<c​ompdec:matrix.org> Rucknium: Do you you have examples of wallets that produce the defects in your paper? Exodus I think has there own implementation, maybe others

compdec: It's easier to see the evidence of nonstandard transactions on chain than it is to discover which wallet implementations may be creating them. About 10% of recent transactions have nonstandard fees: github.com/Rucknium/misc-research/tree/main/Monero-Nonstandard-Fees

isthmus has identified some nonstandard txs, too: github.com/Mitchellpkt/monero_fingerprinting_pipeline

If you look at the nonstandard fee data, almost all txs had standard fees for about a week after the August 2022 hard fork. Many Monero wallet implementations were nonfunctional during that period.

There are a lot of closed-source Monero wallet implementations. That makes it harder to figure out what is making these txs. Some of them could be coming from centralized exchanges, too.

We know that MyMonero and lws used to produce nonstandard fees, but the issue is supposed to be fixed now:

Users of different wallets and services could test them to see if their wallets are producing nonstandard fees.

<c​ompdec:matrix.org> I'll try some things too then. I made distributions over the Wasserstein distance for those persistence diagrams of rings I produced. There seems to be a short term, medium term, and long term structure to it.