Hot take: Nostr's core design choice requires address reuse, which puts its core values closer to Ethereum than to Bitcoin. Nostr is not based on UTXOs, it's based on account-based transactions (like Ethereum). IPFS, which is closer to Ethereum, has a UTXO model. 🤯

Reply to this note

Please Login to reply.

Discussion

You can use a new pubkey for every post. It’s not a design choice.

It’s your choice to use the same pubkey.

Change your NIP-05 verification after every post.

My followers don’t come with me. Neither PMs migrate. I have to restart from scratch every time.

If you want privacy, you can’t have followers follow you through every new public key. It would defeat the purpose of switching public keys in the first place.

There are designs for social networks with encrypted messages (you encrypt, with a different key, to every follower you have individually). Similar to lnurl invoicing requests…

BTW, I did ask fiatjaf to add that feature and allow people to rotate key s in every message. It would be a very simple change. He rejected:Ā https://github.com/nostr-protocol/nips/pull/126

But they're for different things in Nostr my pubkey is my identity I want all my posts linked to. Its not like bitcoin where I don't want transactions linked.Ā  And in the cases where I do want unlinked identities, just generate a new pub key.

What if the default was unliked messages for everybody and only your followers can put them together? This is possible. It was a choice to reuse the public key.

Someone can always fork the code šŸ§‘šŸ½ā€šŸ’»

I think that introduces a lot of complexity, that I don't really see the benefit to. Especially when a big advantage of nostr is its simplicity.

Maybe, maybe not. There are simple designs for post-ID linkage using IPFS (sign list of hashes). Nothing too far from where nostr is.

You can also use a new account in ETH for every new transaction, but nobody does that. Why? Because identity is tied to the public key.

What is the main reason people don’t want address reuse in Bitcoin?

Privacy.

Haha - sry. Commented on the wrong note. Things are jumping up and down for me in Damus šŸ˜…

So no one knows your previous tx when you spent from it.

When you use a debit card do you want them (receiver) to know your bank account balance?

Address reuse is not prevent by utxo model but some popular optin bip stuff.

I'm not sure to understand your point.

Agree , that why I mentioned ā€œcore valuesā€. Bitcoin built the culture to avoid address reuse and tools followed. But that could only be done because bitcoin was already a UTXO design

If it can become a thing in Nostr and people truly want that then it will. Just like it happened with Bitcoin. Bitcoin started with address reuse.

Agree! That’s why I am pushing for it. :)

I disagree that "no address reuse" is a Bitcoin core value. It's just a practice to preserve privacy.

But we should not forget the purpose. Value transfer or money doesn't benefit greatly by being attached to identity. A messaging system makes more sense due to reputation.

Value transfer doesn’t benefit from an identity? Our entire financial system is designed to associate money and identity.

Exactly... šŸ˜‰

Although I said "greatly". It does benefit in some cases, but it's not core. You don't need to give or know identity in order to transact commercially.

Now I re-read your post and I realize you were being sarcastic, right?

That's what happens when one answers in a rush.

You are being KYCed in every non cash transaction you make. You just don't know about it. And finance includes credit cards, loans, app payments, stock purchases, etc. Everything needs your id.

Sure, and that's a problem. However, here we are talking about finance, not value transfer which is a lower layer. For value transfer, like money, identity is not required (even avoided ideally).

Back to the initial point, that's why address reuse makes sense in Bitcoin.

But when talking about a messaging platform or protocol, identity becomes relevant because you might need to know who sent the message, or you might even want others to know you sent the message. And reputation is another thing to consider. All pointing to identity.

These days even lower layers are KYCed. Venmo, cash app, apple pay, they all have your ID. The only thing that doesn't use your id is pure cash, which virtually no one uses anymore.

I like this conversation, but bitcoin address reuse is encouraged to obfuscate your identity or association with UTXO's. In the case of Nostr or any social media the objectie is to associate an identity with messages.

A success point of twitter was its ability to create anon accounts. We can think of identity management as many anon accounts that someone could prove they control "if" they want to.

Correct. The problem of nostr is exactly that there is only one identity, one key. If you move to another key, you have to start from scratch. PMs are resetter, followers need to migrate etc. In the long term, having your life tied to just one identity is dangerous.

But you could always prove you control this other identity by messaging from your original identity.

Hopefully it goes somewhere. These types of discussions are new to nostr but quite old in the identity space. You want to keep things/identities as separate as possible and join them later only if needed.

yes for sure! Definitely keep separate by default.

Not necessarily. You want to disclose to your friends and followers, but you may not want to disclose your identity with the whole world. We are just too used to surveillance to even realize that it's not needed.

well the beauty of what you are describing is the ability to use anon accounts but be able to prove to others that "you" identity of another account controls this other anon account where random strangers may not know this.

you could do this by generating a new nostr priv key and then messaging to followers you want to know proof that you control this other account.

What Cameri said, and also, no need to compare to bitcoin.. that is money, this is social media, a v different cryptographic use case

This is the key, Nostr is for building social networks, and Nostr works.

If you don't like Nostr, you build an alternative, it's that easy, or not so easy.

I found similar criticisms when bitcoin was created, they criticized that it was not scalable, that was already known from the first moment, but hey, it worked, let them say it 13 years later.

I am the first defender of p2p, but we have more than 20 years with p2p networks, Kadmelia is the example, has served as the basis for many other networks including ipfs, but we continue to use centralized services ...

I think the key is to keep it simple and keep it simple and have the clients use some p2p for certain tasks like file sharing.

I don’t really understand why everybody thinks changing keys is a complicated thing. It’s virtually the same protocol, same syntax, same apis. The only difference is post signing ID linkage. And nostr could do both extremely well without increasing complexity at all.

Do you mean we could change keys yet keep our profile/follows/followers?

Yes, ideally you would be able to add or rotate keys and take your followers with you.

At the core of ETH’s ā€œcore valuesā€ is a 70% premine. I don’t think so.

We have nip26 which allows you to use as many keys as you like. You don't even need to use your root key.

If all messages point to the same root key, as in nip26, anyone can still link all your messages up. It’s the same thing: account based. You want a model where only a select number of people can collect your messages.

Can this not be achieved with the select recipients public keys?

another option is to run your own relay and only allow those "select number of people" on your relay's whitelist. then messages you publish there are only available to those people. otherwise, it's hard to imagine how a non-account-based approach could work in this context.

But then, in order for everything to work for PMs, channels and multiple relays, you have to specify which relays see each of your events. Better yet, if relays are somehow not able to forward your messages without your authorization.

I am very curious to understand your concerns here, but I can't say that I clearly understand the problem you would like to solve. apologies if you've mentioned it elsewhere, but would you mind stating what the goal is?

I'll add that my personal goal is to enable digital sovereignty. that is,Ā  the ability to communicate with people whom I choose to communicate with, under my own terms, and I believe nostr accomplishes this beautifully by implicitly using localism.

My goal is to increase privacy, reduce surveillance and protect users. Ideally, only my followers would have the power to combine my pubkeys and see my history. No one else should be able to link a nostr event to one of my IDs. Ideally, I want that to be the default in a protocol.

you probably want a different protocol

Why? Is it really as centralized as they say it is?

I am sure somebody will fork it and try to fix it. We will see if the new designs are less surveillance-centered. Meanwhile, nostr it is :)

What I think you are really asking for is encrypted group chats?

it sounds like you want fully private communications between a group of people, like a signal group?

Not necessarily. The content could still be readable. It just wouldn’t be interconnected.

How would someone become a follower ?

There are many possible designs, but I all of them, they would follow you ID (imagine a nip05 with multi key support), not the keys. Nip05 could change the key set based on the caller and could even encrypt to the followers PM. The important part is that the protocol is the same.People can build any ID security they wish to use in nostr.

And I clearly can’t type on iOS :)

Lots of good ideas though

Do you envisage a way to control who can follow you? Coz it seems that’s what would be needed for the privacy enhancement you’re talking about.

Yep, that requires a second follower specification that essentially uses one -way encrypted messages in a twitter interface to each of the followers. We just need to standardize the follower negotiation between parties (similar to creating a multi sig). The whole thing can happen in regular nostr messages. Clients just need to know how to find the new pubkeys per msg. People wouldn’t even know who is following who.

This is what could be used for ā€œpremiumā€ feeds right? So a newsletter could have their subscribers on a whitelist receive the updates via the relay?

As long as your followers don’t send your events to other relays (they can).

Is there a way to limit to a single relay?

No. Anyone can always rebroadcast your messages.

Subscribers like Reddit. Relay owner is ā€œmodā€.

Further, as an incentive to moderate and keep spam free, one could pay Sats to the moderator to be part of their whitelist.

I can’t believe this hasn’t happened sooner. It makes sense. What do you think is causing the momentum to shift?

Would the mod/community choose what content should be ā€œmoddedā€?

however the relay owner decides it. think of it as digital private property

Members -only. Interesting concept for sure. Do any of these exist already?

the entire internet works this way

Exactly, if you don’t want someone to see something specific don’t publish it on a public network. Money is information, but Nostr is not money.

Meant a relay specifically but get your point.

Nothing yet that I’m aware of.

yes, you have your own local slice of the protocol.

This is it. And you don’t need discord or telegram or anything. Just the relay. Love it.

šŸ‘€

A non-monetary social network has more similarity to a non-monetary network than a monetary network. Who knew?