Sean

Version 1 of this player, and to be honest, my first player design. How could I improve it?

#buildinpublic #dataviz #music

Sure, exploring music data in a force graph is great but you should be able to listen too right? 🔊

How would you do it differently? 🤔

#buildinpublic #uidesign

Filtering logic and UX was a little challenging because there's like 2,500 genres.

To simplify, I grouped sub-genres under their parent genre, while still allowing the parent to be selected on its own

Slightly unconventional, but avoids clutter from duplicate parent checkboxes

Filter artists by any number of genres

Big unlock: you can now view all artists, capped with a node limiter control.

Right now, you're seeing the top 200, 500, 2000, etc. of ~180k artists, but how do you jump to the last 2,000, or the middle 2,000, in a UI with no “pages”?

#buildinpublic

Two big-ish updates:

1 → Finally got around to adding color. Right now it indicates which major genre-tree the node is a part of.

2 → Added a fun *auto-zoom on select* behavior. Makes it so much easier to traverse the graph

#buildinpublic #dataviz #uidesign

More music data coming in means more real estate needed to display it.

Making the drawer fully responsive from lg → md → sm screens was a a little tricky, but it's so satisfying seeing it snap in to place now. Still working on the layout of the content

#uidesign #buildinpublic

Forget about dang ol’ Cracker Barrel. Look what @figma.com’s done to this super cute toolbar button in Figjam 😢

Looks like @arcbrowser.bsky.social took away the AI-Assisted Find and Replace filter 😔

Since when did Spotify have a comment section?

The genre relationship data is all from MusicBrainz at the moment. UI made with React, Motion, and Shadcn

#buildinpublic #uidesign #dataviz

Finally able to add some clustering controls to the genre graph!

So far I have Hierarchy, Influence, and Fusion. You're also able to toggle 'DAG Mode' which is a more a way to view connections more linearly. The physics on that are super funky still 😆

Next up is setting up some clustering logic. Some I've been thinking about are clustering based on shared artists, cultural proximity (time and place), popularity (cumulative listeners and number of artists), and genre hierarchy.

What other ways of clustering could be useful for discovery?

New look for the genre graph, obviously inspired by @obsidian.md, and dark mode! #buildinpublic

Very cool! Can I ask what tool you’re using for these for graph visualizations? I’m working on an app right now that lets users explore music data spatially using react-force-graph

Implemented a command menu (search, if I'm being honest) for quicker switching between genre and artist selections

Looking into using the url-rels endpoint from MusicBrainz so we’ll see how that goes. Will definitely check out Deezer. Thanks for tips 🙌

Still getting the hang of @motion.dev and it took me longer than I’d like to admit to get this animation to where I want it. Also, just learned that @last.fm deprecated their artist image endpoint 😭

Any ideas for alternative approaches to sourcing artist images?

What were the apps? Curious to see what you could produce in two days :)

Messing around with some nav concepts in Figma for this music exploration tool I’m working on.

This one won’t stick because I want the list view to work along side the graph, updating to show the nodes in the viewport. Thought I’d share anyway just to see if it sparks any ideas for anyone ☺️

Bangin’ out this control panel doohickey for a project that lets users explore music data spatially with a force graph. Made with #react and #motion

One thing I’m a bit concerned about is walk-up observers not getting the full context of our conversation — which could mean we either miss the chance to gather feedback from them entirely, or we end up getting feedback that isn’t accurate.

I'll be gathering feedback on a new enterprise application feature in a busy expo-style booth.

Has anyone run quick discovery or usability sessions in a customer-facing conference setting? What formats or tactics helped you collect honest, actionable insights in just a few minutes per person?

Anyone have any experience doing discovery work or usability testing in a customer-facing conference setting? What worked for you?

Very cool! 👏

How did you find GPT was at working with Swift? I tried to get it to help me with a project and found it pretty miserable, but that was a couple of years ago

Is there any way to get around @figma.com Slides requiring people to log in before they can edit? I’d like to use the interaction widgets to collect feedback on something I’m working on, but I don’t want to force anyone to create an account

This is the way

I don't have as much of a problem with throwing away work as I do continuing to work on something that I know I'm going to have to throw away

Incredible work 👏