Thread

while I agree it would be good to have more content to explain nostr, it won't really matter given the current onboarding and product experience. it's way too confusing for folks, and generally not satisfying instantly. people will and do churn out. if there's one thing to focus on (for the social media use case) it's search. both for people, but more importantly, topics. Twitter won because of search. it wasn't a social network. it was an information network, and it excelled at real-time because of real-time search. that's a base requirement now. what nostr adds, and what will make it sticky, is the multi-app/use-case ecosystem. but each use case is going to have different needs for attracting people to it. search is table stakes for the social media one. it won't matter how good the tutorial content is until this is done right. nostr has the benefit right now of having a completely open and wild API. that matters as every other service is closing down. it's perfectly OK that we mostly have devs and bitcoin-obsessives at the moment. we have time to get all the kinks out and make something that's truly hard to replicate. where the only way to compete will be to join. View quoted note โ†’

Replies (47)

in your opinion, what made onboarding on twitter so good, and what specific parts of onboarding in existing nostr clients (take Damus for example play the biggest role in people having second thoughts and drop off? is it complexity in language, are there too many steps to join, too little, or something missing completely?
In this vital discussion, please don't overlook mid-level people.๐Ÿ™ I've been in tech most of my life, but have been an inventor and a designer of software systems that run on microcontrollers with less emphasis on the network side. Even with rather significant tech bones, I struggle to keep up with nostr developments and with understanding how to pull everything together and use different apps in a coherent way. It would be great if there were some centrally available tutorial-type information that would help me get a handle on things. Given that, I could be much more effective at onboarding normies.๐Ÿ˜† View quoted note โ†’
It's interesting this is at top of mind today, as I was just thinking about the npub/relay discovery problem and its tendency to centralization/federation. This seems like a solved problem, but it hasn't yet been integrated into the nostr model, simply out of maintaining simplicity. What I mean by the discovery problem is that a npub's present preferred relays is not trivial to discover, while at this time it depends on concentrator services like nostr.band. Bitcoin already has the notion of a mempool and it uses a DHT to share a common set of data. TOR also has integrated a pubkey discovery mechanism to find .onion rendezvous nodes. To find the location of a person's preferred relay(s) and/or simply a list of relays to find a npub's most recent profile data in a decentralized way, it seems to me one of these techniques is appropriate. It is also imminently compatible with the gossip/outbox client model. This way, nobody really needs to know who is at what relay to find their most recent posts and latest profile update. Then it becomes feasible for someone to have only a self-homed relay yet reach the full audience regardless of which relay their friends and followers are using.
guess which way do hundreds of millions Chinese youngests search? not Baidu apparently, nor Weibo (twitter alternative) Answer is Xiaohongshu, where real active young plebs sharing their daily loves or struggles! where there's a real active user base, it's the de facto search engine. Let's hope nostr would be the decentralized optimization.
I probably talk about the bitcoiner content too much, the problem is not at all that weโ€™ve got tons of bitcoiners or that theyโ€™re talking about bitcoin too much. More itโ€™s that itโ€™s hard to discover the other people and conversations. The UX experience is where we need to do work. Itโ€™s not so much fun, but itโ€™s what will make the difference. So when it comes to contact discovery, Iโ€™ve always wanted to do something Private Set Intersection. Anybody interested in working on getting PSI for privacy preserving contact discovery that we could use with Nostr? Thereโ€™s code which does the first 90% of the work:
GitHub
GitHub - OpenMined/PSI: Private Set Intersection Cardinality protocol based on ECDH and Bloom Filters
Private Set Intersection Cardinality protocol based on ECDH and Bloom Filters - OpenMined/PSI
 Regarding search, my initial thought is throw elastic search at the problem. But then Iโ€™m wondering what LLMโ€™s could do to make search better. I think weโ€™d want to have multiple search providers with a common api, the way that Damus and others support plugging in to translation engines. I think the needs of search and indexing are different than a normal relay, and we probably need those broken out in to a separate service or an add-on which only some relays provide.
Partially agree with both statements, but not convinced about the priorities. I think there is a bit of a state of operational blindness reached. At the very least, other questions should be asked like how do you guarantee equality and equal opportunities here, especially because Nostr could be more than social media. Also, external input should be considered relevant, and there should be the courage to incorporate it here. Right now, Nostr is a men's club, you can ignore that and continue to code into the void or start with a fresh infusion of healthy vitamins to heal and close this gap. There are both other needs these days from ppl outside and more similarities than you might think with the Bubblers. Building bridges, creating a healthy base for thinking and participation first? I'm still here with ideas for a broader focus on conferences to do that and also for clever PR and film, which can arise from it. Freedom here is real. But creating a healthy base should not be optional.
People who can't hear about chains that must not be named cover your ears... Okay check out what farcaster and warpcast are doing for discoverability and community finding. They got it right or at least in the right direction with easy to add topic lists. I asked before about social clients adding an easy click list of popular tools and clients that do music, streaming,stores etc. I was sent one website that keeps a list, it needs to be in each app.
Absolutely. Search is paramount. Some Nostr clients, such as online Nostrgram, have difficulties displaying the last notes posted from an npub. This naturally depends on which relays have been utilized, but the user experience results in a situation where there is no properly working timeline. When a user can't see new notes from an npub they follow or visit, interaction and engagement will be impacted. Zaps, quote-tweets, replies + likes, all function as feedback data streams, giving users some information about the extent of their reach. This is one of the reasons why 'likes' have a purpose for small accounts with limited reach. Users have an interest in measuring their reach in some way, not the least to verify that their relay setup works properly. Having some idea of how many people that are able to see a note is extremely important when an individual compare different social media options to judge where it is makes sense to be active. If the reach on Nostr is worse than on a social media where you are shadowbanned, then people will choose the app where they are shadowbanned, if that means a better reach in relation to Nostr.
View quoted note โ†’ Bluesky already won the search, not in the official app, but in 3rd party apps thanks to their access to the firehose, granted, that is easy when you have a single data store, so we will see as federation is introduced, but regardless, search can't be decentralized, it is always a function of crawling and aggregation, and that itself is very expensive. It can be competitive, but it will always be as centralized as Web Search.
I find it fascinating to be an early adopter to the point that there is actually a chance of Jack Dorsey himself reading my comment. I wonder how this will scale when A LOT of people onboard. Just look at Elon's Twitter. Must be impossible to find the needles in the haystack of comments without proper algos.
Decentralizing search itself seems impossible. We're likely stuck with private search providers. The thing that centralized Google as our legacy web search provider was simply that they produced better search results than all the others by a wide margin. Like all other decentralized systems, the only way I see to stop that happening to private search providers in Nostr is for them to share a search distribution protocol. Do you think it would be possible for Nostr devs to work together and build a single search protocol for all their apps that includes results from all present and future private search providers? Or will the problem always remain that a new private search provider could pop up that is so good they'll steal eyes away from the in-app search results?
Right now Nostr is what Gravatar AND Trust Pilot should have been. You don't need search to connect with content / people you love when you can find and follow them in any app/website (if that app chooses to integrated Nostr - which they should!) That following / timeline is then portable. "Oh I just saw a post from that guy I found on Walmart dotcom.. that comment he made about the weed whacker was hilarious.. oh! I better order more beans.. *navigates to Walmart dotcom*" People and communities sell products.. Advertising is still dying and dying faster and faster. Nostr should be marketed to brands. People can start discussions around products or a specific website and connect with each other. Nostr turns every piece of online real estate into a trusted open discussion with cryptographically proven identity. Welcome to Nike dotcom - log in with Nostr to see what people are saying about this products. โ€ขThey can run a node on the site and add support to select third-party nodes for trust/backup/failsafes. โ€ขThey can monitor comments and feedback in real time ON THE SITE with the MAJOR added value of people taking those social connections away from the site. โ€ขThey turn EVERY customer into an influencer Sure clients can work on onboarding and search improvements but every social site is doing that 24/7.
More you delay the mass adoption - better would be the solution ! Simply because once there are general users (in millions and billions) , the most of dev cycles are spent in scaling and Ux - focus is shifted from building a "great" product to building a "useful" (and mostly dumbed down product) .. This is the primary reason Windows or Mac could not compete Linux on product quality. The key is thus sustaining the product even without general success metrics such as MAUs. A "great" product, by definition must solve complex challenges. It does so by embracing the complexity. Simplicity, in this case, is outcome of resolving the complexities , NOT of avoiding them. Thus - in the intermittent phase - you would notice lot of churn. Devs should not be concerned of dumb users leaving (churn) .. they were never meant to stay! Real users will stay with or without documentation - they will figure out the things even without search. A "great" product must be targeted to their needs till it resolves all the complexities and becomes simple enough for dumb users .
I believe there is lots of group think worry going on here, regarding "education". I just downloaded Amethyst and created an account. The bitcoin banter was useful in helping me gain interest and figuring out how to (want to) attach an alby address. I am not a fan of KYC and have not taken the next step of getting an actual "wallet". For now just tell people to download an app (or go to website) and create an account. Those people will make demands on the dev's that will(or not) be met and build retention. Having people, like me, able to see this banter makes me feel a part of something new and helps build that same community and educate us as well. Just keep up the good work! PS, my demands are that apps are available in a open source repository or directly from the developers website, or directly usable on a mobile browser. I hate default redirects to an app store and QR codes. If an app demands total access to my phone including camera to function GFY. I can toggle permissions just fine and if that breaks the app I go looking for something else. And of course there is more View quoted note โ†’