Possible systemic weakness

[marcus_of_augustus]

Running through my head a scenario where an entity that wanted to gain by destroying the network could have the resources to manipulate difficulty and price simultaneously, albeit temporarily. There are three cases I can see;

i) drives difficulty and price higher together
ii) locks difficulty and price in tandem (presumably some kind of feedback mechanism)
iii) drive price lower and difficulty higher

They would achieve this through expending resources in both market activity and mining activity.
Case i) trivial, I don't see this as anything other than increased participation so no weakness there.

Case ii) interesting proposition but not sure if this is actually an attack or what would be the outcome, positive or negative.

Case iii) is the one to worry about I think. They would have to first buy up or otherwise accumulate BTC resources or have access to short-selling futures mechanisms as well as have access to significant mining resources that could be bought on-line in a co-ordinated manner. Then through concerted market selling and futures selling depressing the price and simultaneously they drive mining difficulty high by increasing mining to the max they create a dislocation in the natural pricing signal.
Net effect being that honest miners begin dropping off-line as they become uneconomic due to the pricing/difficulty dislocation. I guess the question to answer here is, how long could they keep it up and is it possible to sustain it sufficiently to destroy all value in bitcoin?

[bittrader]

I have another case to add to your three.

I am the archnemesis of the Bitcoin network, and I am trying to do scenario iii above. As part of this plan, I want to - let's say - double the current difficulty. Theoretically, the network is currently solving blocks every 10 minutes on average. To double the difficulty, I need to double the rate at which blocks are being solved. To do this, the computing power of the network needs to be double -- or, in other words, my own computing power needs to be equal to the power of the rest of the network combined. Quite a feat, but possible for a well-funded archnemesis such as I.

However, if I had enough computing power to do this, I think I would use my evil strength to bring down the network in another way. Since I have half the computing power, I could start forking the real block chain and go on a double-spending spree. If network users couldn't depend on the network guarding against double-spending, the whole Bitcoin system would be doomed, I think (no one could safely receive BTC, so there wouldn't be much use for it).

[marcus_of_augustus]

Probably should have spelled it out but I was discounting the >50% attack since that is well described elsewhere. Using the word "manipulate" to imply that the hostile had some resources to influence difficulty but not over-whelming, i.e. dominant resources.

Anyway, how big does the network have to be to be safe from the >50%?

Are we there yet?

[bittrader]

Hm...it should be hard to have much of a significant effect on difficulty without a corresponding significant amount of computing power at your disposal. I guess it depends on what you call "significant" though.

I hadn't thought much about the issue of manipulating price though, until you mentioned it. I suspect that wouldn't be nearly as hard to do as manipulating difficulty. If you do it at the right time, spending a few hundred dollars on Mt. Gox can make a noticable impact on the exchange rate, though only temporarily. There was a post from you a few days ago comparing the Bitcoin network's computing power to the computing power of several of the top supercomputers combined. But comparing an exchange like Mt. Gox to any of the big financial exchanges is a different story, so it may be that that's more of our weak spot.

[kjj]

No, nothing in this thread is right.

Price manipulation is possible right now because volumes are low, but it will cost you more than it costs everyone else, so you'll lose.

The scenarios involving technical manipulation are entirely founded on misconceptions.  The network really doesn't work the way you imagine it does.  Someone would need several orders of magnitude more computing power than the rest of the world combined to pull off a block chain manipulation, and it would gain them very, very little.

[marcus_of_augustus]

The scenarios involving technical manipulation are entirely founded on misconceptions.

Clarify.

Technical manipulation as I describe in OP is simply adding a significant portion of power to the network, say 40%, over and above "natural growth", or adding and removing a significant portion (simultaneously with pricing manipulation).