We have too few users and barely any content, so there is nothing to filter and no users to do the filtering. Many good ideas will only show their true potential at a bigger scale.
Discussion
I use several custom feeds regularly, it's not hard to build useful ones. My UX is horrible though
different scales require different solutions. outbox will shine at scale as well. you could argue its pointless complexity with the current number of users.
id say that not adopting outbox is preventing growth .. thats why everyone is gravitating to primal (one big relay), cause if you don't you're just noting to mostly the void.
how many real messages get stuck in damus throttling? i suppose yes you can queue and retry, but damus is the only client that does this.
anyway!
just my daily future trip.
what does this mean that damus client is the only one that does this? this is not true. the client doesn't do any throttling.
damus is the only client that retries
this can't be true, retrying is the only way to have reliable delivery. email does the same thing.
pretty sure its true, ive used a TON of nostr clients, none retry. many dont even surface the error to the end user either.
thats insane
it prevents growth,.. when you started to throttle, couldnt really use damus relay anymore.. pay to unthrottle would be nice.
yeah just need a noteguard module that bumps the reputation to those who zap the relay
public relays need to rate limit, spammers filled like 200gb into the relay without me noticing, thats what prompted me to do that
commies don't like money, ya know
the irony is, they paid a fortune for their iphone
and wen android damus? been waiting a year for that
turns out your rust is a bag of shit apparently
sounds like the clients should just fix that
almost no clients ever show NOTICE messages either, which is kinda lame
If we get big without widespread outbox support and a lot of relay decentralization capture by one big company becomes super easy and an obvious move.
I mean primal literally can't implement outbox since their clients aren't nostr clients, so that is pretty much guaranteed at this point.
If you are right then Nostr is the biggest waste of time in the universe and we should just give up, no?
How else do you reconcile the certainty of failure with keeping actively working on a thing?
maybe because I don't have a negative and black and white view on everything? I see it as a progression. outbox is inevitable at scale unless you want to silo yourself.
I thought you said capture is inevitable.
I can see them capturing the most users and not being able to implement outbox, but that doesn't mean nostr is pointless or dead. it just means its being used in a specific way that will become more isolated over time.
I believe the Primal people are as idealistic as you or I, and they want implement outbox, but of course laziness and business demands will probably take precedence over those ideals in a normal day. Now if they get really big and do an IPO or something like than we're screwed.
But the sooner more clients implements outbox and more people start using Nostr in weirder ways (like the simple act of publishing to small relays, which is basically impossible today if you want anyone to read your note) the sooner Primal will be forced to cooperate.
I think we are further along outbox implementation than we are being given credit for:
1. When onboarding new users, Primal ensures that the default set of relays includes a combination of larger relays and “long tail” smaller relays. We are already contributing to the spread of content over the entire public relay infrastructure. In contrast, Damus uses the same 4 relays for all new users.
2. We leave relay hints in all user and event mentions, as well as “copy id” actions, so that outbox clients can find things.
3. By default, all Primal clients write directly to the outbox relays the user has specified.
Therefore, when it comes generating content on the network, Primal already does all the right things. Please correct me if I’m missing something.
What remains to be done is the ability to diversify ways of reading/discovering content. We can’t solely rely on the caching service; our clients need to be functional without it. We are working on it.
As for the “Primal isn’t a real Nostr client” claim, that's a bad look nostr:nprofile1qyfhwue69uhhyetvv9uju6nzx56jucm0d5qs6amnwvaz7tmwdaejumr0dsqzqvhpsfmr23gwhv795lgjc8uw0v44z3pe4sg2vlh08k0an3wx3cj96l2ln2. Let’s focus on building. Damus still needs proper outbox and blossom support. Maybe that’s where energy is better spent instead of fudding another project.
how come ross@primal.net doesnt have a outbox list kind 10002? just a kind3.
I don’t think ross used primal to create his nostr account. if he did he would have had proper NIP65 relays set.
gotcha. i read your message here a few times, it made me realize, your version of outbox is not what I had hoped an outbox model would do.
i had hoped that if a client supported outbox, that people who follow me would see my notes. but ive heard from friends, they never see me in their timeline, only if they go to my profile page or to a 'feed' that im included in..
from reading your note here it sounded like that's because of the reading from outbox on primal is 'long tail?'.. 🤔
im just tryin to understand. installed it today for moar testing, and using amber 🎊
Interesting and that does make sense. 🫂
It's nice to see the turnaround in Primal's approach. Has potential to become a real nostr client after all.
I hope it’s true. The app is good. I just want it to be real NOSTR on the backend. No shenanigans.
You're doing great and I appreciate the transparency with your path towards Outbox/Inbox. Primal is as much of a Nostr client as any other that exists today. Primal's approach may be different in some areas, but that's still fine in my eyes. It's just different and that difference can be used to market to and appeal to different types of people since not all users fit inside the same box. Just like Damus or Amethyst has their own use cases, marketing, and target users. We're building for 8 billion people and we're not going to all use the same app.
Just implement outbox and blacklist the mega-relays.
the user experience suffers when there are multiple large modes without outbox support. I immediately noticed it on damus when primal started getting popular. You are going to have to implement it for user experience or become a silo.
I feel that. But even at this scale I feel the need for custom feeds to filter the noise. I'm probably not nostring efficiently enough
It’s make sense
nevent1qqs2zsdur7sufvnu2h0yycyg3fxxe0e06k2n2ejuelz88tk0ejt3n3qpupmhxue69uhhyetvv9ujuerpd46hxtnfduhj2v3swaehxw309aex2mrp0yhxumm5daeks6fwwa5kute9xgc8wumn8ghj7mn0wvhxcmmv9ujnyvrhwden5te0wfjkccte9eekjctdwd68ytnrdakj7ffjxpmhxue69uhhyetvv9ujuvrcvd5xzapwvdhk6te9xgc8wumn8ghj7mnxwfjkccte9eshqup0y5erqamnwvaz7tmjv4kxz7tjwvhxumm5daeks6fwwa5kute9xgc8wumn8ghj7un9d3shjtnwv4u8getj0ghxxmmd9ujnyvrhwden5te0vejkuunfwgkhxtnwda6x7umgdyh8w6tws3mwe7