Anatomy of a Nostr DM spammer.

67K DMs since the 28th. No other events.

https://api.nostrgraph.net/beta/identities/eb6f4df621815c168ac0c9942eecc5e522b64969fdcf1d596babbc66cc489000.json?pretty=true

Reply to this note

Please Login to reply.

Discussion

Most of them were self-DMs. Obviously it’s all encrypted.

Most likely a way to try get free storage.

What does this mean for actually users of Nostr?

Well in this case they don’t seem to be DMing random people, however around 4K by memory were sent to a different pubkey.

For normal users, nothing except this stuff clogs relays and makes your experience worse performance wise mostly.

So basically in order to unclog and make a users expirience optimal would it be wise to run our own nostr nodes?

Depends. Most the the job of a relay today isn’t querying your own events, it’s events of people you follow and conversations that stem from that. I’d suggest trying a paid relay before running your own.

Nostr.wine is decent.

And the key issue isn’t just one guys 250,000 self DMs, it’s 1,000 or 10,000 of people who do the same. And when we grow again by 10x or 100x. It hurts smaller relays the most.

A personal server can make sense as a local backup. It doesn’t need to be public.

We will have useful bots that may only interact with DMs.

Yep. But if the bot only DMs itself, host your own data on a private relay - don’t ask the network to baby you.

Pay to relay and proof of work will likely offset this issue anyway.

Sorry, missed the DMing to itself part.

Yes, fair enough; a legitimate use case calls for a dedicated paid relay for that.