For all you know, half the people you've traded with in the last month already have a special client with a blacklist/whitelist in it. How would you identify them so that your client doesn't let them back on the "taint-free" fork of the blockchain? I think you might have a good answer to this, but it will be one which demonstrates a misunderstanding of the original proposal.
I've received at least one personal request to continue developing this idea from someone who suggested that the information in the blockchain will probably be an important part of the future of private justice. That is, in fact, why I came up with the idea.
No one has yet pointed out explicitly that if a thief were on this discussion, that thief would make every effort to discourage people from even thinking about the proposal.
This is an interesting perspective. I only meant that anyone, regardless how they feel about fungibility, can compile a list of all the address that currently hold coin that was in an address that contained allegedly stolen bitcoins. So perhaps "taint" is the wrong word for westky to use in describing this fact.
To confuse the issue even more, I have developed a theory (silly, if you want to call it that) that the reason no reward has been offered by the victims of heists is that they weren't really victims:
- A large dealer creates a hidden (fake) member
- The fake member "steals" bitcoin
- The dealer "loses" all the records indicating who they owe.
- The dealer announces that they will reimburse all customers who can prove their claim
- Less than 100% of the customers prove their claims
- The dealer reimburses claims using the stolen bitcoin (after mixing it up enough to avoid suspicion)
- The dealer ends up with some stolen coin that didn't get reimbursed to customers who failed or didn't bother to prove a claim - customers who have abandon bitcoin because it's too insecure.
Anyone can do that if they have the marketing power and the lack of morals. But they're in danger if we start analyzing the blockchain more closely.
Given the vehemence of the responses to the proposal, I've become ambivalent on the likelihood of this theory. I asked Gavin for advice in setting up Python so I could enhance his BitcoinTools to do this backtracing, but he hasn't answered. So, large dealers who are unethical thieves, you might want to pressure him into ignoring me, and perhaps removing the tools lest some other curious soul use them to demonstrate that my theory is true. I sure hope to hell it isn't.