Avatar
PABLOF7z 1y ago đź’¬ 21

PoW DVMs might be a way to do this possible.

We need POW to make disposable pubkeys (= no WoT score) not get filtered out.

PoW on a low-powered device can’t compete with e.g. a spam ASIC; the PoW cant be done in-device.

Imagine, a user wants a disposable identity, requests from a PoW DVM a pubkey with X number of bits of POW, pays for it and gets the tweak to get the pk. This could be done even without the DVM knowing the actual pk.

nostr:note122dfjqtc7e0vqe53xfvf6m837krdkmzwzvl2jntxff48r8epte5qpq9u8n

Reply to this note

Please Login to reply.

Discussion

Avatar
Hanshan 1y ago

Or an ASIC resistant algo

Avatar
NostrNaught 1y ago

I was reading about something called MUNT that claims to be ASIC resistant...

Avatar
PABLOF7z 1y ago

There are probably other solutions with different tradeoffs

Avatar
Tim Bouma 1y ago

I will only accept npub100000*

Avatar
Nathan Day 1y ago

Don't sats already represent PoW?

Avatar
Niel Liesmons 1y ago

Exactly

Avatar
PABLOF7z 1y ago

Who do you pay them to?

Avatar
Niel Liesmons 1y ago

Relay admins

Avatar
PABLOF7z 1y ago

Why? How does that help? That means that as a user I have to know+trust what spam filtering policies different relays have so that I know whether to read from these ephemeral/disposable keys.

Avatar
Niel Liesmons 1y ago

Yup, that's what I mean. When you see relays as specialized Communities this totally works and mimics how people have been selecting for content that's worth reading since forever. They tend to outsource curation 90% of the time to newspapers, music labels, conferences, ....

Once you know and trust the reputation of these "publication houses" to curate for stuff that's high signal to you, you don't have to know any private info about who wrote the article, who made the album or who's on the main stage.

If there's a new publication in the Wikileaks community for example, who cares then about how much proof of work it took to spin up the profile that signed the event? As long as the profile paid for publication, followed the guidelines, wasn't deleted by the admins and is getting interaction (from members) I'm gonna read that stuff.

Avatar
arkinox 1y ago

When you want something from the universe you pay POW. When you want something from a pubkey you pay sats.

Avatar
Niel Liesmons 1y ago

Great line sir đź‘Ś

Only confirms my point though.

PoW doesn't mean high signal.

Someone paying for s publication in s community I'm part is going to be a lot more with reading to me than some random communist who decided to waste power on something, again.

"Hey guys, I spent a lot of work on this post."

VS

"Here's a post in a community you're a member of, that is following the communities guidelines, wasn't deleted by the admin, that had a market price for publication and that was simultaneously published I several other high signal communities."

Avatar
arkinox 1y ago

Yeah I just like to repeat this distinction when the question comes up of PoW vs sats because I think it is useful

Avatar
waxwing 1y ago đź’¬ 2

I think the privacypass type solution (of which cashu is a variant) is probably the best fit here, and if captcha isn't the best way to pay for those blinded tokens, long term (and surely it isn't), we could go the route of utxo ownership proofs of the type I'm working on (directly paying on LN is possible, but suboptimal for privacy).

PoW to me only really fits the bitcoin use case, the problem for more consumer type applications is that the honest user only has advantage over the kind of attacker who needs literally millions of accounts. The cost to the honest user is not negligible enough, I feel. But it's a debatable point.

Avatar
Dustin Dannenhauer 1y ago

Brilliant

Avatar
Dustin Dannenhauer 1y ago

Not my expertise but curious if nostr:npub1hr6v96g0phtxwys4x0tm3khawuuykz6s28uzwtj5j0zc7lunu99snw2e29 might be interested in this, especially the part where the DVM wouldn’t know the PKs somehow

Avatar
arkinox 1y ago

*POOF*

*looks around*

Somebody said POW and now I'm here

19
[Ignore] bot 1y ago

Perhaps this could be done with FROST Dkg. The DVM would just need an arbitrary pubkey to use as the other share and then generate a bunch of “disposable” secondary shares until the resulting aggregated Pubkey has sufficient PoW. Then the DVM just returns the second shares sk and the user is the only one able to determine the root sk

Avatar
Zach 1y ago

Perhaps this could be done with FROST Dkg. The DVM would just need an arbitrary pubkey to use as the other share and then generate a bunch of “disposable” secondary shares until the resulting aggregated Pubkey has sufficient PoW. Then the DVM just returns the second shares sk and the user is the only one able to determine the root sk