An interesting feature is that this simplifies the validation process. All that needs to be done is to parse the block list (of hashes) once. As each hash is parsed you simply look it up in a hash-set. If it doesn't exist you add it. If it does exist you delete it. When you are done parsing the block list, you will have the minimal set of valid and unspent out-points. You might even be able to keep the whole set in memory. (at least for a while!)
Is this is the same idea as "With "Balance sheets" most of the block chain can be forgotten"?
http://bitcointalk.org/index.php?topic=505.0I think the problem with your proposal to make transactions less traceable is that the network has to be able to check that someone is not trying to spend someone else's coins before incorporating the new transaction in the block list. This, combined with the requirement to prevent double spending, seems to preclude greater opacity.
You must also discourage people from forcing the inclusion of arbitrary data in the block chain.
ByteCoin