Nostr Web Client

Saturday night launch. 🚀

Nostr's UX problem isn't a secret.

30-day retention trends to 0%. Daily active users stuck at ~10k. Posts disappear. Followers vanish when you switch apps. Every app feels like beta software.

I've spent 2+ years arguing that great UX wins. So I designed a meta-study, threw a few hundred dollars at Claude, and put together what the evidence indicates.

tl;dr: Ship working experiences, then add features.

🔗 https://nostr-ux.com

6 critical patterns backed by 100+ citations:

‣ Onboarding, content discovery, core interactions

‣ Performance, progressive complexity, cross-client consistency

‣ Anti-patterns to avoid + validation checklists

I'm not a designer, but I know what good looks like. This is opinionated, evidence-based, and focused on retention.

What am I missing? What did I get wrong?

Send PRs this way: https://github.com/shawnyeager/nostr-ux-research

cc: nostr:nprofile1qyxhwumn8ghj7mn0wvhxcmmvqyg8wumn8ghj7mn0wd68ytnvv9hxgqpqr0rs5q2gk0e3dk3nlc7gnu378ec6cnlenqp8a3cjhyzu6f8k5sgsy67l5w nostr:nprofile1qy8hwumn8ghj7mn09eehgu3wvdeqzrthwden5te0dehhxtnvdakqqgqh2wxu9f38d8gfgsl33smuhc6cl26mh7vpzu6592nutlchrmthcsdtdqz4 nostr:nprofile1qydhwumn8ghj7cmgwfhku6trd3jjuer5dahx7m3wvdhk6tcpr9mhxue69uhkcctwvuh8yetvv9uhxtnvv9hxgtmfwsqzq77777lz9hvwt86xqrsyf2jn588ewk5aclf8mavr80rhmduy5kq92xkt0h

Garbage nsec 1mo ago 💬 1

Have you considered that it might be time to throw in the towel on Nostr 1.0 and start looking towards Nostr 2.0? An outright refresh?

I think the protocol is too paralysed to be able to implement these recommendations in any serious way. It's been 2 years to remove NIP4 DMs, which everyone agrees are bad (and critically insecure to boot). If something that everyone agrees is bad, and has agreed as such for two years, takes this much work to be rid of (and is STILL here) then what hope for the things you're proposing? There's just no mechanism, and no mechanism to make a mechanism.

And the proof is in the numbers. They are flat for 2 years at 10k, as you point out, and they will likely drop overall this year. And that's *with* all the work that's gone in. And there's no reason to think 2026 and 2027 will be different. Many hopes seem pinned on some Deus ex machina, but there are no signs what that Deus ex machina might be.

I think some thoughts on what Nostr 2.0 could look like would the best thing at this stage.

Reply to this note

Please Login to reply.

Discussion

hodlbod 1mo ago 💬 2

UX is not fixed by starting over from scratch. It takes many years to develop all the basics needed to make good software. The protocol has problems and a successor will come someday, but we have real technical momentum we shouldn't just dump.

Shawn 1mo ago 💬 1

I agree. The way forward is to push and grow, not abandon and restart.

Garbage nsec 1mo ago 💬 1

I disagree. The problems are too fundamental and it's only dragging out the inevitable. A refresh though, with all the many lessons learned, could be the start of something fundamentally transformative.

Shawn 1mo ago 💬 1

Get to work, fren. Pull requests are the way.

Garbage nsec 1mo ago 💬 1

Working yes, but I'm working with others on a hard fork for a different use case.

Pull requests for this current incarnation and use case are not the way, I feel. It's just not scalable.

Shawn 1mo ago

Do keep me posted. 🤝🏻

Garbage nsec 1mo ago 💬 2

Much of the work can absolutely be ported over. If you remember Mozilla Suite, it became this bloated application that combined a browser, email client, and other junk. Some devs forked it and started the Firefox project as an experiment to create a leaner, faster, standalone browser alternative. That worked well enough. Many such cases. It can be the way.

What we have now is not what people want. It just isn't. Not only normies but most people. Even Ross Ulbricht left to go back to posting couple pics on Instagram.

I'm not saying there isn't some Deux ex machina waiting in the wings, but the numbers do not look good no matter how you frame them.

hodlbod 1mo ago 💬 1

I agree, not because of the numbers, but because there are several unsolved or poorly solved technical problems. But what is great about nostr can be preserved in a fork/port if the protocol designers understand how to sort out what is good from the garbage.

Garbage nsec 1mo ago

Agree. The bigger unsolved technical problems are basically Fields Medal calibre. It’s very possible to take what’s good (which is a lot, and which is obvious) and leave the Fields Medal stuff behind.

Offbeat Neglected Prawn 1mo ago 💬 1

And now Firefox is bigger and more bloated than the suite now called Seamonkey.

Garbage nsec 1mo ago 💬 1

SeaMonkey, are you kidding me? SeaMonkey is used by like 17 Simpsons comic book store guy bearded weirdos who get a kick out of 1990s retro interfaces.

If the end-game for Nostr is to preserve the messy origins of what we have now and then chug along in a SeaMonkey-esque way, beloved by every one of its 17 loyal users, then we may as well all just give up now.