Topic: analysis on Shared Coin? (Read 904 times)

Has there been discussion on how an analyst would be able to trace transactions through Shared Coin?

It seems to me that the weakness in Shared Coin transactions is that the type of transaction on the blockchain per iteration would be fairly unique (multi-sig, multiple inputs/outputs)

If an analyst wanted to connect a payment from a wallet to a payment address,  the analyst (with code) could form a tree of SharedCoin transactions starting with the final payment transaction, to a
large number of input addresses that connect to a Shared Coin transaction that is no more than 10 nodes removed from the final payment. (the maximum number of Shared Coin iterations)  In reality, all of these transactions would be close together in time, most likely end up in the same block.   

Assume the analyst has knowledge of the source address(es) and the address used to receive change in the transaction, thus knowing the amount.  Is it easy to determine that the transactions are connected if the amount of the transaction is non-trivial?

Even without knowledge of the input amount, wouldn't the number of suspected input addresses be manageable for higher scrutiny by the analyst?  (traceable to a withdrawal from a FineCEN-compliant institution linked to your identity)

I personally like the idea of Shared Coin, but I'd really like to know how much anonymity is being gained by using it.