As much as I love #nostr I think one of the biggest issues that needs addressing is the poor loading of accounts and notes.
For instance, on Amethyst just now (my go to client even above my own) I loaded a profile to see who'd written a note, because their profile pic and name won't resolve, and lo & behold, the profile view shows no more info amd just 7 notes and 2 replies load.
That's on a perfectly fast Internet connection on a perfectly capable device on what is arguably the most comprehensive nostr client.
What's the cause?
Let's surely solve those issues before getting in to arguments over changes that break things at the protocol level.
If I introduced anyone else to this ecosystem now they'd see that happening and give up. 2 people I brought here already have.
I don't think they'd give two flying fucks if replies to kind 1 notes are also kind 1 or kind 1111. It's unimportant.
#TellNostr
Replies (27)
Yes! 10 minutes average for a bitcoin settlement is good, 10 seconds for a lookup on a nostr profile is not going to cut it for the average person.
Exactly. I just looked at your profile for about 3p seconds, it didn't load. A few of your notes did, but thats it. People won't put up with it.
I have citrine of my phone, the device which I'm currently using to not see people's profiles or notes loading.
Matt can you see my notes loading . All my iPhone frens left only 3 are still here. The frens on android are still lurking but not active 🫂 good point! I hope the next wave wil stay and not get annoyed :purple:
Yeah, your notes and profile are loading for me, at the moment. Of course, in a couple of hours that might not be the case. That needs fixing.
strange, what are your relays indexer?
Unfortunately, I have no clue how this could be solvable.
_Perhaps_ part of the root cause is the decentralized structure and different routing to actually access any given profile/note ...
So the easy way out would be not only a faster information protocol but also nodes with more locally stored data - aka centralization, which simultaneously steer small node operators away from the network (cost).
Maybe better client side caching could solve it more sustainable - e. g. when reading a thread the profiles of the thread participants might already be loaded. With the cost of increased network traffic. Having all clients behave as small nodes would be cool but probably not feasible.
If that is, how its already how it works, I'll better shut up :D I am not a technical person really.
Of course, the 0% uptime response is BS, or there'd be no profiles loading.
That doesn't make a blind bit of difference. Stuff either loads, or doesn't, in what appears to be an entirely random manner, regardless of what devoce I'm using, be it the laptop, desktop, tablet or mobile. That's absolutely no good.
I'm pretty sure the centralised routing of content is a major part of it but i have to wonder if the physical servers being used are up to the job. Some of it will be down to clients, and the various ways they pull, deduplicate and prioritise loading. It's even harder for web clients. But these are the things that definitely need fixing. Not kind 1111 replies. That's just so not important at this point.
I think hist is only for kind 3
Not sure, I'd tag semisol in and ask him but every time I tag someone in a note in Amerhyst it crashes. Then when I go to send a crash report, it crashes. So I give up and carry on with my day.
Fiddling with the browser websocket connections limits is a cock already, just trying to get the web client to make the appropriate number of connections to draw data and close them when done is a song and a dance. But these issues are happening on native apps too, so it's not a websocket connections limit issue in amd of itself. I'm convinced it's the servers, at this point.
Thanks for tagging him in. Coincidentally I don't think I added that, I'm pretty sure that was a default added by Amethyst after one of the updates.
Yes, sometimes are the relay servers, when are too crowded.
I notice that almost all nostr users are using same old common public relays, posting same notes on same bunch of relays.
I use this public account only with 3-4 relays but usually I use nostr in a more private manner, using my own relay, with a different account, only for private stuff. And works perfectly fine.
On public relays, if you browse them individually (and not by "following" feed) you will see a shit ton on garbage. I do not understand why people are using these public relays for sending encrypted notes from apps or wallets when they can use a private relay.
Just go and browse nostr.wine or nostr.lol relays. Hundreds of garbage notes. So ofc your client then is loading slow.
relay.nostr.lol for example is full of these kind of notes
So expect other public relays to be the same or worse...
I don't know who runs this relay but they can set a filter threshold at 60-8% and not see them
Wallet apps using a NWC should use a private relay for this stuff.
they use ephemeral events that you shouldn't see in the notes
That makes sense. But then, would better quality infrastructure improve things? Outbox should make decentralisation easier, but is it really feasible to ask people to run their own personal relays?
Maybe in the future a nostr client (mostly native apps) could have built-in relays.
You open the app - your personal relay is up.
You close the app - your relay is down.
I'm building a non social focused nostr based app atm which includes an internal relay. Issue with that is that to be available you have to be online, so data can't be shared between contacts unless they're both online. Which means, as well as direct p2p connections, I need to establish a method for people to set up a shared private relay.
>shared private relay
I did a test with a friend.
Each one run a simple relay on his PC.
Each one set the nostr client to write on others relay and on his one in the same time.
If one relay is offline, the other still get the notes/chats.
It could be a simple p2p communication.
I am fine with "not being online" so others can't sent me messages. It forces users to not stay online all fucking day and just do other things.
I just tried to zap you and it failed with "failed to zap" 🙁
To be honest, in my apps use case, you wouldn't expect people to be online all the time and at the same time, randomly, but they do still need to be able to sync data, so that's why I'm building it with the idea of a shared relay. They can sync p2p when possible but it'll fall back to checking the relay when it's not. That also means the app will be usable even without a shared relay but people would need to coordinate.
Just another day in nostr land - it's the thought that counts though, so to that I'll say thankyou 🙏
