Replying to Avatar Shawn

I'm seeing references to "key package relays" in nostr:npub1whtn0s68y3cs98zysa4nxrfzss5g5snhndv35tk5m2sudsr7ltms48r3ec and nostr:npub1h0uj825jgcr9lzxyp37ehasuenq070707pj63je07n8mkcsg3u0qnsrwx8, maybe others. I kind of have to laugh. I'm a nerd, and I barely understand the current rats nest of relay management. Now we have more. Something has to give here.

Making an app user have to *think* about relays early and upfront is bad UX.

You are making a tradeoff: cognitive burden for censorship resistance.

When 99% of your users do not face a direct censorship threat, you only add overhead.

Especially now with multiple categories of relays, that are confusing to even nerds like us.

All this effectively does: annoys the users who want nostr, and repels those who are curious.

Instead of offering on-ramps to newbies, we are offering on-walls.

Relay management is a must *and* it belongs in advanced settings. The word "relay" should be used sparingly and deliberately.

Convention over configuration all the way.

nostr:nevent1qqstndx77ca4j83mq64dkcvtfj8yj6y5amh4wxlhyra7056dgvlyxrcpz4mhxue69uhhyetvv9ujuerpd46hxtnfduhsygx8aknxpf4usfc9xr5zkjnhz2kdaghrrhq22mude926cqy7lktuscpsgqqqqqqsy29pvt

Reply to this note

Please Login to reply.

Discussion

A very important point you have just made.

nostr:npub180cvv07tjdrrgpa0j7j7tmnyl2yr6yr7l8j4s3evf6u64th6gkwsyjh6w6 says Nostr without relays front and center makes little sense. I kind of agree. Which would make hiding relays under advanced settings a kind of protocol self-harm. If you go that route then let's just build one gigantic nos.lol on the largest AWS instance money can buy and be done with it.

I think it may be about rethinking the entire presentation of relays, to the roots of the roots. Even the word "relay".

I don't agree and it does not lead to one gigantic relay

Hyperbole of course. But it does lead to a scale pattern that rips the heart out of Nostr by the time you get to around 500k daily users (should that day come).

Complexity maxies will say I'm oversimplifying, but yes, hide it away. Just like Signal does.

The problem is the nature of what Signal is hiding away there is different in kind.

What is the point of Nostr without relay awareness? It'd be like abstracting away the flavours at Baskin Robbins. You come, you get Vanilla, and if you ask then someone will take you to the back to see the other 31 flavours. Makes no sense.

If do you that then just go IROH and WebTransport for the stack.

You remind me of a friend, who takes no privacy precautions online, and his argument "why care if they know everything about us anyway?"

It's a nuanced subject and I believe there is a sweet spot between UX and censorship resistance

I think I'd like your friend.

For me it's not about a sweet spot between UX and censorship resistance. Tying relays to censorship resistance exclusively was a bad idea from the start. There's so much more to relays.

Also without maximal user awareness of relays then I can't see what the point of Nostr's underlying stack is. Websockets for what?

There's a more to relays, sure, what does that have to do?

Because coming at relays a different way means relays can be the first thing in every user onboarding flow. Like the very first. Which they should be. But super digestible.

And if they are the first thing then this whole sweet-spot spectrum doesn't really exist anymore.

Basically if we have to find out where relays sit on some kind of sweet-spot UX spectrum then we've already lost. The entire concept of Nostr rests not only on relays in a technical sense but on users being aware of them too. Like that is the most fundamental thing, by a long, long shot. It's almost the entire justification.

so according to you, the sweet spot is where the user doesn't have to think about relays at all?

Of course not

if users don't have to think about relays upfront, they'll probably never think about them. that's why i believe every nostr client should ask for the relays to connect to on first launch.

defaults are centralizing, so many people won't change them.

how you make onboarding as smooth as possible is a matter of good UI. nostr:nprofile1qqs84k6jpsav0jmdeqjn2zxlpnsajaw6f8l0m2d4e9t8gjsyn53s4nspzamhxue69uhhyetvv9ujuumfv9khxarj9e3k7mgpz3mhxue69uhhyetvv9ujuerpd46hxtnfduq3vamnwvaz7tmjv4kxz7fwdehhxarj9e3xzmnyhat8ky seems to be doing an amazing job with 0xchat lite. users should be *aware* of what is going behind the scenes, they deserve to know, let's give them free will.

making decisions for other people is always bad unless they're not able to, and this is not rocket science, everyone uses the internet nowadays and has some idea of how it works. btw, nostr:nprofile1qqsyvrp9u6p0mfur9dfdru3d853tx9mdjuhkphxuxgfwmryja7zsvhqpz9mhxue69uhkummnw3ezuamfdejj7qgswaehxw309ahx7um5wghx6mmd9uq3wamnwvaz7tmkd96x7u3wdehhxarjxyhxxmmd9ukfdvuv are any UI changes in that regard coming to outbox amethyst?

Fully agree. If we’re going to put relays behind advanced settings then we know that less than 1% of new users will have anything to do with that setting.

Ergo, as you say, the default relays will be the forever relays.

Which leads to the question: What are the default relays?

To which there are only bad answers.

Right, like no-one uses VPNs because that's not the default in any OS.

No amount of UI in your face will motivate you to understand what a relay is, it will just frustrate and you will put whatever just to get by.

People motivated enough will find the setting.

There's a lot of reasons why relay variety and supply in an open protocol will thrive.

That doesn't acknowledge how fundamental relays are to nostr. It's in the name! VPNs are not some fundamental element of MacOS or Linux.

Without relays absolutely front and center, you have to ask: Why Nostr? If relays are just a credible exit then choose another protocol that is way faster, and way more consistent, and that also offers a credible exit?

What's the answer? For users for whom relays are some shadowy thing that they may or may not ever discover, why Nostr?

We need rational defaults + fine-grained configs.

default relays should be based on who the user wants to follow

they should get a list of top 100 accounts on NOSTR and asked to follow 10 of them, then relays are chosen from that

I have faced a direct censorship threat before, but I'm still concerned with how poorly the relays are presented.

there should be some kind of automated relay management for beginners that can be over-ridden manually by advanced users

kind of like in windows there is a the dumb-fuck apple-esque UX and then you can dig deeper to classic UX and even deeper to command prompt

a new user should understand nsec and npub but should not even know what a relay is until the time is right

Google is your friend or not 😅

There's a sat split on this event and it made me open the wallet twice to pay both recipients. Not cool nostr:nprofile1qqsf3d23f5vqdnkfrem3kmvfq5hf4puwszgfnvurn4syu0s6w53s50cpzamhxue69uhhyetvv9ujumn0wd68ytnzv9hxgtc6uz24k

you are probably wrong but who cares

True. We don't think about DNS, when we type an URL into the browser's address bar. We don't even think about the protocol (https) when we do that. Our focus is straight on the goal.

BUT, we can use some special DNS, if we have special needs like adblocking or privacy. The settings are there from the start, but not so obvious on the surface.

And I think most Nostr clients are taking the right approach. Each has its default relay, just like every ISP has its DNS. I don't see why a beginner should have to deal with that unless they want to.

If other clients do it differently, well then they're just doing it wrong, against the convention.

>I don't see why a beginner should have to deal with that unless they want to.

I think it's a more fundamental question of why is this beginner even here? If not for the relays? Relays are what make Nostr Nostr. Fiatjaf calls them Nostr's superpower, and I think he's right. If beginners are not coming specifically for a network that's built this way, and one that explicitly gives them these novel choices, then what are they coming for? The memes?

I see it like an electric car manufacturer saying let's ease people into the fact that these cars are electric over the course of a test drive. Okay but you won't get people signing up for the test drive in the first place.

This is a fair and valuable analysis.

However, selling someone on a car that's electric is a far cleaner, more valuable proposition than selling someone on a social network that's relay-based. What does that mean relative to where they're coming from?

Gas/combustion --> EV <> X/FB/Tiktok --> relay-based

I think you hit on the exact question there. How can relays themselves be a marketable thing? Be *the* thing? Be the reason people come to try out Nostr. The thing they heard about that makes them think, cool, okay, that's new, I'm going to try this all out. Because if relays can't be that then nostr has no chance.

But if they are to be that then the whole relay narrative needs a major overhaul.

Replace "relay" with "community" and I think it makes sense. Social has always been about community. Relay's need to dial in on this.

The narrative does need to change cause it is too focused on client UX. Clients can be vibe-coded in a weekend by a jr dev. The real nostr is on community based relays (paid or WoT). Get a good relay going with an engaged community and a client for it will rise organically.

Yes these are the lines to think along. New users can't be asked to choose a relay with no mental foothold at all.

I agree if you want expansion then you need to give the newbs the least amount of config possible.

i think even the private and public keys are confusing and should come in focus after the user opens the app the 3rd time or so. if the user is really engaged, then the app should offer to backup keys.

Exactly, we should definitely keep keys and relays as part of this culture, while maximizing for user retention

Yeah, nothing loaded properly in my Whitenoise app, then I found out there are no default relays included with it… hope the ones I chose work