mikedilger on Nostr: I didn't say that. hodlbod says this around 43:20 "I'm pretty sure that Tony doesn't ...
I didn't say that. hodlbod says this around 43:20 "I'm pretty sure that Tony doesn't believe in nostr as like a censorship-resistant long-term solution, I think he said that in his blog post from last year when he joined." He thought he read this in your blog. Maybe we should look to see what your blog says to see if hodlbod was right. Of course you can tell us directly what you think and that would be far more accurate.
I don't believe copying events is a problem. If relays have space and are willing to accept them, it's fine. It seems like a waste of space and bandwidth, but maybe it's not if that space/bandwidth would otherwise be underutilized. The only way in which copying events presents a problem I describe with my "training wheels" analogy, that we cannot see if our clients are working properly with the outbox model because the events are everywhere so we don't see the problems and can't fix it. And so if events somehow didn't get copied for a time so we could see how bad things really are, that would be nice... but also that I have no idea how to make that happen so it's a pipe dream and let's move on. That was the gist of what I was trying to say starting at 41:47.
The real problem is not the event copying (which is a solution, but IMHO the wrong solution), but is that clients are not seeking events from the relays that the authors wrote them to in the first place (as fiatjaf says after hodlbod).
I don't believe copying events is a problem. If relays have space and are willing to accept them, it's fine. It seems like a waste of space and bandwidth, but maybe it's not if that space/bandwidth would otherwise be underutilized. The only way in which copying events presents a problem I describe with my "training wheels" analogy, that we cannot see if our clients are working properly with the outbox model because the events are everywhere so we don't see the problems and can't fix it. And so if events somehow didn't get copied for a time so we could see how bad things really are, that would be nice... but also that I have no idea how to make that happen so it's a pipe dream and let's move on. That was the gist of what I was trying to say starting at 41:47.
The real problem is not the event copying (which is a solution, but IMHO the wrong solution), but is that clients are not seeking events from the relays that the authors wrote them to in the first place (as fiatjaf says after hodlbod).