The system is quite "self healing" in that once any attack is over, the scraps will be quite efficiently picked up and woven into new fabric again. No money can be "spawned" in the system, anything adverse that slipped into the block chain will be sorted out after. The expense of changing a transaction increases over time, if you want to be sure of a transaction: wait it out. Up to a point that reverting the transaction would hold no profit. That said it's relatively hard to organize such an attack and it would most likely leave large traces. Lewis 2011/6/12 Eugen Leitl <eugen@leitl.org>

----- Forwarded message from Ian G <iang@iang.org> -----

From: Ian G <iang@iang.org> Date: Sun, 12 Jun 2011 23:12:08 +1000 To: Crypto discussion list <cryptography@randombit.net> CC: Eugen Leitl <eugen@leitl.org> Subject: Re: [cryptography] attacks against bitcoin User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:5.0) Gecko/20110528 Thunderbird/5.0b1

On 12/06/11 8:16 PM, Eugen Leitl wrote:

...

How safe is the bitcoin cryptosystem and the communication network against targeted attacks?

It depends on what the intention or objective of the attack is. And that depends on the threat actor.

For example, a phishing threat actor would be looking to steal money. Whereas the state actor would be looking to figure out who is doing what, to attack out of band. A competitor would look to attack the reputation, by e.g., disruption to reliability or mud flinging. A retailer / consumer would look to dump liability. Insiders would look to extract rents.

Each of these interests from diverse parties result in different attractivenesses to different threat scenarios, many of them uncorrelated. Some of the scenarios and assets can be protected ("mitigated") by tech/crypto, but typically most cannot, and require non-tech mitigations.

To go any distance on this, you'd rapidly end up doing a major risk analysis, a lot of work.

The alternative is to start from the classical CIA, etc. The problem with that is it that it is someone else's threat model, not yours. You really don't want to discover who that someone is after you've built your system, it'll so ruin your appetite.

iang

----- End forwarded message ----- -- Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org ______________________________________________________________ ICBM: 48.07100, 11.36820 http://www.ativel.com http://postbiota.org 8B29F6BE: 099D 78BA 2FD3 B014 B08A 7779 75B0 2443 8B29 F6BE