Built and am still improving a public web tool to explore music as a graph instead of a list.
Artists, bands, albums, instruments, collaborations — all navigable as an interactive node network on top of MusicBrainz data.
Music isn’t linear. It’s a web. So this lets you wander it.
https://overtone.kernelpanic.lol
Check it out and let me know what you think, all feedback is welcome :)
#music #musicdiscovery #DataVisualization #graphs #MusicBrainz #IndieWeb #OpenData
@state I haven't (and probably won't) had time to do a thorough review, but it seems like you covered everything I would have expected! In particular being able to browse via band members and tags.
Though I would like to see EP's and singles, or even be able to load groups with no releases so that I can browse their members/tags. I haven't checked if you can browse things via more obscure or technical relationships like engineers, events etc.
@state I wonder if you could do a cool '6 degrees of separation' game or feature? Seeing in how many clicks you can get from entity A to entity B. And/or show how two artists are interlinked.
By the way, I'm love giving feedback but I (and MusicBrainz staff in general) probably aren't your target audience - we are experts at wrangling MusicBrainz'.... utilitarian.... UI and it would be weird if we started using a different site to browse 😜
Keep up the good work -a
@musicbrainz
I'll also add in the engineers, that definitely should add a layer of interesting complexity.
In terms of the 6 degrees feature, I love the idea, had thought about it after reading up on the Erdős number while working on a parallel project. Something I've been toying with that would be cool is to either allow to build and listen to playlists on the platform or at least build exportable playlists with a selection of the visualization, so people can listen to new stuff easily.
@state Have you looked into ListenBrainz playlists? You should be able to use the LB API to produce and mix various artist/album/tag/etc playlists.
You can play with it here: https://listenbrainz.org/explore/lb-radio/
(our apologies, I believe you have to be logged in to use that now, it used to be open but bots/AI scrapers ruined that) -a
@musicbrainz
Hi again! Sorry for the delay, I wanted to take some good time to reply!
First of all, I'm happy you had time to review it at all, so thanks for that :D I opted for using tags to represent genres everywhere. Genres themselves felt a bit limiting and having both displayed made a mess out of the visuals. I figured tags contributed by people would convey the same as the genres and even more. I hadn't thought about adding engineers, but events are definitely a must.
stateless
MusicBrainz