The cold start problem is one of the most brutal problems for any beginning project, and there really aren't any easy ways around it. Right now, nothing really beats manual curation and direct engagement. You're right that changing habits is hard, which is why I think most in the kind-1/social nostr space aren't the best to market towards, even if it seems like a good starting place. I think it's a crowded place to market to (you're competing for attention from kind-1 users, in a space with lots of kind 1 clients). Odds are they've found the type of app that suits their needs. Finding fresh eyes without nostr/freedom goggles (😅), who are already enthusiastic but fed up with their current alternatives are where to go. Working for them, and building for what they're crying for is where the growth happens. Knowing you're invested in _specifically them_ helps them invest in you. But I'm also someone stuck with many cold start problems, always refining my message and audience - so take the answer with salt 🤣

Make your community an island. Carefully look into the open waters, or at least make sure your users are equipped before they do. ⛵+🛡️ > 🛶

Off the top of my head, learning through the simplest models maybe, or agents as proxy. - Student wants to get something done, asks the big model - Big model knows smaller model's capabilities and directs it to accomplish something very limited, maybe even a partial answer. - Big model explains what happened and what might be needed next, maybe anything helpful for the student to manually do to cement concepts

For now, even manual curation works. You can broadcast recipe events to your relay, whether that's just you and your team, or your users. Minimal incentives like getting back the sats they spent on the relay could help as well. By default, searching those words to just your relay and/or the user's relays mitigates the spam. Automated discovery from the wider nostr ecosystem can come as infra evolves. Lot's of behind the scenes projects, I'm even working on one. I think when those projects are ready, that's when you should consider looking at that level of openness. Because otherwise, you're stuck with things that can be broken easily like key words and hashtags. Thinking it through, I think there are two things to keep in mind to balance out a quality community with discovery. 1) start however small and narrow you need so you can guarantee a good experience. If you can keep quality high by follows of follows (possibly further) from zap.cooking users, that's a lot already. Keywords and hashtags have the most effectiveness when you're confident in the sources you're pulling from. 2) Every step you take looking into the open nostr ecosystem opens you up to more potential bad experiences, so you need to deeply consider every step you're making, because what you display from your app is representative of the community you're cultivating.

Single relay with free read, cheap/trivial membership cost will take care of your problems, it's also the starting expectations for most users unfamiliar with nostr - "come look! We have a quality community with great content/recipes! Your membership keeps our community running and maintains its quality." Adding more relays is what users can do after, and that understanding comes from when users know what nostr is, which is pretty advanced (what is a relay? why would I want more? which ones to connect? how do I add more?) You're not gatekeeping anything. Rather, you're giving them a garden before they knowingly and willingly venture in the wild for more.

https://www.youtube.com/watch?v=B8oHoLL0O20 #music

We traded digesting the real raw food for digesting artsy predigested food. 😋 Oh how far we've fallen.

That's some gnarly gain they got going on 😳

This is how i speak I also don't know what I'm saying either halp

"the problem with time & timezones", aka: Man has existential crisis talking about timezones https://www.youtube.com/watch?v=-5wpm-gesOY

Gm Nos :)