Is this just a case of using a relay that supports NIP-03? If so, which relay?

iirc @fiatjaf has a relay somewhere that will timestamp your events, and @Amethyst as a client will allow you to timestamp stuff manually, and will also display if and when stuff was timestamped; but in general its just an action 'someone' has to perform, ideally quickly after the original event occured; it does not matter who does it, because ultimately relies on the bitcoin blockchain and not on the person who caused the (indirect) reference to end up in said blockchain. For example i timestamped this note at the time: View quoted note → Sorry i don't have a super satisfying clear cut answer for you at this time, part of my advocation for NIP-03 is to get us there; i think many people/devs are snoozing on this issue.

Or maybe i did not timestamp that event, or its lost, ah well that one has more than enough replies to tie it down time wise🤷‍♂️

I'm kinda confused tbh, but more by the fact I don't understand the tech. I assume you dont mean the '1 yr' in the example you showed? But as you mention actually on the blockchain? What is the technical hurdle? On the surface it seams like a simple thing.

Oh let me explain. Tldr: you can proof something happened before a certain moment in time by anchoring it to bitcoin. Nip03 is just putting those proofs in an event. Nostr events have a time/publish date field, where it is written when the creation supposedly took place. But thats just a number, and the author is totally free to lie about that number, nothing is really preventing that. But we do have something else, called 'Open Time Stamps' (OTS). It provides a method to proof that something is ATLEAST a particular age old; i.e. it proofs something existed at that moment in time, but could be older. It does that by putting a cryptographic hash via a bitcoin transaction into the Bitcoin blockchain; given the nature of Bitcoin's proof of work chain, an outside observer can be very certain as to the approximate date/time associated with the block that particular transaction/hash resides in. Now all that is in the blockchain is this simple hash; but that hash is the product of a process where a (potentially large, say millions) collection of files/things are condensed into said hash, and for each of those files/things there exist a proof that allows an outside observer to verify that that file/thing indeed was part of the creation of the hash; ergo the file/thing must have existed at the time the hash ended up in the blockchain. (Hope im still making sense here). In short: its a method to tie countless of things to the bitcoin blockchain in a very cheap/efficient/scalable manner to say something verfiable about time about those things. I.e. you can't produce nostr events after the fact, pretending that they are old (well you can, but you can't and also provide such an OTS proof, because you are not powerfull enough to rewrite the blockchain). NIP-03 is a simple standard where you put the timestamp proof of an event in a seperate event. This for example could help you defend yourself in situations where your keys are stolen; assuming you consistently timestamped all your stuff, and people expect such OTS proofs, the key-thief cant create legitemate events pretending to be statements made in the past.

That's why i support the idea that absolute everything stored at IA must be shared through torrents

That makes sense and thanks for spelling it out. I'm learning all the time on here, yet again nostr makes me smile. So why don't most clients implement NIP-03? Is it technically taxing to timestamp every note or is it a decision made by client devs? Or as ypu hinted at earlier, devs are actually unaware of the need?

I don't think people realize the point. I did a podcast with fiatjaf about open time stamps, when he was busy writing the implementation, and it was only during that conversation when i explained that he realized why its important, lol. I just re-listened it to see if it was worth sharing it with you, but i think you are better off with my wall of text haha. Anyway the point was context; some old book, is physically an old book; but data is just data, it has no smell etc. So, especially in the age of A.I., you can not only conjure up data, but also a whole bunch of fake context surrounding that data. So something pretending to be very old, might as well be created mere minutes ago, including its surrounding conversation; and when over time a large quantity of keys get compromized and public knowledge the mess only becomes greater. We are better off just structurally timestamping all the things, future historians will thank us for adding some smell to the data.

I just realized i did make this little article once: View article →

How do you know the Internet Archive will be burnt down? The US government can't do that to a 501(c)(3).

I made it up

That makes so much sense. Cheers man. Smelly data, that's what we need.

What OTS and Nostr show isn’t just that something existed at a certain time. Once you can prove existence and authorship without relying on a trusted party, the same mechanism applies to later changes: updates, commentary, or even walking away. Proof-of-creation is probably just the first step.

Timestamp deez nuts, future historians.

It will probably break at some point because of the crazy volume, but @npub1z4d4...j7rz is now timestamping everything from a bunch of popular relays instead of just stuff from wss://pyramid.fiatjaf.com.

Ever note is not timestamp already?

Anybody willing to try this nostr web archiver? https://chromewebstore.google.com/detail/nostr-web-archiver/aegkhiohbkiminnfpaekgpedkiookikg?pli=1

yes, after my hangover-day is done