I think Pablo gets it right, in that we don’t actually know #nostr true nature yet.. it’s too early to tell.

But defining the protocol by the clients built on top, seems to be a pretty basic trap we can avoid

Reply to this note

Please Login to reply.

Discussion

Isn’t the whole idea to avoid becoming too dependent on the client?

that’s definitely one benefit. I’d avoid the binary kind of thinking where possible.. “The whole idea”

One with an IT lean might say, to avoid becoming too dependent on the server(s)

If there was a “whole point” I think it would be something like ‘to enable information to flow freely, and to try and prevent one entity from accumulating a disproportionate ability to control the flow of information.

but maybe that doesn’t even properly capture it.

Yes, let’s say you had a falling out with a clients CEO, you simply take your key elsewhere and lose nothing and keep your feed, followers.

on one client or one relay or one algorithm, and not have one entity control over all. I suggest you check out a extension like nos2x // alby in a browser… or #[5]​ on iOS so you can experience many different clients.

It’s too early to know what will be built in the future, but I would hope that client that tried to violate user trust, i.e., creating accounts where the nsec is not accessible, or selling user data, etc. would be quickly abandoned. Right now, we are in a place where the development happens out in the open. We know who operates our relays and builds our apps. They’re part of this community and we can call them out if they do something we don’t agree with. I don’t think anything like this has ever been attempted before.

many reasons other then the client being a bad actor for using a signing extension… this explains it better than I ever could.

https://orangepill.dev/nostr-guides/guide-nostr-key-generation-and-management/#tools-and-clients-to-manage-nostr-keys-and-authorize-signing-of-events

It’s frustrating that we don’t have iOS native signing extensions yet.

you mean for apps? That would be nice. But I also think generally the community here would like to move to platform agnostic options, jacks always talking about trying to fund a web os.

One protocol but many clients nostr is.

There are so many different clients, not only twitter-like clients and you don't need to be locked into one. It's early days, but people are building P2P markets and wechat-like "super apps". Great thing about building on Nostr is when you build a client, you already have an existing user base. One account for all clients and one native currency of the internet for all of humanity.

And with that great power comes the great responsibility to properly manage one’s keys to both.

Definitely need better key management tools