgrade to their hive accounts in the future
- scrobbles are feeless, although I'm sure you guys already expected that
- songs you start playing don't broadcast until a 60% threshold is reached, then only after a new song starts playing does the hive tx go out (it waits in case you're rewinding the same song)
- rewinding works but is a bit more complex to track and not bulletproof, if you love binge-listening to the same song we may still be able to credit you the appropriate scrobble credits
The extension supports 16 of the most popular platforms currently, do note that this only works for browser activity however and only chrome currently. To cover songs listened through software/apps we'd have to build a software ourselves in the future.
Zingit
Worth noting, the naming isn't written in stone yet~
The other side of this is our web app that reads scrobble data off of Hive and does something with it
Right now the site shows this:
- top scrobbles - most listened to songs across all accounts
- latest scrobbles as well as new scrobbles - latest songs/latest new songs that haven't been scrobbled before
- latest comments - comments placed on a song or timed comments placed within a timespan of a song
- youtube embed to check out the songs from our stat rankings
- lyrics fetched from LRCLIB to read while listening to a song
- some more stats with probably more to come
When a new song is scrobbled for the first time, we try and fetch as much data on it as we can such as ISRC (the music industry's standard unique track identifier) from Spotify - this helps with matching the songs across all the different platforms. This gets posted as a separate operation on hive to build a canonical record of every song that our users have listened to.
The timed comment system is something we're mostly interested in to develop further. The idea that you leave notes at a specific moment in a song and can discuss that further. We have it set up that while playing the song through our site the comments pop up whenever that moment hits, it adds a social layer to listening to songs. The infrastructure for it is already using @peakd's Snaps meaning that the comments are also on-chain.
We're currently working to make sure it works through both platforms, i.e. leaving general comments or timed comments on both Zingit or through Peakd places them in the right place on a song - as long as you're registered with us and are using the extension.
What, but why?
Some of you may be wondering that. Naturally, this is opt-in and not forcing anyone to participate, nor will there be any tokens to entice you to.
Discovery is a big part of it. Aiding people to get into new songs, artists and potentially genres. Most users, according to studies, fall back to what they're comfortable with and know after a certain age, and coming into contact with new material isn't easy. Platforms mainly recommend you songs that pay them, and many artists go unheard without listeners ever finding out they wanted to hear their craft. The social side, for starters, helps bridge some gaps, you may be curious what your followers are listening to, and you check out some of their top scrobbles until you get that "zing" feeling of falling in love with a new song, band or genre you didn't know existed.
With all the data, we can focus on listening behavior over tags/algorithmic guesses or other methods from centralized parties. it also enables smarter recommendation algorithms over time, especially with AI diving into it to get to know you as an individual listener better. By checking your scrobble history, your social cues on songs, and later even personalized agents you can talk to about your recommendations and listening habits.
Music is the beginning, once things are set up, we may add other formats to the social layer as well as more platforms where scrobbling is enabled - such as @threespeak, etc.
I'm looking forward to see how this can enrich the social side of our users while they're being entertained by music, videos, podcasts, etc, while making discoverability better for them.
While things aren't ready for a public launch yet, I'll be offering the extension to a few people who wanna test it out.
Here's a quick video on what the website looks like currently along with testing some functions:
https://youtu.be/DMtkfZIy9tE?si=JHCXTHezxAOSHWKb
Anyway, a lot more left to do before we get a domain going and make a public release so you can start filling up your song history. Let me know first impressions/thoughts/feedback/etc.