00:34:26
atomfried:matrix.org:
https://ccs.getmonero.org/proposals/emsczkp-research-folding-gbp.html
00:34:26
atomfried:matrix.org:
If this succeeds could it be used for something like "miners collect all proves and add one proof for all transactions in the block" ?
00:45:35
DataHoarder:
this was brought up in MRL meetings. Specifically “Stream proving” part would, and might allow for a different class of "full" verifying nodes with pruned txs, but with aggregated proofs per block (so they can be fully verified), and archival nodes that keep the full version of the original transactions (with individual proofs)
00:47:56
atomfried:matrix.org:
Why would there be a need to keep the aggregated proofs?
00:47:56
atomfried:matrix.org:
Couldnt the consensus force miners to Aggregate the proofs for every block?
00:48:31
DataHoarder:
they can. but you must be able to reorg
00:49:04
DataHoarder:
which involves the full transaction, and tx hash also involves the full tx + pruned part atm. someone might want to verify that other part
00:49:09
atomfried:matrix.org:
Ah i see, forgot about that
00:49:43
DataHoarder:
current pruned nodes download all, verify, then prune after a while
00:50:18
DataHoarder:
these aggregate "pruned" nodes could do that, but also, be able to sync pruned data and verify
00:50:47
DataHoarder:
makes full nodes quite nice in limited bandwidth environments, without skipping verification of the proofs
00:50:59
DataHoarder:
archival nodes would be doing both
00:51:11
atomfried:matrix.org:
I see.Why stop there? Couldnt we Aggregate like n blocks into one proof?
00:51:17
atomfried:matrix.org:
For pruned nodes
00:52:00
DataHoarder:
I don't have more information on that. I'd recommend you wait for them or gets brought up next MRL
00:53:27
atomfried:matrix.org:
Ok, thank you.
00:53:27
atomfried:matrix.org:
Maybe one could even fold the all proofs into one, so initially nodes have to only verify this one folded membership-mega-proof 🤔🤣
07:00:09
rbrunner7:
"One folded membership-mega-proof": Don't these people here, mentioned recently on Reddit, try something similar? https://minaprotocol.com/blog/22kb-sized-blockchain-a-technical-reference
07:40:04
kayabanerve:matrix.org:
Mina's whole thing is a single proof for the state, yeah
14:27:59
atomfried:matrix.org:
Is something like this achievable for monero?
14:27:59
atomfried:matrix.org:
If the math stars align ofc
14:32:54
kayabanerve:matrix.org:
With years of R&D
14:33:07
kayabanerve:matrix.org:
I'd say an end-goal of Monero should be completely succinct proofs.